University
stringclasses 19
values | Text
stringlengths 458
20.7k
|
|---|---|
Curtain
|
Curtin University WASM
7.1 THESIS SUMMARY AND CONTRIBUTION TO KNOWLEDGE
Traditionally, open-pit mining and underground mining have been studied as a
separately project in cases where shallow deposits extend to a considerable depth. For
shallow deposits, usually, open-pit mining acts as a primary mining method in the
beginning of the project. When open-pit mining is approaching its end of mine life, the
‘transition problem’ becomes apparent. It is a decision-making process of whether to
extend the pit or make the transition from open-pit mining to underground mining. If
open-pit is no longer an economic mining method for the remaining mineral resource, the
decision is to make the transition to underground mining. If the final decision is to make
the transition to underground mining, a combination of open-pit and underground mining
strategy is then adopted by the mining project.
In the practice of combination of open-pit and underground mining strategy, the
inability to maximize project value (discounted or/and undiscounted) and resource
utilization using open-pit mining and underground mining are studied independently. The
main contributions for the failures are arbitrary crown pillar location and over-mined pit.
Literature shows the significance of considering both open-pit and underground
concurrently in the mine planning and optimization process.
This research used mathematical modelling techniques to approach the transition
problem and proposed mathematical models for generating the transition point, transition
period and crown pillar placement to solve the transition problem. The proposed models
can provide a clear guidance on where an operation should consider making the transition
from open-pit to underground mining.
Given the computation complexity of the problem, this research utilized data
clustering techniques in open-pit mining and stope-based methodology for underground
mining. The main reason was to decrease the size of the problem for both open-pit mining
and underground mining. For open-pit mining, the data clustering technique was used to
aggregate blocks with similar properties in order to decrease the number of entities.
Meanwhile, the size reduction strategy for underground mining aims to retain the
profitable stopes and eliminate the unprofitable stopes.
77 | P age
|
Curtain
|
Curtin University WASM
In this research, two mathematical models were developed and presented. The first
model is Transition Point Model which is used to generate the optimal transition point by
maximizing the undiscounted cashflow of the mine operation. Following that, the
modelling process expands to achieve optimal transition point and period by maximizing
the discounted cashflow of the mine operation by developing the Transition Period Model.
The expansion considers the underground mining advancement, operation capacities,
blending, capital expenditure for making the transition, underground mine design
parameters, etc. The results generated by this expanded model provide a comprehensive
transition plan to the engineers. These two models were tested with a simplified two-
dimensional dataset for verification purposes. The results showed that all constraints were
satisfied.
The two models were implemented on a synthetic block model which consists of 7,200
blocks. Upon solving the mathematical models, the optimal transition point, optimal
crown pillar placement, and/or optimal transition period were generated. The Transition
Point Model provides the optimal mining layout along with maximized undiscounted
cashflow of $3.74 billion. One the other hand, Transition Period Model provides the
optimal mining layout along with maximized discounted cashflow of $2.60 billion
without considering any delay period when transitioning from open-pit to underground.
In contrast, while considering a two-delay period, the discounted cashflow is $2.51 billion.
Although the implementations demonstrate the capability of the models, it is unable to
be implemented in a real case study. This is due to the computational complexity of real
cases. A series of case studies were presented to show the limitation of the proposed
models.
The hierarchical clustering algorithm was proposed to handle the computational
complexity and reduce the size of the problem for open-pit mining. The proposed
hierarchical clustering utilized the similarity index to perform the clustering action. The
similarity index is calculated based on coordinate, slope factor, grade factor,
neighborhood factor and level factor. A two-dimensional case study was utilized to test
the validity of the model.
The two models along with the hierarchical clustering algorithm were implemented
on a larger scale dataset which consists of 83,025 blocks. The open-pit blocks were
78 | P age
|
Curtain
|
Curtin University WASM
clustered into 1,507 cluster groups and underground mining generated 2,136 profitable
stopes. The proposed clustering algorithm successfully reduced the size of the problem
for open-pit mining by approximately 85%. The implementation of the hierarchical
clustering algorithm along with the proposed transition models were successful. The
solution time for the Transition Point Model and Transition Period Model reduced
significantly.
In conclusion, a new methodology for solving the transition problem was developed,
verified, and implemented. It can optimize the undiscounted cashflow and NPV of a
mining project and can provide the optimal mining strategy for the project.
7.2 RECOMMENDATIONS
The first area of enhancement is a stochastic mining model which can generate a high
confidence mining plan. Geological uncertainty has remained as a topical discussion
within the mining industry and it will directly affect the mine planning and optimization
process. As presented by Chung, Topal, and Erten (2015), geological uncertainty will
affect the transition problem significantly. Hence, a ‘grey area’ will appear which is
named as the ‘Transition envelope’. Future studies could consider transforming the model
into a stochastic mining model where it can handle multiple inputs and generate a result
with higher confidence levels.
Furthermore, future improvement by introducing flexibility for underground mining
advancement can be considered. In the current model, the underground mining
advancement is top-down approach. In the future studies, it can consider expanding the
underground mining advancement into bottom-up and complete flexibility. Besides, the
model has the potential to further improve to allow for simultaneous open-pit and
underground mining operation/production.
Next, the aspect that should be concerned with in future studies is the underground
development mining schedule. Underground development scheduling can directly affect
the stope mining sequence and accessibility. Hence, it will help to generate a more
detailed and practical mining plan if there are dependencies/links between development
and stopes. Moreover, the sublevel stoping cycle should be involved in the future studies.
79 | P age
|
Curtain
|
III
I would like to express my appreciation to my supervisors Associate Professor Mostafa
Sharifzadeh and Professor Youhei Kawamura (Akita University, Japan) for their
invaluable supervision, support, and encouragement. Without their support, this
research would not have come to its current stage of development.
I wish to express my gratitude to Associate Professor Emmanuel Knox Chikutwe
Chanda (The University of Adelaide), Dr. Hyong Doo Jang (Curtin University), and
Dr. Markus Wagner (The University of Adelaide) who served as a co-operation to our
teamwork. Their support, guidance, and inspiration were greatly supportive during
the course of research.
I also want to give my thanks to Professor Roger Thompson (Curtin University) as the
thesis chairperson, who makes my earlier scientific connections with my supervisors
and provides invaluable supports to my field trips, conferences attending, and other
administrative assistants.
I would like to thank the managements of Angas Zinc Mine (South of Australia) and
Nickel West Leinster (Western Australia) for permission to use the mines as experiment
site and to publish this thesis. Also, the assistance of their staff in setting up the
underground experiments is hereby acknowledged.
I would also like to convey my sincere thanks to the managers of Hokkaido electronic
equipment (HDK Company, Japan), Mr. Kamijima and Mr. Sawada, for the software
and hardware developments and other technical supports about electronic devices.
I am very grateful to all of academic and support staffs of the Western Australian
School of Mines (WASM) in Kalgoorlie Campus and thanks to my co-research fellows.
Special thanks to my siblings and in-laws for their encouragements and devotions.
|
Curtain
|
VI
In the challenging environment and dynamic topology of an underground mine,
reliable and effective communication is a high-stake issue, along with the objectives
of safe and efficient mining operations. Automation through remote and automatic
systems has delivered improvements in workplace health and safety for employees,
operational management, energy and cost-effectiveness, and real-time response to
events. In this context, Wireless Sensor Networks (WSNs) have been widely employed
in underground monitoring and communication systems for the purpose of
environmental monitoring, the positioning of workers and equipment, operational
monitoring and communication system. Considering the capabilities of WSNs, a
ZigBee network is adopted in this study.
The aim of this study is to propose a reliable and effective monitoring and
communication system in underground environments, using new technologies. To
achieve this, a ZigBee network-assisted Geographic Information System (GIS) is
analysed to establish an integrated system for monitoring underground environment
features, thereby preventing underground incidents, and facilitating bilateral
emergency communications between surface operators and underground miners.
It is demonstrated how ZigBee network performance is optimised for such
environments. ZigBee radio wave attenuation is investigated to evaluate a stable
communication distance between ZigBee nodes at straight and curved tunnels in a
real mine scenario. Experimental measurements of ZigBee radio wave attenuation
are validated by simulation results. Based on the analysis of the experimental and
simulation results, factors affecting radio waves attenuation in the junctions,
curvatures and fields near and far from the source are also assessed. Stable wireless
communication distances between developed ZigBee nodes in the experiments of
underground Angas Zinc Mine were obtained 100 m and 70 m for straight and curved
tunnels, respectively.
Various sensor node arrangements of ZigBee networks for underground space
monitoring and communication systems are also investigated. The performance of
ZigBee topologies is analysed in 12, 20, 30, 40 and 50-node scenarios for stationary
|
Curtain
|
VII
node deployment in underground environments. The metrics used for the
performance evaluation include throughput, packet delivery ratio (PDR), end-to-end
delay, energy consumption and network security. The results of evaluation confirm
that the mesh topology is preferable in WSN design in relation to higher throughput,
packet delivery ratio, and network security, whereas the cluster-tree topology is
preferable in relation to lower end-to-end delay and lower energy consumption.
Attempts are made to make the system more effective and smarter in order to
process and manage underground ZigBee data received into a control centre
database using the GIS. ZigBee nodes are developed to sense environmental
attributes such as temperature, humidity, and gases concentration, switching ON and
OFF ventilation fans, and texting emergency messages. A trigger action plan for
attributes monitored above the normal or threshold value limits is programmed into
the surface GIS management server. It is designed to turn the auxiliary fans on
remotely or automatically in orange (caution) conditions, and to send evacuation
messages to underground miners in red (unsafe) conditions.
The approaches are generalised by modelling ZigBee networks with the more
comprehensive and realistic representations of a monitoring and communication
system for underground mines. A system design and model is developed based on
the classification of results from an experiment undertaken at an underground mine
in Western Australia.
|
Curtain
|
1
INTRODUCTION
BACKGROUND
Mining is one of the earliest industries to contribute to human civilisation (Hartman
& Mutmansky, 2002). The importance of mining as a foundation for most industries
has led to rapid progress, utilising the latest technological advancements. Increased
demand for mineral products continues to drive global expansion of the underground
mining sector, even though the substantial majority of ore deposits have already
been exploited from the near-surface.
Underground mining is a risky operation due to its workplace characteristics. Many
research papers have investigated factors leading to human casualties and injuries,
recurring hazards, and unexpected costs in underground mines (Khanzode et al.,
2011; Molina et al., 2011; Saleh & Cummings, 2011). Typical hazards associated with
underground mining are poor lighting conditions, narrow spaces, rock falls, poor
ventilation, wet conditions, communication constraints and structural complexity.
Some studies have focused on operational problems such as fleet management
including dispatching, routing and scheduling (Gamache et al., 2005). Others have
explored geotechnical considerations such as ground movement and tunnel collapse
(Ghorbani et al., 2012; Li et al., 2010) in order to improve safety and management
efficiency in underground mining.
In the challenging environment and changing topology of a mine, reliable and
effective communication is a high-stake issue, along with the objectives of safe and
efficient mining operations. Automation through remote and automatic systems has
delivered improvements in workplace health and safety for employees, operational
management, energy and cost-effectiveness, and real-time response to events. In
this context, Wireless Sensor Networks (WSNs) have been widely employed in
underground monitoring and communication systems for the purpose of
environmental monitoring, the positioning of workers and equipment, operational
monitoring and communication systems.
WSNs form one of the most stimulating fields of computer science research and have
contributed to home and industry communication and monitoring solutions over the
CHAPTER 1
INTRODUCTION
|
Curtain
|
2
past five decades (Silicon-Laboratories, 2012). However, they have only been
appraised and applied in the mining industry during the last two decades.
Research on WSNs in underground mines has advanced for in the areas of gas
detection and predicting collapses in coal mines. This industry has high event
statistics. Reported annual statistics of fatalities from coal mines, especially in the
USA and China, underpin the importance of utilising WSNs to minimise injuries,
deaths, damage to equipment and unexpected costs (Li & Liu, 2009; Molina, 2011).
Underground mining has benefited from the implementation of WSNs for strategic
positioning of miners and equipment (Chehri et al., 2008; Tadisetty et al., 2003; Yin,
2011) and communication both in vocal and visual formats (Müller & Noack, 2011;
Sicignano et al., 2013).
Considering the capabilities of WSNs, a ZigBee network is adopted in this study.
ZigBee, based on IEEE 802.15.4 standard, is a new wireless technology which delivers
greater benefits for monitoring and communication systems in underground spaces,
compared to other WSNs. The first version of a ZigBee node was developed by ZigBee
Alliance in 2004 (Longkang et al., 2011). It has acceptable communication distances
between nodes, substantial node capacity within a network, low energy consumption
by sensor nodes and low overall complexity. Also, ZigBee nodes, network installation
and maintenance are very cost-effective compared with other underground WSNs.
ZigBee does not require any access point or central node to transmit data between
clusters. Although ZigBee network has very low data rate (250 kbps) for image, voice
and video communication, it can deliver high-performance networking applications
for data transmission between nodes (node to node relays) based on multiple
wireless hops (Sharifzadeh et al., 2015).
The proposed system integrates a ZigBee network with a geographic information
system (GIS) to enable the monitoring and control of underground mining
applications from a surface office. GIS is a new technology and is used for spatial data
analysis in order to capture, store, analyse, manage, and present data that is linked
to locations (ESRI, 2012). GIS allows users to view, understand, question, interpret,
and visualise data in many ways, revealing relationships, patterns, and trends in the
form of maps, globes, reports, and charts.
CHAPTER 1
INTRODUCTION
|
Curtain
|
3
This study demonstrates the innovative integration of ZigBee and GIS technologies
into one system for communication and monitoring in mine tunnels. Underground
safety and health concerns are significantly ameliorated through enhanced
ventilation management and improved emergency text messaging. The integrated
system receives data from developed ZigBee nodes and maps information in the GIS
management server. The system can sense the mine tunnel environment and can
communicate, and control the operation of the ventilations. Temporal ZigBee
environment data including temperature, humidity and gas concentration readings
are processed in the surface GIS management server. A trigger action plan is
programmed to sound the alarm and remotely turn on auxiliary fans to clear the
unsafe conditions when the monitored parameters exceed threshold values. The
system enables emergency text messages to be communicated between
underground tunnels and the surface.
PROBLEM STATEMENT
The core problems addressed in this study are the mitigation of underground mining
incidents and the management of underground mining operations from a surface
office by an effective monitoring and communication system.
The death toll in the underground mining industry over the last decade is presented
in Figure 1-1. The narratives of the fatalities occurring between 2003 and 2010 were
obtained for three countries (USA, India and China). It is obvious that despite the
progress of safety technology over recent years, there are still considerable fatalities
in this industry. Therefore, underground mining remains one of the most dangerous
occupations (Chakraborty, 2012; MSHA, 2014; Wu et al., 2011).
In this study, an integrated system for underground mine monitoring and
communication based on new technologies using ZigBee and GIS is proposed. ZigBee
network is employed to sense the environmental attributes of tunnels and to manage
ventilation system as well as establishing emergency communication with
underground miners by texting message. This is merged with GIS services to create
an automated system to provide 3D visualisation, programmed trigger action plans
and multi-user operation.
CHAPTER 1
INTRODUCTION
|
Curtain
|
4
Figure 1-1 Death toll in the underground mining industry
The high quality of service and reliable message transmission through the network
are crucial issues in dense industrial WSNs. This study also investigates an adequate
sensor node arrangement of ZigBee networks for underground space monitoring and
communication systems.
OBJECTIVES
In order to achieve a reliable system integration in the underground mines as a
remedy for the difficulties and issues mentioned in the problem statement, the
objectives of this study are as follows:
• Review the state of the art regarding the use of WSNs and particularly ZigBee
networks in underground mines.
• Evaluate common WSNs for applications in underground mines and demonstrate
how a ZigBee network performance is suitable for such environments. ZigBee
radio wave attenuation is investigated to evaluate stable communication ranges
between ZigBee nodes at straight and curved tunnels in a real mining scenario.
Moreover, experimental measurements of ZigBee radio wave attenuation are
validated by simulation results. Based on the analysis of the experimental and
simulation results, factors affecting radio waves attenuation in the junctions,
curvatures and fields near and far from the source are also assessed. Finally,
CHAPTER 1
INTRODUCTION
|
Curtain
|
5
stable wireless communication ranges between developed ZigBee nodes in the
underground Angas Zinc Mine is posited to be 100 m and 70 m for straight and
curved tunnels respectively. The development of a ZigBee network application
compared to other WSNs in underground mines is also endorsed.
• Investigate various sensor node arrangements of ZigBee network for
underground space monitoring and communication systems. The performance of
ZigBee topologies is analysed in 12, 20, 30, 40 and 50-node scenarios for
stationary node deployment in underground environments. The metrics used for
the performance evaluation include throughput, packet delivery ratio (PDR), end-
to-end delay, energy consumption and network security. The results of evaluation
confirm that the mesh topology is preferable in WSN design in relation to higher
throughput, packet delivery ratio, and network security, whereas the cluster-tree
topology is preferable in relation to lower end-to-end delay and lower energy
consumption. The analyses show that the mesh topology creates a more reliable
monitoring and communication network with an adequate quality of service in
underground spaces and tunnels. Therefore, greater end-to-end delay and energy
consumption are not major concerns for the mesh topology in underground mine
applications, based on the acceptable data latency and the use of mine power.
• Analyse an automated underground mine monitoring and communication system
based on the integration of new technologies to promote safety and health,
operational management and cost-effectiveness. The proposed system
integration of a WSN-assisted GIS enables monitoring and control of underground
mining applications from a surface office. Based on the capabilities of WSNs, a
ZigBee network is adopted for near real-time monitoring, ventilation system
control and emergency communication in an underground mine. ZigBee nodes
are developed to sense environmental attributes such as temperature, humidity,
and gases concentration, switching ON and OFF ventilation fans, and texting
emergency messages. A trigger action plan for attributes monitored above the
normal or threshold value limits is programmed into the surface GIS management
server. It is designed to turn the auxiliary fans on remotely or automatically in
orange (caution) conditions, and to send evacuation messages to underground
CHAPTER 1
INTRODUCTION
|
Curtain
|
6
miners in red (unsafe) conditions. Multi-user operation and 3D visualisations are
other benefits achieved in the proposed system.
• Assess the controllable and uncontrollable parameters for the establishment of a
ZigBee network for underground mines. Accordingly, the methodology for
designing and modelling an underground mine monitoring and communication
system is generalised. Further procedures for a ZigBee network implementation
in an underground mine are proposed. For the physical verification of these
procedures, an experiment is carried out to design a ZigBee network model based
on the results obtained in an underground mine in Western Australia for
communication distances under different conditions. In addition, another
experiment is performed to validate this model by testing system functions and
applications including temperature, humidity and illumination readings, message
texting, and the control of ventilation fans through the tunnels of this
underground mine. The system operates successfully and demonstrates the
reliable outcomes of the system functions and applications.
SCOPE AND LIMITATIONS
This investigation was conducted to develop underground mining monitoring and
communication systems so as to promote safety and health, operational
management and cost-effectiveness. The proposed system integration of a ZigBee
network-assisted GIS enables monitoring and control of underground environment
attributes. The potential functions and applications of the ZigBee networks for
underground communication and monitoring is illustrated in Figure 1-2. These
functions and applications assessing sensor nodes’ abilities are classified as follows.
(a) Safety and health approach
• Air quality and quantity measurements
• Determination of workers’ locations
• Emergency and safety communications
• Gas detector and fire alarm
• Geotechnical monitoring
(b) Operations management and control
CHAPTER 1
INTRODUCTION
|
Curtain
|
7
• Real-time monitoring of underground mine operations from a surface control
centre
• Improving the underground operation cycles (scheduling)
• Traffic control (signals)
Consequently, approaching high security in the safety and health matters and
improving operation management based on the proposed system will considerably
increase the cost-effectiveness in underground mining projects.
Figure 1-2 Potential functions and applications of the ZigBee network in
underground mines
The limitation of the study is first concerning about ZigBee nodes employed in the
tests. These are laboratory version which have less ability in the feature of antenna
gain compared with updated industrial ones. Second, the outcomes are generalised
based on the underground mine sites of the case studies. Finally, there is an
uncertainty where uncontrollable parameters such as tunnel wall distortion and
roughness, the detail of the dielectric constant and conductivity of rock mass effect
on findings. Also, within the scope of the research topic, there is little prior research
on using system integration of ZigBee and GIS for the applied functions and
applications in underground mines.
CHAPTER 1
INTRODUCTION
|
Curtain
|
8
SIGNIFICANCE AND RELEVANCE
The study proposes a new system integration for the monitoring and communication
of underground mines using ZigBee and GIS. This will significantly improve health and
safety issues and management difficulties in underground mining. The finding of this
study will benefit the engineering profession, considering the important role it plays
in the development of today’s industries. Hence, the greater worldwide demand for
ore deposits has necessitated safer and more productive approaches through the
automation of mining techniques. Therefore, the significance of the study is
summarised as follows:
• This study could be beneficial in this era of system integration engineering where
new technologies are being utilised.
• The study could contribute to the expansion of knowledge of WSNs. The emerging
functions and applications of smart wireless sensors in mining activities could
influence relevant research in fields such as modern computer science, wireless
communication and mobile computing.
• Furthermore, the study could promote mining automation and mitigate the
current predicament of safety and health issues in underground mining.
• The study demonstrates and confirms a reliable system for underground mines
to sense environmental attributes such as temperature, humidity and gases
concentration, to manage ventilation fans, and to text emergency messages as a
supportive communication system.
• The results pave the way for future researchers to think about utilising WSNs for
other functions and applications in underground mining and similar industries.
THESIS STRUCTURE
The logical structure of the thesis is shown in Figure 1-3. The thesis comprises seven
chapters, summarised as follows:
Chapter 1 presents an introductory statement regarding the enhancement of safety
and health, as well as operation management in underground mines. It also includes
the objectives, scope and limitations of the research and offers conclusive and
significant research contributions to the mining industry.
CHAPTER 1
INTRODUCTION
|
Curtain
|
10
Chapter 2 presents the state of the art of WSNs in underground mines, in relation to
the innovation of remote and automatic systems against the backdrop of fewer
skilled and unskilled mining personnel. It also surveys the essential features of ZigBee
networks in underground environments compared to other WSNs, and presents the
relative strengths and weaknesses of ZigBee networks in these environments.
Chapter 3 analyses the attenuation theory of radio waves in underground mines. It
investigates the influences of openings, features and operations of underground
mines on the behaviour of WSNs’ radio wave propagation. Then, it discusses the
measurement of ZigBee communication distances by comparing experiments and
simulations results obtained in an underground mine.
Chapter 4 highlights the placement of the sensing nodes, an important factor in
allowing efficient transmission as well as maximum security throughout the wireless
network. It also investigates an optimal arrangement of ZigBee nodes by creating
various scenarios of mesh and cluster-tree configurations, as well as LQI-related
metrics evaluation in mine tunnels using simulation programmes to assess ZigBee
network performance and security in underground mines.
Chapter 5 provides a design of system integration based on the ZigBee and GIS for
the underground mines monitoring and communication. It develops a trigger action
plan, programmed in the surface GIS management server, for the automatic response
when monitored underground mine attributes exceed normal or threshold value
limits. It also includes a design to turn the auxiliary fans on remotely or automatically
in orange (caution) conditions, and to send evacuation messages to underground
miners in red (unsafe) conditions.
Chapter 6 generalises the ZigBee network model for underground mines. It
demonstrates what controllable and uncontrollable parameters influence a ZigBee
network model design and how a ZigBee network is physically established in an
underground mine. Then it presents a verification of the system design by physical
experiments in the underground mine.
Chapter 7 concludes the study by re-presenting the essential findings, methods and
designs for the establishment of ZigBee networks in underground mines. It then
offers a contribution to the direction of future research.
CHAPTER 1
INTRODUCTION
|
Curtain
|
13
STATE OF ART OF WSNS IN UNDERGROUND
MINES
INTRODUCTION
Underground mining is the process of extracting deep ore-materials from the ground.
Operations could be improved through a more comprehensive means of interaction
between underground and surface communication networks. This two-way
interaction would cover information regarding operations, risk and hazard analysis,
safety management, and service and maintenance supports. Owing to the dangerous
nature of underground mining, techniques of communication to monitor, control and
ameliorate incidents are considered. It is claimed that monitoring and
communication systems in underground spaces can be developed entirely using
ZigBee networks, one of the WSN technologies. However, the reliability and
performance of such systems are analysed in chapters to follow. The conceptual
process of the literature survey of WSNs focusing ZigBee network is illustrated in
Figure 2-1. According to this process, a critical survey of prior research on WSNs for
underground mines is carried out before reasoning how ZigBee networks are adopted
for underground mines in this study and pervious research. The claim is supported by
reviewing and classifying the features of ZigBee networks and developed with
arguments on the strengths of such networks in underground mines. The result of
this process is the promotion of safety and operational management in underground
mines through ZigBee network developments.
The remainder of this chapter is organised as follows: WSNs’ architecture designs and
applications in underground mines are surveyed in the second section. Then,
common underground WSNs are evaluated and compared in order to assess an
efficient WSN. The third section is a survey of essential features of ZigBee networks
in underground environments. Finally, the related strengths and weaknesses of a
ZigBee network establishment are presented.
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
15
equipment such as telephones (1913), and Walkie-Talkies (1966), supported mining
operations considerably and reduced potential risks in underground spaces. Today
communication systems are an integral part of mining and assist underground mining
operations in achieving a greater degree of safety. In fact, assets have been
maximised with the addition of advanced communication techniques compared to
the traditional methods of exchanging information between sub-surface and ground-
level offices.
Remote and automatic systems, operated by fewer skilled and unskilled workers,
have brought more dependable health and safety to the workplace. This has meant
greater cost-effectiveness, better management of technical problems, energy savings,
more real-time responses to accidents, and better environmental monitoring of
underground mines (Fisher & Schnittger, 2012).
The study of wireless sensor networks (WSNs) is one of the most stimulating fields of
computer science research. This has contributed to home and industry
communication and monitoring solutions over the past five decades (Silicon-
Laboratories, 2012), whereas WSNs have only been appraised in the mining industry
during the past twenty years. They are built from a considerable number of low cost
and low power consumption nodes. These nodes are able to collect data and relay it
to the base station or sink. Then, all base stations or sinks transfer the received data
to a coordinator in the control room. These attributes of WSNs’ nodes bring
significant benefits to underground mine monitoring and operations. They are able
to sense underground environmental variations, perform video surveillance, vocal or
message communications, and trace miners and equipment location and operation.
Research on WSNs in underground mines has advanced for gas detection and
collapse prediction in the coal mine industry. This industry has recorded a large
number of events. Reported annual statistics of fatalities from coal mines, especially
in the USA and China, underpin the importance of utilising WSNs to reduce the
number of injuries, deaths and damage to equipment in order to minimise
unexpected costs (Molina, 2011). Underground mining has benefited by the
implementation of WSNs for strategic positioning of miners and equipment (Chehri,
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
16
2008; Wan et al., 2013) and communication in both vocal and visual formats (Junhua
& Xiaozhou, 2012; Li & Lu, 2011).
Studies about the application of WSNs in the underground mining industry started
gaining attention in 2003. The prominence of WSNs in academic and commercial
publications from 2003 to 2013 is illustrated in Figure 2-2. This is based on an
extensive electronic survey of over 382 peer review journal papers, 97 conference
papers, 30 books, 367 dissertations, and 64 newspaper articles and trade materials.
From the survey, it was realised that WSNs have mainly attracted research attention
on the evaluation of WSNs’ performance and their possible applications in
underground mines. As seen in Figure 2-2, students also made an increasing
contribution to the WSN benefits during these years. In particular, journals and
conferences papers as contemporary sources of WSN investigation reached a zenith
of 71 in 2012. These statistics demonstrate that significant academic and commercial
investment has been placed in assessing WSNs over the past decade.
80
70
60
Type of Document:
50
Newspaper and Trade
40 Journal Paper
30 Conf. Paper
Dissertation
20
Book
10
0
Figure 2-2 Studies on wireless sensor network (WSN) in underground mines from
2003 to 2013
Why WSN in underground mines
Increasing safety and health in underground operations is a priority in mining
automation. WSNs have played a significant role in applying automation in the
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
stnemucoD
fo
rebmuN
3002 4002 5002 6002 7002 8002 9002 0102 1102 2102 3102
|
Curtain
|
17
underground mining industry and are able to collect and transmit information that is
characterized by physical and environmental attributes (Wei et al., 2007). WSN
capability can also increase by attaching a variety of sensors for different applications
such as mechanical, thermal, biological, chemical, optical, and magnetic monitoring.
Sensor features can also benefit underground mine networks (Karl & Willig, 2005)
and are largely involved in wireless communication, self-organisation and flexible
topology. They allow easy expansion following mining progression as well (Haifeng et
al., 2011).
I) Ad-hoc network ability: WSNs can be structured as autonomous networks by
automatically setting up a considerable number of nodes via wireless links as well as
not requiring technical infrastructure in establishing WSNs (Buratti et al., 2009). Every
node can be a router for the nearby nodes simultaneously while it runs its performing
standard functions. Such a structure supports a robust monitoring and
communication system comprising many nodes compared with a wired network
which is very likely to fail due to cable damage in harsh and narrow underground
spaces.
II) Self-configuration and self-healing: WSNs provide flexible topology arrangements
and fault management. In the case of node failures or any disconnection in the
network coverage, the system will enable other possible routes to maintain network
functionalities. The self-configuration and self-healing abilities have removed the
dilemma of diagnosing and frequent repairing and monitoring network failure
without compromising mining operations.
III) Scalability: the number of sensor nodes deployed in a WSN can range from
hundreds to thousands in one network. This sensor networking is sufficient to cover
all required underground mine monitoring and communication applications (Al-
Karaki & Kamal, 2004). In such a wide network, data transmission can proceed in a
well-arranged node deployment when an event occurs anywhere within the mine.
IV) Power usage: WSN nodes are usually battery-powered. This enables them to
continue monitoring and communicating regardless of mine site power fluctuations.
There are substantial challenges regarding energy-saving mechanisms such as radio
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
18
optimisation, data reduction, sleep/wake scheduling, energy-efficient routing, and
battery depletion. Furthermore, recently manufactured batteries can operate on
either DC or AC power based on their battery charge or the mine site power supply,
respectively. Switching between battery and the mine site power results in extended
battery life, and nodes are thus able to continue longer data transmission during
power outages (Moridi et al., 2015).
V) Sensor nodes capability: sensors are responsible for sensing environment features
while nodes act as wireless data transmitters through the network and these
combined operations centralise in a single device called a sensor node. Low cost, low
power consumption are prominent capabilities of WSNs sensors providing the
sensing the surrounding environments and interacting through wireless
communication nodes (actors). More importantly, WSN nodes have a reliable
transmission capability with small size, low power, and low complexity of
establishment (Akyildiz & Kasimoglu, 2004). Hence, WSNs are advantageous and
profitable for underground mining.
The primary advantages of wireless networks compared with other communication
techniques are real-time data transfer, and simple and cost-effective establishment
without the need for wire-transmission. Therefore, wireless communication
mitigates problems of line breakage through underground collapse or machinery
malfunction whilst stabilising operational functions.
WSNs architecture design in underground mines
Functional practicality is an advantage of the individual node system architecture of
WSNs in underground mining. As main tunnels in underground mines are kilometres
long, a WSN can enable connections of small branch tunnels networks to the main
tunnel network. Moreover, the network architecture of WSNs is based on restrictions,
sensor node features, and required applications within the system. Each node sensor
performs a sensing task for detecting specific events. For example, a single ZigBee
node can contain up to 240 sensors and run multiple applications (Daintree-Networks,
2006). A sink which also is a specific node is responsible for collecting sensing data
reported from all the sensor nodes located in branch tunnels and transmits data to a
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
20
The architecture design of recent underground WSNs is based mainly on the
technology chosen. For example, ZigBee technology is selected for communication
between nodes and gateways (or sinks) in the high frequency range, while Wi-Fi or
UWB are employed for medium frequency communication. Alternatively, for the
purpose of reducing risks associated with power cables in underground mines, long-
range technologies such as internet or WiMAX have been adopted for the
communication between a control centre and clients’ operators (Gisbert et al., 2013).
WSNs applications in underground mines
The applications of WSNs have to some extent brought significant improvements to
safety and health, risk assessment, enhanced production, and operating costs
reduction in most underground mining aspects. WSNs’ applications directly rely on
sensor technology capability for detecting physical environment attributes. It is also
notable that the ability of real-time data telemetry makes WSNs’ applications vital in
predicting and preventing incidents. Based on the author’s literature review,
applications in underground mines are classified into three groups: monitoring,
communication and tracking. Figure 2-4 illustrates the proportion of WSN application
studies in underground mines.
Figure 2-4 Percentage of conducted studies on WSNs’ applications in underground
mines
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
21
As shown in this chart, 47% and 36% of these studies have concentrated on
monitoring and tracking, whereas only 17% on communication because of numerous
required underground applications in monitoring and tracking. This was impractical
using technologies that existed before WSNs. In addition, the importance of
monitoring and tracking is becoming more evident as they have delivered an
integrated solution to the major concerns in underground mine management,
particularly in coal mines. On the other hand, the successful use of semi-wireless
communication systems such as Leaky Feeder-Based System and Ethernet, and
wireless communication services like Walkie-Talkie System and Bluetooth reduces
research in the area of WSNs. The possible WSNs’ applications in underground mines
that were examined and classified in studies during the last decade are illustrated in
Figure 2-5. The details of each area are investigated in the sections that follow.
Figure 2-5 Possible WSNs’ applications in underground mines during last decade
Environmental monitoring
The most important application of WSNs in underground mines is to help minimise
the effects of natural disasters by monitoring the environment. Various types of
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
22
sensors are being used to measure environment attributes, such as gas concentration,
temperature, humidity, and ground acceleration (geotechnical instrument), to
provide real-time safety and health data from underground operations. Analysis of
the data is vital in mining management and incident prevention. Although other
monitoring systems could be utilised in underground tunnels, WSNs significantly
refine the surveillance for overall care of mining personnel. For example, sensors can
be more easily and densely installed compared with wired ones, and managing a
specified underground area in emergency conditions will be more efficient, based on
the received local detailed data.
Minute sensor nodes also render WSNs more useful to sense dark, dusty and narrow
underground spaces which are travel paths for large mine machinery. These nodes
could be deployed in top corners or ceiling of tunnels. In addition, deploying sensor
nodes, that are in places distant from incidents, helps maintain a durable and
sustainable wireless network. Monitoring the concentration of toxic gases
particularly in coal mines is another beneficial application of these sensors. As they
are able to be developed and employed in wet areas, advantages of WSNs become
more conspicuous in the presence of aquifers and mining operation water by which
chemical or mechanical devices lose their sensitivity.
Studies in underground monitoring using WSNs have widely focused on evaluating
applications of fire detection, concentration of noxious gases, and other
environmental features such as temperature, humidity and air pressure in mines
(Pandit & Rane, 2013; Sun et al., 2010). Continually measuring dangerous gases such
as methane, carbon dioxide and monoxide which can lead to explosions or fire in coal
mines makes WSN research important in the context of underground mining.
Operation management is another perspective of WSN research in underground
mines. In 2009, Li & Liu suggested a Structure-Aware Self-Adaptive (SASA) WSN
system for rapid structure variation detection, caused by collapses in underground
mines, regulating mesh deployment of sensors. As fires are still a major factor causing
fatalities and injuries in the mining industries, research has examined various
methods and WSN types for abrupt fires in underground coal mines (Bhattacharjee
et al., 2012; Wei, 2007). El Kouche Alma'aitah, Hassanein, & Obaia (2013) investigated
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
23
a WSN platform to monitor the erosion conditions of mining machinery to prevent
ignitable sparks. Zhang, Yang, Han, & Kim, (2014) also introduced an integrated
system capable of repeating inspections based on the WSNs and Cable Monitoring
System (CMS) to improve underground coal mine safety.
Operational monitoring through WSNs is also considered to improve safety issues
and operation management in underground mines. For example, Yin (2011)
simulated a program to evaluate underground scrapers’ tele control autonomous
walking, merging WLAN and ZigBee technologies. In addition, Wan et al. (2013)
presented a derailment monitoring system in endless rope continuous tractors. Real-
time monitoring of shuttle car operations, derailment detection, alarming and
parking were expected from this system. The ability of audio sensors equipped with
powerful, high-sensitivity and low-consumption microphones, also supports search
of trapped miners after collapses in tunnels (Akyildiz & Stuntebeck, 2006; Molina,
2011). Furthermore, since air ventilation deficiency in underground mines is a critical
issue to the occupational safety and health for mine personnel, air quality is improved
by adding auxiliary fans to the ventilation system equipped by the automatic or
remote reaction of sensor nodes (Moridi, 2015). It is generally recognized that studies
focusing on operational monitoring should be performed in actual cases to achieve
meaningful results in mine production scheduling.
Finally, an efficient underground monitoring system is usually designed with multiple
requirements of sensing and reacting applications to prevent gas explosions and
collapse (Bhattacharjee, 2012; Li & Liu, 2009; Pandit & Rane, 2013; Wei, 2007).
Communications
Leaky Feeder (LF) is a semi-wireless communication system which has been recently
used for the distribution of voice and data in underground mines. This system has
very costly network establishment and maintenance and lacks standardization.
Furthermore, its functionality to transmit data is limited to line-of-sight transmissions
as transmission cannot pass through solid rock. It also becomes damaged and
inoperative as a result of underground collapses or accidents.
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
24
An alternative solution for underground communication systems based on WSNs are
to incorporate video, picture, voice and text messaging communications (Pandit &
Rane, 2013; Qu et al., 2012; Yarkan et al., 2009). The proposed optical WSNs for
underground communication systems are not reliable for emergency situations due
to the presence of smoke and dust, and the shortage of lighting. Expensive cameras
are also required. Hence, studies have focused more in the ability of voice
communication (Sicignano, 2011) and text messaging (Moridi, 2015) using WSNs in
underground mines.
Researchers have dealt with signal propagation techniques of WSNs in confined
environments like tunnels and underground mines to propose real-time response as
well as provide communication effectiveness in emergency situations. Specific
conditions of tunnels such as rough surface, different shapes, sharpness edges and
curvatures have an influence on signal strength reduction. Under these conditions,
cost-effective designs on the software and hardware of underground voice
communication can be investigated using Wi-Fi under the IEEE 802.11 standard
(Müller & Noack, 2011; VAMVU & BARBU, 2013), and ZigBee under the IEEE 802.15.4
standard (Pandit & Rane, 2013; Yin, 2011).
Tracking
Tracking to estimate miners’ and equipment location in underground mines has been
another challenge. Since WSNs have a potential ability to include such applications,
some theoretical models of sensor node localization have been proposed for
environment monitoring, miners’ localisation and vehicle tracking (Song et al., 2011).
In order to achieve these purposes, sensor deployment strategy involves creating
extensive and dense monitoring and tracking systems. Improvement in the energy
efficiency of nodes in order to prolong the lifespan of WSNs is achieved by this
strategy (Chen et al., 2008; Haifeng, 2011; Wu et al., 2010). In addition, an optimal
positioning algorithm will effectively make the network secure and time saving (Salap
et al., 2009).
There are several factors such as communication mode, location method, topology
and routing protocol which would be considered when estimating methods of
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
25
stationary and mobile node locations (Chen et al., 2009). Thus, positioning algorithms
are categorized using a triangle measuring method based on triangulating arrival
signals parameters, or employing the scene fingerprint method through scene
features and the adjacent method to determine the zone of any moving node. The
triangle measuring methods including signal intensity (RSSI), time of arrival of signal
(TOA), time difference of arrival (TDOA) and angle of arrival (AOA) are widely used
for positioning systems in underground mines (Longkang, 2011).
Analysing studies on localization show that RSSI and TOA methods are also pertinent
because of their precision and their suitable hardware. The RSSI method is cost-
effective and generally used with a DV-Hop algorithm to calculate the distance for a
mobile node from closed anchor nodes in underground mines (Chen, 2009; Wang &
Shen, 2009; Xu et al., 2012). The TOA method has also been investigated for higher
positioning precision (Chehri et al., 2009).
Underground WSNs evaluation
Wireless sensor networks (WSNs) have been utilised in underground mines as a way
to enhance safety and productivity and reduce operational costs (Bhattacharjee,
2012; Chehri, 2009). The common WSNs for monitoring and communication systems
in underground mining are mainly comprised of Bluetooth technology, ultra-
wideband (UWB) technology, Wi-Fi technology and ZigBee technology. The system
architecture of these WSNs in underground mines is illustrated Figure 2-6.
For the investigation of applicable and reliable wireless systems, the main features of
underground WSNs are illustrated in Table 2-1 (Bandyopadhyay et al., 2009;
Bluetooth SIG Inc, 2013; Jinyun et al., 2009; Kawamura et al., 2013). As shown in the
Table 2-1, Bluetooth has a limited applicability because of its short communication
distance between nodes (i.e. requiring a high number of nodes per tunnel) and its
low network capacity. However, UWB meets a sufficient data rate, network capacity
and low power consumption per node but the communication distance restriction
can cause congestion of nodes in tunnels. Thus, traffic routing would be a major
problem in utilising a UWB system.
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
27
Table 2-1 Comparison of common underground WSNs
Parameters Bluetooth UWB Wi-Fi ZigBee
Communication distance
10 < 10 50-100 50-500
(m)
Frequency range (GHz) 2.4 3.1 - 10.6 2.4 or 5 2.4
Data rate (Mbps) 1 100-500 11 250 10-3
Network capacity (nodes) 7 10-500 32 65×536
Power consumption
1-100 30 500 -1000 20-40
(mW)
Complexity High Medium - High High Low
On the other hand, Wi-Fi is a common wireless technology utilised in underground
mines because it has adequate communication distance and a high communication
speed. Some negative aspects of this network are high power consumption nodes,
the need for infrastructure access points for clusters, continuous power supply and
access point connections for cabling. Additionally, there is no multi-hop network
topology between Wi-Fi nodes even though data is capable of being transmitted
between nodes and access point.
ZigBee is a new wireless technology which combines current technical advances
compared with other WSNs for monitoring and communication systems in
underground mines. It has greater communication distances between nodes,
substantial node network, low energy consumption and low complexity, as shown in
Table 1. Also, ZigBee technology has very cost-effective nodes, network installation
and maintenance compared with other underground WSNs. nor does it require any
access point or central node to transmit data between clusters. Although ZigBee
network has very low data rate (250 kbps) for image, voice and video communication,
it is capable in providing networking applications for data transmission between
nodes (node to node relays) with high performance based on numerous hops.
Recently, ZigBee has been used in the field of mine safety for a range of applications
mostly in underground coal mines as an automatic meter reading system, security
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
28
system and for remote control whilst supporting other WSNs (Chehri et al., 2011;
Hongjiang & Shuangyou, 2008). This study proposes a monitoring and
communication system for underground mines based on ZigBee network
performance without other supporting WSNs. In fact, utilising central (sink) nodes or
access points through another WSNs because of high power consumption and cable
damage risk is eliminated. ZigBee is selected for its powerful networking capability
through ad-hoc and multi-hop topology and considerable network capacity and cost-
effectiveness. Based on this system, data from sensors (fixed nodes) and workers and
vehicles (mobile nodes) locations in an underground mine could be transferred to a
surface gateway for monitoring and bilateral communication.
A SURVEY OF ZIGBEE NETWORK IN UNDERGROUND MINES
ZigBee is a new wireless communication technology based on the IEEE 802.15.4
standard. This standard introduces ZigBee as a low rate and low power consumption
technology for wireless personal area network (WPAN). The first ZigBee node was
developed by ZigBee union in 2004 (Longkang, 2011). Due to the benefits of the
features detailed below, ZigBee network is adopted for underground mine
communication systems ZigBee nodes can last six months up to two years. They are
also very low-cost compared with other WSN nodes and qualify for a free licence in
the frequency bands, with the flexible range in operating frequency of 2.4 GHz
around the world and regional bands of 868 MHz in Europe and 915 MHz in the USA.
ZigBee provides reliable infrastructure based on a mechanism of collision avoidance
to prevent competition and conflict in sending data. It also avoids interference with
other ZigBee networks which communicate on the same frequency nearby (Longkang,
2011). ZigBee networks can support up to 65,536 sensor nodes in one system. Real-
time data aggregation enables it to provide bounded delay guarantees, so it is
capable of collecting, processing and transmitting data in a very short delay time. It
is capable of ad-hoc networking. In other words, sensor nodes including stationary
and mobile ones can join the network autonomously and communicate together
without existing infrastructure and central control. In addition, ZigBee nodes
commonly have self-organization and self-healing abilities. Another prominent
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
29
feature of ZigBee is multi-hop communications. Some ZigBee network topologies
such as mesh networking are promoted with multi-hop ability. This allows a wireless
network to forward data by hopping node to node until it will be securely received
by the destination node. This ability, indeed, increases fault tolerance through the
network, as well as having alternative routes for more secure and effective
communications particularly in emergency conditions.
These significant features of ZigBee networks have substantially drawn researchers’
attention of this technology to the area of underground mining. Figure 2-7 illustrates
the trend of different studies in underground mines using ZigBee between 2003 and
2013. The bar chart is extracted from 23 newspapers and magazines, 138 journal
papers, 86 conferences papers, 34 university dissertations, and 19 books including
book chapters. Therefore, in the earlier years of ZigBee development, underground
applications were introduced primarily via newspapers and trading company
advertisements. However, research papers and conferences started to focus on this
technology. There also have been substantial publications in this period such as
dissertations and books on ZigBee, specifically focusing on the context of
underground mining. The trend of studies has still increased after 2013 because of
the electronical development of the ZigBee nodes in terms of functions and the
considerable extension of antenna gain.
40
35
30 Type of Document:
25 Newspaper and Trade
20 Journal Paper
15 Conf. Paper
10 Dissertation
5 Book
0
Figure 2-7 Studies on ZigBee in underground mines
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
stnemucoD
fo
rebmuN
3002 4002 5002 6002 7002 8002 9002 0102 1102 2102 3102
|
Curtain
|
30
In these studies, it is clearly observed that safety was the major aspect considered in
a considerable number of research works on ZigBee technology for underground
mining, particularly coal mines. The main investigations of these studies are analysed
in the following sections.
ZigBee applications
This research presents three categories of underground ZigBee network applications
including environmental monitoring, tracking and communication. Figure 2-8
chronologically illustrates the description of these applications based on analysing
studies from 2003 to 2013.
11%
Environmental monitoring
Tracking
50%
Communication
39%
Figure 2-8 Breakdown of studies conducted on ZigBee applications in underground
mines
As seen in Figure 2-8, more than 50% of studies have mostly focused on monitoring
underground mine coal environment attributes, and 39% of studies investigated the
most accurate algorithms as well as efficient and cost-effective methods for the
positioning of mobile and stationary ZigBee nodes. Lastly, only less than 11% of
studies examined the communication between surface and underground personnel
by ZigBee network. Further explanation with regard to academic work on the ZigBee
network follows.
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
31
The application of ZigBee networks relies on the sensor network applications which
are able to communicate with physical layers of the ZigBee protocol. Sensor networks
may consist of many different types of sensors which can be used for continuous
sensing, event detection, event ID, location sensing, and local control of actuators
(Leccese et al., 2014). The concept of micro-sensing and wireless connection of these
nodes is a promising area for new applications. Underground mine applications based
on stationary and mobile sensor nodes can be categorised as follows:
a) Potential applications of stationary sensor nodes (positioning):
• Secondary (supportive) Communication: based on voice communication or
message texting in emergency conditions
• Environmental monitoring: measuring air quality and quantity as well as
detecting gas and fire
• Ground movement monitoring: the periodic transmission of geotechnical
instrumentation data
• Production management: the real-time transmission of equipment
operational information
b) Potential applications of mobile sensor nodes :
• Emergency mustering: tracking miners and tagging systems
• Traffic management: monitoring mobile plant equipment and using traffic
signals control
• Production measurement: improving the trip cycle for jumbo drills and
longwalls, or configuring the payload of production vehicles
ZigBee stack
The protocol stack of ZigBee networks is comprised of four main layers. The logical
arrangement of these layers is illustrated in Figure 2-9. The application layer and the
network layer are supported under a ZigBee specification, and the media access
control (MAC) and the physical layers (PHY) are supported through the IEEE 802.15.4
standard. ZigBee layers are often formed as upper layers of a ZigBee stack. There are
sublayers, each of which have different functions.
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
32
Figure 2-9 Logical arrangement of a ZigBee stack
I) Application layer: This layer provides a framework for distributed applications and
it is divided into three sub-layers comprised of the application objects, the ZigBee
device object (ZDO) and the application sub-layer. The application objects are the
endpoint software which includes the tasks of each ZigBee node, for example,
periodic reading of environmental conditions. A single ZigBee node can support
between 1 to 240 application objects. At the same time, the ZDO defines the different
roles of ZigBee communication within the network like coordinator, full function
device or end device. It also allows nodes to detect each other and establish a reliable
network. The application sub-layer supports a data delivery service and secures links
for the application objects and the ZigBee device object.
II) Network layer: this layer is responsible for ZigBee network addressing and routing
by broadcasting through a MAC layer. So, this layer makes sure that sent data packets
are being received by the destination node, and joining or re-joining is secure.
III) IEEE 802.15.4: This standard, defined in 2003, specifies two layers of MAC and
PHY for low-rate wireless personal area networks (LR-WPANs).
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
33
IV) Media access control (MAC): The MAC layer provides addressing and channel
access control mechanisms (16 channel) that make it possible for several terminals
or network nodes to communicate within a multiple access network.
V) Physical layer (PHY): Defines the means of transmitting raw bits rather than logical
data packets through a physical link connecting network nodes and converts data
packets to wireless signals (over-the-air) and vice versa. This layer supports three
frequency bands of 2.400 GHz-2.484 GHz globally at a maximum rate of 250 kbps,
902 MHz-928 MHz at 40 kbps of data rate in the United States, and 780.0 MHz -868.6
MHz at 20 kbps in Europe. Link quality and energy detection measurement are also
other functionalities of a PHY layer (Lu, 2011).
ZigBee routing protocols
A routing protocol determines the routes selected between sensor nodes in order to
communicate. Routing as mentioned above is one of the major functions of the
ZigBee network layer which is highly influenced by lower layers of MAC and PHY
layers. Therefore, analysing these lower layers, such as the required communication
range between nodes, plays a significant role in designing an efficient and reliable
routing protocol for WSNs. Routing protocols are commonly categorized into two
groups called proactive and reactive routing.
Under the proactive routing, also known as table-driven route discovery, nodes have
to be set up in a certain topology before establishing the network. Each node ought
to aware of nearby nodes in advance. Thus, nodes become active to discover
surrounding destination nodes before transmitting data packets. Although this is a
very efficient and reliable way to communicate, energy consumption and bandwidth
occupancy are increased as all nodes are automatically updated and are always
discovering other available routes.
The reactive routing, also known as on-demand route discovery, is where a node
discovers routes on demand. Here, a node initiates a route discovery and adopts a
path to a destination node when it as the source node has a packet to deliver. It is
suitable for mobile nodes which have to alter communication topologies and paths
over time through the network. As a result, the lesser consumption of energy and
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
34
bandwidth is a salient feature of this routing protocol, but packet delivery becomes
longer owing to routes discoveries for each transmission (Hamid et al., 2013).
These challenges have drawn academics’ attention to investigate a wide range of new
algorithms and solutions on ZigBee routing optimization largely for decreasing cost
and power consumption and improve the reliability of packet delivery through the
network.
ZigBee network topologies
Topology refers to the configuration of nodes (hardware) that establish a wireless
network and how the data is transmitted within the network. ZigBee networks under
the IEEE 802.15.4 standard support three different node functions including
coordinator (gateway), full-function (router) and end-device. These functionalities
are as follows:
I) Coordinator: This node sets up and controls the network as well as storing
information required from other nodes. It operates as a terminal for other nodes
through the network and is also referred to as a personal area network (PAN)
coordinator.
II) Full-function: This node relays data transmission between the coordinator and
other participating nodes as well as fulfilling duties such as environmental sensing.
This type of node extends network area coverage and strives to maintain
communication routes despite network congestion or possible node failure.
III) End-device: This node only can receive or transmit data. Accordingly, it must be
set up for direct communication with the nearest full-function node or coordinator.
Because of the function of these nodes types, three network topologies namely star,
cluster-tree and mesh (peer-to-peer) are predominantly supported by ZigBee
specifications. Figure 2-10 illustrates how these topologies are differentiated to
establish a network between nodes. First, in the star topology, a PAN coordinator has
a responsibility to communicate with every single node through the network. This
topology is appropriate for systems which need centralized and real-time
communicable applications. Second, in the cluster-tree topology, every full-function
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
36
ZigBee network reliability
Ensuring the reliability of WSNs is mainly concerned with the possibility of temporary
or permanent node failure within the network largely resulting from the deficiencies
of battery-power and radio-based communication. This is of particular importance
when designing an extensive and reliable ZigBee network for underground mine
monitoring and communication systems. To provide higher degree accuracy and
more reliable exchange of data through such a robust network, it needs to be secured
and equipped along several possible routes. To avoid an unreliable communication
in the linkage of sensor nodes, network design is not commonly focused on each
single end-to-end delivery. In fact, the adjustment of network features and higher
layers of application and network layers in a ZigBee stack can definitively improve the
reliability of lower layers of the PHY and MAC layers. For example, different intervals
of periodic readings must be managed for the application layers in order to avoid
overloading or data-remembering failure by lower layers. It also is essential for sensor
node applications to remove unnecessary data records and aggregation.
Consequently, a number of characters such as worth-data priority, reading intervals
management, and environmental effects on radio wave propagation all influence the
network reliability (Baronti et al., 2007).
ZigBee network security
Sensor nodes in a wireless sensor network are limited in their computational power
and communication resources. Due to these strict resource constraints, present
network security mechanisms are inappropriate for this field of operation. Efficient
encryption of measured data can be achieved at the cost of increased overheads in
the length of the message. However, radio communications are the most energy
consuming function performed by these nodes, hence the communications
overheads have to be minimised to achieve system longevity. The security
requirements of wireless sensor networks are defined as data confidentiality,
authenticity, integrity and current validity.
I) Data confidentiality: Data confidentiality means keeping important transmitted
information undisclosed from unauthorised personnel. This is particularly important
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
37
in the case of wireless networks where data is transmitted using a radio frequency
whereby any radio receiver can intercept data. Data confidentiality is usually
achieved by encrypting the information before transmission so that only authorised
personnel can decrypt transmitted information. Encryption is therefore classified into
two categories: symmetric encryption and asymmetric encryption. In symmetric
encryption, a secret key is shared between the authorised parties, while in
asymmetric encryption, the sender encrypts the data with a public key and the
receiver decrypts it using a private key. A strong encryption mechanism not only
prevents message recovery but also prevents uncategorised parties from decoding
even partial information about the message. This property is called semantic security,
which implies that the encryption of the same plaintext two different times should
give two different cipher texts (Perrig et al., 2002).
II) Data authenticity: Data authenticity provides a means to detect messages from
unauthorised nodes thereby preventing such nodes engaging the network, that is,
data authentication allows a receiver to verify that the data is sent by the claimed
sender. This is particularly important in sensor networks where a hostile node can
easily implant a large number of messages into the network (Baronti, 2007) causing
other nodes to process these messages thereby decreasing their power resources.
Therefore, a receiver of these messages should ensure that the message is desired
from an authorised source. Data authentication can be achieved by calculating a
Message Authentication Code1 (MAC) using a shared secret key for the transmitted
data. This MAC is also sent simultaneously with the data. The receiver would also
calculate the MAC for the received data using the shared key, and then compare this
computed MAC value to the one sent by the sender of the data. If the two match,
then the receiver recognises that the data has been sent from a valid sender (Perrig,
2002). This affirms message authenticity.
III) Data integrity: Communications in wireless sensor networks are based on
broadcasts, hence messages can be easily intercepted and/or tampered via audio
reception through the wireless medium. Data integrity provides a way for the
receiver of the message to know if the data has been tampered while in transit by an
attacker (Perrig, 2002). Data integrity is closely related to data authentication since
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
38
the MAC also provides data integrity. The receiver of the data calculates the MAC and
compares it to the one transmitted by the sender. If the two MACs match then it
ensures that the data was not tampered with. In other words, if an adversary has
tampered with the message then the MAC calculated by the receiver cannot be equal
to the MAC that was initially calculated by the sender at the time of sending the
message.
IV) Data freshness: Data freshness ensures that the received data is recent and that
an adversary has not repeated old messages subsequently. Data freshness can be
divided into two categories: weak freshness and strong freshness (Perrig, 2002).
Weak freshness provides partial data ordering which prevents data from being
replayed but carries no delay information (Baronti, 2007). Strong freshness, on the
other hand, uses a request-response model to provide complete ordering of
messages and delay estimation to prevent the data being held by an unauthorised
user. Weak freshness is required for sensor measurements while strong freshness is
required for time synchronisation within the network. One of the most common
methods to provide data freshness is to use a monotonically increasing counter with
every message and reject any messages with old counter values. However, every
recipient would need to maintain a table of the last counter value from every sender.
This method may be unfeasible in wireless sensor networks where the sensor nodes
are memory constrained and would not be able to store such a table for even a
moderately sized network.
DISCUSSION
It is strongly believed that the advantages of ZigBee networks to underground mining
outweigh its disadvantages. On one hand, there are some negative viewpoints in
using ZigBee networks in underground environments. The significant infirmity of it
through WSNs is that it is capable of data transmission with a very low data rate of
250 kbps. Although this is efficient for digital data telemetry, text messages and to
some extent voice messages, video-data transmission might be impossible. This is
because photo and video streams in dusty and dark environments is normally
impractical in underground spaces.
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
39
Another concern regarding an extended ZigBee network might be an increase in
multi-hops. As a result of the lack of central routers and gateways (coordinators),
nodes are themselves routers. Thus, packets follow multi-hop routes and pass via
mobile nodes before arriving at their final destination. This feature causes a serious
vulnerability of wireless communications in underground mines owing to the
possibility of violation of such nodes. Nevertheless, a ZigBee network possesses
various features that enable it to mitigate these dilemmas in underground mines. The
features of ZigBee are analysed in the following paragraphs.
I) Energy-effective: ZigBee is known for very low power consumption in WSNs. For
one thing, a ZigBee network is very low energy-consuming both for node and protocol
(IEEE 802.14.5) alike; see Figure 2-9. A ZigBee node also is much more energy-efficient
with 20-40 mW power usage, compared to a Wi-Fi node with 500-1000 mW power
usage, as shown in Table 1. This key feature definitely prioritises ZigBee with a
debatable alternative choice of Wi-Fi for wireless network installation in industries
and in particular, underground mining. In addition, the recent manufacture of new
ZigBee nodes are associated with the lower power consumption of 1 mW.
ZigBee nodes are known for low power consumption because of efficient energy
usage while transmitting radio signals, and more importantly due to intelligent
battery power management in sleep mode. Such ability enables any ZigBee node to
be programmed so as to switch automatically to the sleep mode when it does not
need to record or transmit data. Power consumption during waiting time to
communicate with surrounding nodes, while it is in the sleep mode, is even negligible.
For example, for an output power of 1 mW of radio transmission, a ZigBee node
normally consumes 75 mA at 3.3 V whereas it increases to 150 mA at 3.3 V for an
output power of 100 mW. In other words, a high-power node consumes twice the
power to transmit a data packet compared with a low-power node (Cirronet, 2007).
In this situation, if these ZigBee nodes are awake only 5% of the time that is very
active period of radio telemetry, the approximate average power consumption would
be 5% as well for both cases. The battery, as the result of this, will have a life span of
five years with a low-power node (1 mW) and four years and nine months with a high-
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
40
power node (100 mW). Now, it is clear why it is claimed that ZigBee nodes with such
a battery power can last many months to several years.
II) Cost-effective: ZigBee technology is most cost-effective for several reasons. Firstly,
it has inexpensive modules compared with Wi-Fi modules. The details of this are
illustrated in Table 2-2, (Rahman, 2014). As seen in this table, the establishment of a
ZigBee network not requiring any access point is a more cost- effective solution for
wireless networking in underground mines.
Table 2-2 Comparison of costs between ZigBee and Wi-Fi networks
WSN Costs
Module: ~ $2.75 - $3.5
Cable: $0
ZigBee
Access point: $0/switch
Module: ~ $8 - $16
Cable: $0
Wi-Fi
Access point/switch: $20 - 50
There are different kinds of ZigBee nodes which give the flexibility to design even
more cost-effective wireless networks. There is a low-cost node with a minimal
memory requirement called Reduce Function Device (RFD). It can only function as a
network device to record and send data, but unable to receive data or data telemetry.
Secondly, a ZigBee network also comes with free licences to broadcast in the
frequency spectrum. Finally, it provides very low maintenance costs owing to
facilities with an inherent configuration and redundancy of nodes within the network.
That is why a ZigBee network is introduced as a low-cost system for monitoring and
communication in underground mines.
III) Frequency range Flexibility: ZigBee utilises the 2.4 GHz frequency band to support
global operation, and affords other regional operations such as 868MHz in Europe
and 915 MHz in the USA. Such flexibility contributes to improve the ZigBee network
adaptability to the specified applications that need stronger output power or where
less energy consumption is required.
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
41
IV) Multi-hop: ZigBee utilises a multi-hop routing that enables nodes to operate as a
relay in order to deliver data from nearby nodes, and pass it to the final node
(coordinator). This means that the range of communication between a node and
coordinator can be extended. Therefore, it is one fundamental component in
underground spaces for long distance wireless communication(Qandour et al., 2014).
V) Ad-hoc network: a ZigBee network is established spontaneously as nodes turn on
and connect, and does not rely on base stations to coordinate the routes of
communication between nodes. As a result, the nodes can be placed anywhere taking
into account the restrictions of underground mine activities and environments, and
the ZigBee network will automatically figure out the routes to communicate. ZigBee
nodes also have self-organization and self-healing abilities to rebuild wireless
networking at high potential node failure in underground tunnels. Furthermore, this
fortifies the advantage of such a dynamic system for more underground mining
applications based on the mobile nodes connecting together or alternatively to the
fixed nodes through the underground wireless network.
VI) Large network capacity: ZigBee connection and communication among 65,536
wireless sensor nodes in one system is another proficient ability compared to other
WSNs used in underground mines. This considerably reduces the costs of network
establishment and maintenance and energy consumption, as well as eliminating
locations and services of system infrastructure through the narrow environments.
However, node capacity is restricted by network coverage, topology structure, and
bandwidth requirements based on the types of applications.
VII) Reliable infrastructure: This is one of the most significant features of ZigBee
networks considered for underground mines. Within such a network, each node
provides reassessing relevant and alternative routes to ensure successful data
delivery to the master point (coordinator). In fact, it includes a mechanism of collision
avoidance to prevent competition and conflict in sending data. This ability of nodes
allows ZigBee to provide a reliable network infrastructure in interference-rich
environments.
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
42
VIII) Real-time data communication: ZigBee is capable of providing bounded delay
guarantees on data delivery that is technically named real-time data delivery in WSNs.
It takes almost 15 ms to complete a bilateral communication between two adjacent
nodes (Li & Zhao, 2009). Such a network performance could certainly revolutionise
underground mining management and safety, particularly in emergency conditions.
IX) Safety: ZigBee provides a data integrity check and authentication function.
Therefore, it is absolutely convincing that the weaknesses of ZigBee for underground
monitoring and communication systems are outweighed by the strengths. These
benefits not only promote health and safety difficulties in underground mines,
particularly in coal mining, but also revolutionise management in underground
mining operations.
CONCLUSION
This chapter made an attempt to discuss and summarize most of the pioneer and the
recent approaches of WSNs’ architecture designs and applications in underground
mines. Clearly, the approaches have undergone an evolution to reach the state of the
art, but there is still a long way to go for a robust and reliable underground space
monitoring and communication system. However, the correlation and linkage
between knowledge scope, existing techniques and practical experiments based on
the past works have been established to prove the possibility of ZigBee networks for
such a vital system in underground mines. It has been found that ZigBee features and
its applications are adapted to the damp, dark and hazardous underground
environments, and they can certainly make an improvement in safety and operations
management in these environments.
CHAPTER 2.
STATE OF ART OF WSNS IN UNDERGROUND MINES
|
Curtain
|
43
ZIGBEE RADIO WAVE PROPAGATION
INVESTIGATION IN UNDERGROUND MINES
INTRODUCTION
The radio waves of WSNs’ communication are more complex in underground mine
environments. As the reliable communication is an effective element for safe and
efficient mining, the analysis of the electromagnetic fields and the investigation of
radio waves attenuation evaluating underground effective factors are essential.
In this chapter, a monitoring and communication system for underground mines
based on ZigBee network performance without supporting other WSNs is proposed.
Use of central (sink) nodes or access points is eliminated because of high power
consumption and cable damage risk. In fact, ZigBee is selected for its powerful
networking capability through ad-hoc and multi-hop topology, its considerable
network capacity and cost-effectiveness. Based on this system, data from sensors
(fixed nodes) and workers and vehicles (mobile nodes) locations in underground mine
could be transferred to a surface gateway for monitoring and bilateral
communication. Wireless network coverage for long distance is mandatory
considering the spatial positions between the surface gateway and ZigBee nodes in
underground mines. Therefore, investigations to prove the proposed system as a
reliable and secure network are delineated as below:
• Stable communication distance for packet delivery
• Evaluation of network metrics
• Accuracy of the position of mobile nodes
Based on expanded knowledge of underground WSNs, the stable communication
distance between ZigBee nodes is analysed in this chapter.
The conceptual procedure of the investigation of ZigBee radio wave propagation in
underground mine in underground mines is illustrated in Figure 3-1. To investigate
the stable communication distance between ZigBee nodes, first the theory of radio
waves propagation models to simulate experimental measurements in the tunnel is
described. Then, the methodology of experimental measurements and simulations
CHAPTER 3.
ZIGBEE RADIO WAVE PROPAGATION INVESTIGATION IN UNDERGROUND MINES
|
Curtain
|
45
(Boutin et al., 2008; Lamminmaki & Lempiainen, 1998). Figure 3-2 illustrates typically
practical measurements for attenuation trend and a theoretical model in a tunnel
channel at the frequency of 2.4 GHz [after Hrovat and Javornik, 2013].
Figure 3-2 Typical attenuation trend for practical measurements and theoretical
model in tunnel [after Hrovat and Javornik, 2013]
In this study, multimode waveguide model is investigated to simulate both near and
far regions from the source by summing the power of rays received from reflections
on the tunnel walls and the source. The excitation plane is used in the geometrical
optics (GO) model of Shooting and Bouncing Ray (SBR) method to analyse different
field distributions and attenuation coefficients (Zhi & Akyildiz, 2010). Based on this
method, rays are first traced from the source then reflected rays from tunnel walls
are calculated. According to this technique, Maxwell’s equations and eigenfunctions,
the received signal power at the coordinate would be obtained by Eq.
3-1: (𝑃𝑃𝑟𝑟 ) (𝑥𝑥,𝑦𝑦,𝑧𝑧)
Eq. 3-1
2
1 𝑒𝑒𝑒𝑒𝑒𝑒𝑛𝑛 −(𝛼𝛼𝑚𝑚𝑚𝑚+𝑗𝑗𝛽𝛽𝑚𝑚𝑚𝑚). 𝑧𝑧
w𝑃𝑃𝑟𝑟 h (e𝑥𝑥r,e𝑦𝑦 ,𝑧𝑧)is =th e𝑃𝑃 t𝑡𝑡 r 𝐺𝐺an𝑡𝑡 s𝐺𝐺m𝑟𝑟 i�tt𝐸𝐸e0r ∑po𝑚𝑚 w,𝑛𝑛 e𝐶𝐶r;𝑚𝑚 ,𝑛𝑛 a.𝐸𝐸n𝑚𝑚d ,𝑛𝑛 (a𝑥𝑥r,e𝑦𝑦 t)h.e𝑒𝑒 antenna gains �of the transmitter
and the
𝑃𝑃
𝑡𝑡r eceiver, respectively. a 𝐺𝐺n𝑡𝑡 d in 𝐺𝐺d𝑟𝑟 icate the field of all significant modes.
𝑚𝑚 𝑛𝑛
CHAPTER 3.
ZIGBEE RADIO WAVE PROPAGATION INVESTIGATION IN UNDERGROUND MINES
|
Curtain
|
46
, , and are the mode intensity in the excitation plane,
𝑒𝑒𝑒𝑒𝑒𝑒𝑛𝑛
𝐶𝐶ei𝑚𝑚ge,𝑛𝑛n fu 𝐸𝐸n𝑚𝑚ct,𝑛𝑛ions,
𝛼𝛼
t𝑚𝑚he𝑛𝑛 attenu 𝛽𝛽a𝑚𝑚ti𝑛𝑛on coefficient and the phase-shift coefficient, as given
by Eqs. 3-2, 3, 4 and 5:
Eq. 3-2
𝐸𝐸0𝜋𝜋 𝑚𝑚𝜋𝜋 𝑛𝑛𝜋𝜋
𝐶𝐶𝑚𝑚,𝑛𝑛 = 𝑚𝑚𝑚𝑚 2 𝑚𝑚𝑚𝑚 2 sin( 2𝑎𝑎 𝑥𝑥0 + 𝜑𝜑𝑥𝑥).cos(2𝑎𝑎𝑦𝑦0 + 𝜑𝜑𝑦𝑦)
𝑎𝑎𝑎𝑎�1−(2𝑎𝑎𝑎𝑎) − (2𝑏𝑏𝑎𝑎)
Eq. 3-3
𝑒𝑒𝑒𝑒𝑒𝑒𝑛𝑛 𝑚𝑚𝜋𝜋 𝑛𝑛𝜋𝜋
𝐸𝐸𝑚𝑚,𝑛𝑛 (𝑥𝑥,𝑦𝑦) ≅ sin( 2𝑎𝑎 𝑥𝑥0 + 𝜑𝜑𝑥𝑥).cos(2𝑎𝑎𝑦𝑦0 + 𝜑𝜑𝑦𝑦)
Eq. 3-4
1 𝑚𝑚𝜋𝜋 2 𝑎𝑎𝑣𝑣 1 𝑛𝑛𝜋𝜋 2 1
𝛼𝛼𝑚𝑚𝑛𝑛 = 𝑎𝑎 (2𝑎𝑎𝑎𝑎) 𝑅𝑅𝑒𝑒 +𝑎𝑎 (2𝑎𝑎𝑎𝑎) 𝑅𝑅𝑒𝑒
�𝑎𝑎𝑣𝑣−1 �𝑎𝑎𝑣𝑣−1
Eq. 3-5
𝑚𝑚𝜋𝜋 𝑛𝑛𝜋𝜋
2 2 2
𝛽𝛽𝑚𝑚𝑛𝑛 = �𝑘𝑘 − (2𝑎𝑎𝑎𝑎) − (2𝑎𝑎𝑎𝑎)
In these formulas, if is even number; if is odd number; if
𝜋𝜋
is odd number a𝜑𝜑n𝑥𝑥 d = 0 𝑚𝑚 if is even numb𝜑𝜑e𝑥𝑥 r.= Th2e t𝑚𝑚unnel cross sectio𝜑𝜑n𝑦𝑦 o=f t0he
𝜋𝜋
𝑛𝑛model is a rectangle sh𝜑𝜑ap𝑦𝑦e= wi2th 𝑛𝑛 width and height, and the origin of a Cartesian
coordinate system is mounted a2t 𝑎𝑎the centre of2 t𝑏𝑏unnel. , and are indicated for
the relative electrical parameters for vertical/horizonta 𝑘𝑘l𝑣𝑣 w 𝑘𝑘aℎlls of t 𝑘𝑘he tunnel and the
wave number as defined by Eqs. 3-6, 7 and 8:
𝜎𝜎𝑣𝑣 Eq. 3-6
𝑎𝑎𝑣𝑣
𝜀𝜀0𝜀𝜀𝑣𝑣+𝑗𝑗2𝑚𝑚𝜋𝜋0
𝑘𝑘𝑣𝑣 = 𝑎𝑎𝑎𝑎 = 𝜀𝜀0𝜀𝜀𝑎𝑎+𝑗𝑗2𝜎𝜎 𝑚𝑚𝑎𝑎 𝜋𝜋0
𝜎𝜎ℎ Eq. 3-7
𝑎𝑎ℎ
𝜀𝜀0𝜀𝜀ℎ+𝑗𝑗2𝑚𝑚𝜋𝜋0
𝑘𝑘ℎ = 𝑎𝑎𝑎𝑎 = 𝜀𝜀0𝜀𝜀𝑎𝑎+𝑗𝑗2𝜎𝜎 𝑚𝑚𝑎𝑎 𝜋𝜋0
Eq. 3-8
𝑘𝑘 wh=
e
re2 𝜋𝜋𝜋𝜋0 �𝜇𝜇
,
0𝜀𝜀0𝜀𝜀𝑎𝑎
and denote the complex electrical parameters for
vertical/h 𝑘𝑘o𝑣𝑣rizon 𝑘𝑘tℎal walls a 𝑘𝑘n𝑎𝑎d the air in the tunnel; , and are the relative
permittivity for vertical/horizontal walls and the air in 𝜀𝜀𝑣𝑣 th 𝜀𝜀eℎ tunne 𝜀𝜀l;𝑎𝑎 , and are
their conductivity; is the permittivity in vacuum space; and 𝜎𝜎𝑣𝑣 i 𝜎𝜎s ℎthe c 𝜎𝜎e𝑎𝑎ntral
frequency of the 𝜀𝜀s0ignal. The three areas are assumed to 𝜋𝜋h0ave the same
permeability .
As seen in
Fig𝜇𝜇u0
re 3-2, the intensity of received signal strength indication (RSSI) is one
of the parameter to analyse the attenuation of radio waves (Chehri et al., 2010). RSSI
CHAPTER 3.
ZIGBEE RADIO WAVE PROPAGATION INVESTIGATION IN UNDERGROUND MINES
|
Curtain
|
47
is an indication of received signal power by a wireless node’s antenna. The unit
conversion between RSSI and received signal power is formulated as Eq. 3-9:
Eq. 3-9
w𝑌𝑌h=er1e0 𝑙𝑙i𝑙𝑙s 𝑙𝑙d1 e0 fi𝑋𝑋ned as RSSI unit in decibels milliwatt (dBm) and is unit for the power
of the received signal in milliwatt (mW) The unit conversion was used to create
Y X
positive values of RSSIs to calculate logarithmical average in this study. The intensity
of the transmitted signal power is also expressed with dBm. The experiments of RSSI
measurements were performed at Angas Zinc Mine in South Australia to evaluate
stable communication distance between ZigBee nodes in underground mines.
EXPERIMENTAL MEASUREMENTS IN TUNNELS
Study area description
Understanding the experimental environment properties is a crucial aspect for the
measurements and simulations of the radio waves attenuation in the underground
mine. Angas Zinc Mine located near Adelaide in South Australia was selected as the
study area. Figure 3-3 illustrates a section view of active and inactive zones in this
mine. The experiments of RSSI measurements were completed in two tunnels of the
inactive mining zones at -160 level and -75 level to avoid any interruption with mining
operations. Tunnel cross sections are arch-shaped with 5.5 m height and 5.5 m width.
The environment properties of the experiments were recorded to evaluate effective
factors on the radio waves attenuation. The tunnels are hosted by the Angas Garnet
Member of the Tapanappa Formation, and the ore-body is mostly composed of zinc
and lead. There is not support system in the most parts of the tunnels due to surround
by hard rocks. The long experiments lines show there are some wet areas caused by
underground water inflow to the tunnels. Also, there are no other facilities such as
cables, pipes, ventilation duct and vehicle access which may affect the propagation
of radio waves.
Apparatus and Setup
ZigBee networks generally consist of apparatus such as coordinators (gateway),
routers and end devices. Coordinator can transmit, receive signals, storage all
CHAPTER 3.
ZIGBEE RADIO WAVE PROPAGATION INVESTIGATION IN UNDERGROUND MINES
|
Curtain
|
49
Experimental procedure
The GSC was connected to the laptop (PC), and the GSRs (GSR1 and GSR2) were
mounted on the tripods with a 1.5 m height to minimize the effect of radio waves
distortion from the surface roughness of the tunnel’s floor and walls. This provides
to present clearly the radio waves behaviours in straight and curved tunnels in the
prototype experiment. Likewise, in real case, sensor nodes would be mounted on the
walls or crown of the tunnel with special spacer to reduce negative effects of surface
on the radio waves propagation. Then, the RSSI measurement between GSR1 and
GSR2 was recorded for a distance of 5 m. The measurements were continued by
increasing the distance between GSR2 and GSR1 at 5 m intervals. For consistency of
the results, the measurements were repeated at least 5 times per each interval.
Figure 3-5 illustrates the procedure of RSSI measurements in the tunnels of Angas
Zinc Mine. The measurements were continued until the power of the received signal
between GSRs was disappeared.
The experiments were performed to measure RSSIs at different openings in straight
and curved tunnels.
Figure 3-5 Experiments procedure to obtain RSSIs in Angas Zinc Mine tunnels
Experimental results
The attenuation of ZigBee radio waves in the underground mine was investigated at
bandwidth 2.4 GHz. The experiments were conducted in straight and curved tunnels
at different levels. Results and interpretations of these experiments are stated as
following sections.
CHAPTER 3.
ZIGBEE RADIO WAVE PROPAGATION INVESTIGATION IN UNDERGROUND MINES
|
Curtain
|
51
The RSSIs trend starts from -44 dBm at the shortest distance between GSRs to -64
dBm at 60 m apart. As a normal procedure, the graph indicates a gradual reduction
of RSSIs at the other points along tunnel. There are two abnormalities in RSSIs trend
in the locations of the junctions and two distribution fields that are discussed in
section 3.5.
Straight tunnel
In the second experiment, the data was collected in a curved tunnel at the -75 level
for up to 100 m length. Figure 3-8 and Figure 3-9 illustrate the layout and results of
radio waves attenuation in this experiment. As shown in Figure 3-8, the line of the
experiment was passing through a NLOS at 0-25 m, LOS at 25-60 m and several NLOSs
between 60 and 100 m. There is an opening at 35-40 m. As seen in Figure 3-9, the
logarithmic average of RSSI values in the curved areas of the tunnel declines sharply
from -45 dBm at the starting point to -63 dBm at 25 m. It is because of the curvature
and the effect of multiple modes in the near region. There is a sudden drop in RSSIs
trend at around 35 m mainly due to the presence of the junction area in the opening
location. The trend of a slow decrease in RSSI values continues in a straight tunnel of
NLOS from 45 to 55 m. After that, there are two main falls in measured RSSIs at 60 m
and 70 m due to the sharp corners of the curvatures. On the right side of the graph,
the decline in RSSIs trend continues gradually due to the field with lower mode
energy in the far region of the source.
Figure 3-8 Layout of radio waves attenuation experiment in curved tunnel
CHAPTER 3.
ZIGBEE RADIO WAVE PROPAGATION INVESTIGATION IN UNDERGROUND MINES
|
Curtain
|
53
summation of rays’ reflections on the walls’ surface and the source. To decrease
significantly the runtime of the calculation, the study area boundary was assumed
adjusting the tunnel walls. The SBR method was employed to trace ray paths through
tunnel geometry solving Maxwell’s equations for consideration of the boundary
conditions.
Figure 3-10 Model geometry and radio waves attenuation in the Angas Zinc Mine
tunnels. (a) Model for straight tunnel and (b) model for curved tunnel
DISCUSSION
Comparison between experiments and simulations
The experimental measurements are validated by simulation results for both
experiments. The comparisons of the results confirm that the junctions of the
branches and the curvatures have a major impact on radio signal propagation. The
extra loss of RSSI values in the junctions occurs because of the sudden fluctuation
and polarization changes in the waves caused by the larger cross section in tunnel
dimension and sharp edges. The tunnel curvatures affect the radio waves
propagation by preventing direct visibility between the transmitter and receiver and
increasing multi-path components. The comparison of the experimental and
simulation results for straight and curved tunnels is illustrated in Figure 3-11.
The curves of the RSSI results could be separated into two parts: the region near the
source with fast attenuation of the signals and the region far from the source with
gradual attenuation. In the former case, fast attenuation may have occurred because
CHAPTER 3.
ZIGBEE RADIO WAVE PROPAGATION INVESTIGATION IN UNDERGROUND MINES
|
Curtain
|
55
Comparison of the measured results
The experimental measurements in straight and curved tunnels are compared to
analysis curves trends for the investigation of stable communication between
wireless nodes. The comparison of the measurements obtained from the
experiments in both tunnels based on the logarithmic trend lines of the RSSI values
via distance are illustrated in Figure 3-12. In these trend lines, a gradual reduction of
RSSI values as a function of distance in both tunnels is concluded. However, in the
curved tunnel the RSSI values are reduced more sharply than in the straight tunnel as
caused mainly by curvatures. According to the trend line equations, the radio waves
attenuation in the curved tunnel is 3.1 times more than in the straight tunnel.
Figure 3-12 Comparison between RSSI measurements in straight and curved tunnels
According to the RSSI measurements as shown in Figure 3-12, A stable
communication distance between ZigBee nodes is calculated in the underground
mine tunnels. It is desirable to have RSSI values over -80 dBm for a stable wireless
communication with the GSR. Based on this information and the trend lines, the
stable wireless communication distances between developed ZigBee nodes by our
group research are determined up to 100 m along straight tunnels and 70 m in the
curved tunnels. These approaches are verified with the calculation of RSSI values at -
80 dBm for the longest distance between GSRs shown in Figure 3-12. Therefore, a
CHAPTER 3.
ZIGBEE RADIO WAVE PROPAGATION INVESTIGATION IN UNDERGROUND MINES
|
Curtain
|
56
stable communication distance for packet delivery as one of the proposed system
proofs based on ZigBee network was verified. Also, it is concluded that 104 GSRs are
required to create stable wireless network for covering the whole levels in Angas Zinc
Mine according to the excavated straight and curved tunnels and decline.
CONCLUSION
Underground wireless sensor networks could significantly improve the efficiency of
environmental monitoring, workers and equipment locations, operational readings
and communication system. In this research was shown that ZigBee is more suitable
for underground wireless monitoring and communication system than the other
underground mine WSNs. The stable communication distance between ZigBee nodes
in underground mine based on the attenuation of radio waves was analysed. To this
end, RSSI experiments in straight and curved tunnels at Angas Zinc Mine were
performed and the results were compared with the simulations of radio waves
attenuation in the tunnels. Evaluation of the experimental and theoretical results
confirmed that the junctions of the branches and the curvatures of the tunnels have
major effects on radio waves propagation. However, in the curved tunnel the RSSI
measurements declined sharply than in the straight tunnel, caused mainly by
curvatures. Regions of the experiments divided into the field near the source with
fast attenuation of the signals due to the congested multiple modes in the near
source and the field far from the source with the gradual attenuation due to the
arrived lower-order modes to the receiver. Finally, the results showed the stable
communication distances between developed ZigBee nodes up to 100 m and 70 m in
straight and curved tunnels, respectively. Consequently, the experiments in this study
prove a stable communication of packet deliveries between ZigBee nodes for
underground monitoring and communication system.
CHAPTER 3.
ZIGBEE RADIO WAVE PROPAGATION INVESTIGATION IN UNDERGROUND MINES
|
Curtain
|
57
PERFORMANCE ANALYSIS OF ZIGBEE
NETWORK IN UNDERGROUND MINES
INTRODUCTION
Wireless sensor networks (WSNs) have recently been proposed for underground
mine monitoring and communication to enhance safety and productivity and so as to
reduce operational costs. Typically, the underground WSNs consist of a few to several
hundred nodes between a surface gateway and specified sensor nodes in the
underground levels. Each node can connect to one or more nodes in order to transmit
data. In particular, the placement of the sensing nodes plays a very important role to
allow for efficient transmission as well as providing maximum security through the
network. It is inevitable for underground WSNs to perform at a high level of network
efficiency with lower energy-consumption and the most cost-effective establishment
and maintenance. Despite the progress of WSNs technologies, they still rely on
infrastructure such as so-called sinks to transfer data from underground sensors to
the management server at the surface.
According to the experiments of developed ZigBee nodes (Moridi, 2015), the study
focuses on the reliability of multi-hop data transmission between nodes in
underground mines. In the following, PAN is technically defined as a low rate-wireless
personal area network (LR-WPAN) in an ad-hoc and self-organising network designed
to serve a variety of applications especially in WSNs. ZigBee, based on IEEE 802.15.4
standard (Chandane et al., 2012), is comprised of PAN Coordinator, coordinator (full-
function device) and end-device. A ZigBee PAN Coordinator forms the only root of
the network. First, it creates the network, and then waits for automatic joining
connections of other nodes. It enables all nodes to communicate within the network
and stores data. Due to a limited communication distance, intermediate coordinator
nodes (full-function devices) are involved to transfer data between sensor nodes (the
actual end-device) and the PAN Coordinator through multi-hop routing. As it is shown
in Figure 2-10 the network architecture of different ZigBee topologies. A full-function
device can sense the environment, as well as communicate with the other nodes. An
end-device is only capable of sensing and sending data to the PAN Coordinator or
CHAPTER 4.
PERFORMANCE ANALYSIS OF ZIGBEE NETWORK IN UNDERGROUND MINES
|
Curtain
|
58
nearest coordinator node. The PAN Coordinator is usually AC powered, while routers
and end-devices are typically battery powered.
ZigBee based on the IEEE 802.15.4 standard has three main types of network
topology for data transmission (the star, the cluster-tree and the peer-to-peer mesh)
as illustrated in Figure 2-10. As seen, end-device nodes may be more beneficial in the
cluster-tree topologies considering energy saving during sleep times, while more full-
function devices have to be employed in mesh topologies as they need to relay the
data of nearby nodes.
A key factor to evaluate the efficiency of the WSNs performance is the routing
protocol. The protocol provides routes for each node (Bhat.M. Subramanya et al.,
2011). Routing is the process of selecting paths within a network to send data from
one node to the nearby nodes.
This chapter aims to evaluate ZigBee network performance and security in
underground mines based on the link quality indication (LQI) for each received signal
or packet using QualNet® 7.3 1. For this purpose, we investigate an optimal
arrangement of ZigBee nodes by creating various scenarios of mesh and cluster-tree
configurations, and LQI-related metrics evaluation in mine tunnels. In the scenarios,
all nodes including the Pan Coordinator, the full-function devices and the end-devices
are assumed to remain stationary. The procedure and methodology of an optimum
arrangement of ZigBee nodes for underground mines is illustrated in Figure 4-1. We
analyse the simulations of the mesh and cluster-tree topologies based on the
network performance metrics of throughput, packet delivery ratio, end-to-end delay,
energy consumption and network security.
BACKGROUND
ZigBee network performance in the perspective of nodes positioning design has
theoretically been developed by numerous research solutions (Chatterjee et al., 2013;
1 QualNet®: http://web.scalable-networks.com/content/qualnet (last accessed 7 September 2015)
CHAPTER 4.
PERFORMANCE ANALYSIS OF ZIGBEE NETWORK IN UNDERGROUND MINES
|
Curtain
|
60
The routing protocols simulation is analysed for the improvement of ZigBee network
performance and applications to select optimal paths to transfer data to the
destination (Narmada & Sudhakara Rao, 2011; Roberts et al., 2013; Sharma & Kumar,
2012; Bhat.M Subramanya et al., 2011; Zen et al., 2008). Routing evaluation is an
important task in ad-hoc networks that do not rely on a pre-existing infrastructure
where the nodes are mobile through the environment. Other studies simulated
different topologies to optimise ZigBee network performance for industrial systems
using stationary nodes (Chandane, 2012; Khan et al., 2013; LAVRIC et al., 2013; Moridi,
2015; Ullo, 2010; Yasin et al., 2013). Reliable and cost-effective networks of ZigBee
topologies require an analysis of quality of services (QoS) metrics such as throughput,
packet delivery ratio, end-to-end delay, energy consumption and network security.
However, even though there are some performance evaluations of ZigBee networks
in underground mines (Bo et al., 2012; Chehri, 2011), the simulation of node
positioning comparing different topologies in such environments is hardly
investigated. In this work ZigBee nodes arrangement considering the mesh and
cluster-tree topologies in underground spaces is analysed based on the analysis of
QoS metrics.
ZIGBEE NETWORK PERFORMANCE METRICS
ZigBee network topologies for the analysis study of optimum nodes arrangement
including the mesh (Peer-to-Peer) and cluster-tree which are challenged in industry
applications are evaluated. Typically, the performance of network topologies are
assessed on the basis of metrics that mainly consist of throughput, packet delivery
ratio, end-to-end delay and energy consumption. In particular any topology involved
with higher throughput and packet delivery ratio, and lower end-to-end delay and
energy consumption is more adequate for ZigBee applications. In these concepts, a
packet is defined as a formatted unit of data carried along a communication channel,
and each packet carries the information that will help it get to its destination. In the
following, we define the basic metrics:
I) Throughput: It is defined as the ability of data packets successfully sent from source
node to destination node in the unit time. In our study, the throughput (bits per
CHAPTER 4.
PERFORMANCE ANALYSIS OF ZIGBEE NETWORK IN UNDERGROUND MINES
|
Curtain
|
61
second) is generated by the ZigBee application within scenario simulation times and
is calculated as Eq. 4-1:
Eq. 4-1
𝑇𝑇𝑝𝑝𝑝𝑝×8
𝑇𝑇 = 𝑇𝑇𝑇𝑇𝑝𝑝𝑝𝑝−𝑇𝑇𝑇𝑇𝑝𝑝𝑝𝑝
where the total packet sent, the time last packet sent and the time first packet sent
are denoted as T, Tlps and Tfps, respectively.
II) Packet delivery ratio: The ratio between the packet number received at the
destination node and the packet number sent by the source node is defined as packet
delivery ratio (PDR).
III) End-to-End delay: Delay or latency through wireless networks is time taken by the
packets to propagate from the source to the destination. The end-to-end packet
delay is comprised of the summation of route discovery (source-processing delay),
queuing (network delay), propagation and transfer time (destination delay). The end-
to-end delay is one of the most critical and fundamental issues for WSNs. Many
applications of sensor networks require an end-to-end delay guarantee for time
sensitive data.
The average end-to-end delay of ZigBee applications for different scenarios is
computed based on the Eqs. 4-2 and 3:
Eq. 4-2
𝑇𝑇𝑡𝑡
𝐴𝐴𝐴𝐴 = 𝑁𝑁𝑝𝑝𝑟𝑟
where the average end-to-end delay, the total of transmission delay of all received
packets and the number of packets received are denoted as AD, Tt and Npr,
respectively.
Eq. 4-3
𝑇𝑇w𝑇𝑇h𝑇𝑇er=e t𝑇𝑇h𝑇𝑇e𝑇𝑇 t−ran𝑇𝑇s𝑇𝑇m𝑇𝑇ission delay of a packet, the time packet received at destination
node and the time packet transmitted at source node are denoted as Tdp, Tpr and
Tpt, respectively.
IV) Energy consumption: Energy efficiency is another critical aspect in the QoS of
WSNs, because nodes are powered by batteries and require time and costs in
recharging once they deployed. Energy consumption of a node in any network
CHAPTER 4.
PERFORMANCE ANALYSIS OF ZIGBEE NETWORK IN UNDERGROUND MINES
|
Curtain
|
63
Coordinator and 12, 20, 30, 40 and 50 nodes located in the shaft and tunnels. The
nodes are selected as coordinator (router) or end device depending on the required
use in the network topology. These scenarios are to simulate a real underground
mine, covering an area of 1000m length and 1000m depth. The remaining simulation
parameters are listed in Table 4-1.
Table 4-1 Simulation parameters and node configurations
Parameter Details
Node placement Stationary
Number of nodes 12, 20, 30, 40 and 50
Network topology Mesh and Cluster-tree
Area of simulation 1000m*1000m
Channel frequency and data rate 2.4GHz and 250kbps
Physical and MAC models 802.15.4 radio
Energy model MicaZ
Battery model Simple linear,1200 mAh
Transmission Power 3 dBm
Antenna model Omnidirectional
Modulation scheme O-QPSK
Routing protocol AODV
Path loss model Two Ray model
Traffic ZigBee application
No. of items and Payload Size 100 and 127bytes
Simulation time 10mins
In the scenarios, the MicaZ model (QualNet7.3, 2014) for the radio interface is
employed. All the nodes in the scenarios are battery-operated devices, and we use a
simple linear battery model for the comparison of the scenarios. Therefore energy is
consumed by those interfaces according to the energy specification of MicaZ model
shown in Table 4-2.
Table 4-2 Specifications of MicaZ energy model
Mode Radio mode Power @ 3V (mW)
Active TX 48.0
Active RX 56.5
Active Idle 10.79
Sleep Sleep 1.50
CHAPTER 4.
PERFORMANCE ANALYSIS OF ZIGBEE NETWORK IN UNDERGROUND MINES
|
Curtain
|
64
Only one PAN co-ordinator is considered as a final sink server to communicate with
other source nodes for data processing and delivery in this multi-hop system. In other
words, a wireless network between the surface PAN coordinator and the
underground sensor are created. The PAN Coordinator and other sensor nodes
including the full function and end devices remain stationary.
Scenarios are separately designed for the mesh and cluster-tree topologies
associated with the different network size including the densities of 12, 20, 30, 40
and 50 nodes given in Table 4-3.
Table 4-3 Simulation scenarios of ZigBee topologies with different network size
Topology Network size (nodes)
12
20
Mesh 30
40
50
12
20
Cluster-tree 30
40
50
Screenshots from the QualNet simulator on 12-node scenarios of the mesh and cluster-
tree topologies are illustrated in Figure 4-2. In these topologies, full-function devices
act as routers to transfer (or relay) data for next source nodes and as a sensor node to
also sense the surrounding environment. An end-device only senses and sends to
nearby nodes. The nodes in the scenarios are manually arranged based on the previous
underground experiments. ZigBee applications defined in the software are used to
evaluate traffic loads between nodes pair with the capability of sending 100 packets,
each packet size having 512 bytes which are active during simulation time.
CHAPTER 4.
PERFORMANCE ANALYSIS OF ZIGBEE NETWORK IN UNDERGROUND MINES
|
Curtain
|
66
RESULTS AND DATA ANALYSIS
The simulation results can be evaluated through various performance metrics in both
the mesh and cluster-tree topologies. By using similar traffic loads, an optimum
ZigBee node arrangement is found for different underground mines. As mentioned
above, the results are analysed based on the performance network metrics of
throughput, packet delivery ratio, end-to-end delay and energy consumption (see
Section 4.3 for the definitions).
I) Throughput
The throughputs between nodes in 12-node scenarios are illustrated in Figure 4-3.
The throughput between source node (SN) and destination node (DN) of (2,1), (3,2),
(4,3), (5,3) in either the mesh or cluster-tree topology is a maximum of 4137 bits/s,
with significant reductions in throughput in the cluster-tree topology compared to
the mesh topology. This is due to simultaneous increase in receiving packets at the
destination nodes (Yasin, 2013). WSNs based on the IEEE 802.15.4 standard
commonly act as displays from sharp throughput drops at higher loads.
Figure 4-3 Throughput versus 12-node scenarios of the mesh and cluster-tree
topologies
The comparison of changes in the number of nodes at the scenarios of 12, 20, 30, 40,
and 50-node with the average throughputs are illustrated in Figure 4-4. The figure
CHAPTER 4.
PERFORMANCE ANALYSIS OF ZIGBEE NETWORK IN UNDERGROUND MINES
|
Curtain
|
67
shows that average throughputs of 3866 and 2079 bits/s are moderately reduced as
the number of nodes increases, with a minimum of 2918 and 1178 bits/s for the mesh
and cluster-tree topologies, respectively. It is also observed that there is an
acceptable throughput within the network for both topologies, however, the mesh
topology performs a better throughput from SNs to DNs due to its path finding
techniques. A drop of throughputs after 12-node scenarios among the mesh topology
has occurred because of rising congestion of packets delivery in full function devices
(coordinators) and because of an increase in the choices of links to nearby nodes and
thus paths through the network.
Figure 4-4 Average throughputs versus varying nodes numbers for the mesh and
cluster-tree topologies
II) Packet delivery ratio
The packet delivery ratios (PDRs) are computed based on a percentage denotes a
ratio between total packets received by DNs and total packets sent from SNs. The
PDRs results for the varying numbers of nodes of the mesh and cluster-tree
topologies are illustrated in Figure 4-5. The PDR in the mesh topology changes slightly
from 81.8% for the 12-node scenario to 77.2% for the 50-node scenario, but it drops
considerably in the cluster-tree topology from 64.5% for the 12-node scenario to 23.4%
for the 50-node scenario. A higher PDR value shows better performance within the
network. Therefore, a visual comparison of the results indicates that the mesh
CHAPTER 4.
PERFORMANCE ANALYSIS OF ZIGBEE NETWORK IN UNDERGROUND MINES
|
Curtain
|
68
topology has a higher network performance at the same traffic loads for the ZigBee
applications.
Figure 4-5 Packet delivery ratios versus varying nodes numbers for the mesh and
cluster-tree topologies
III) End-to-End delay
The average end-to-end delays at each destination node for 12-node scenarios are
illustrated in Figure 4-6. In these bar charts, the node IDs are those as specified in
Figure 4-3. The charts show that end-to-end delays occur at nine destination nodes
in the mesh topology, while it reduces to seven destination nodes in the cluster-tree
topology with the same traffic load. It therefore causes a greater data latency through
the network as a result of the increase in the number of hops, which results in
queuing, channel access delays and transmission delays. As seen in Figure 4-6, there
is no delay for node IDs 5, 9 and 12 in the 12-node scenario of the mesh topology,
while it also does not occur for node IDs of 4, 5, 8, 9 and 12 in the 12-node scenario
of the cluster-tree topology. In fact, the amount of the total delay is reduced with the
increasing number of end-devices through the network.
The tendency of total end-to-end delay of the mesh and cluster-tree network
topologies versus varying number of nodes is illustrated in Figure 4-7. The curves
clearly show that the tendency of end-to-end delay is enhanced with increasing node
CHAPTER 4.
PERFORMANCE ANALYSIS OF ZIGBEE NETWORK IN UNDERGROUND MINES
|
Curtain
|
70
IV) Energy consumption
Next step is evaluating the efficiency of the network by measuring the energy
consumption. Figure 4-8 illustrates the total energy consumption for the mesh and
cluster-tree topologies of ZigBee network versus varying number of nodes. The
trends of the curves in the graph show an increase in energy consumed for more
dense networks. It is also observed that total the energy consumed of 18.4 mWh for
12-node scenario increases to 99.44 mWh for 50-node scenario in the mesh topology,
and it climbs from 15.7mWh for 12-node scenario to 64.2 mWh for 50-node scenario
in the cluster-tree topology. Thus, the cluster-tree topology is more energy efficient
than the mesh topology. This is due to the fact that more end-devices remaining in
sleep mode in the cluster-tree topology. On the other hand, a considerable number
of full-function destination nodes are more engaged in the mesh topology, which
causes higher energy overall consumption. First, such destination nodes have to be
largely in idle mode in order to communicate with nearby nodes. Secondly, the
number of nodes predicted to receive data (receive mode) within a network of the
mesh topology is more necessary than those in the cluster-tree topology.
Figure 4-8 Energy consumption versus varying nodes numbers for the mesh and
cluster-tree topologies
CHAPTER 4.
PERFORMANCE ANALYSIS OF ZIGBEE NETWORK IN UNDERGROUND MINES
|
Curtain
|
71
V) Network security
In a network with mesh (peer-to-peer) topology, all the devices that participate in
relaying the messages are usually full-function devices because end-devices cannot
be as a router and support bilateral communication. The PAN Coordinator might
often be mains powered, while the devices will most likely be battery powered.
Multiple hop communication of the mesh topology with a variety of routing
alternative between nodes provides a higher network security for data delivery
within the network. Underground mine applications such as environment attributes
monitoring and bilateral communication under emergency condition are beneficial
from a higher security of such network topology.
DISCUSSION
The performance investigations of different ZigBee topologies in underground spaces
(mines) are summarised in Table 4-4. The simulation results show that the mesh
(peer-to-peer) topology provides more reliable networking for the arrangement of
ZigBee nodes in underground mine tunnels. This network topology has higher
throughput, packet delivery ratio and network security. Although the cluster-tree
topology is involved with lower end-to-end delay and energy consumption through
the network, such benefits do not play significant roles for underground ZigBee
network communication.
Table 4-4 Comparison of the simulation results of ZigBee topologies reliability in
underground spaces
Metric The reliability of ZigBee network topologies
Mesh Cluster-tree
Throughput
Packet delivery ratio
End-to-end delay
Energy consumption
Network security
As seen in Figure 4-7, the delay time of packet deliveries from the source nodes to
the destination nodes for 12, 20, 30, 40 and 50-node scenarios in the mesh topology
are 3, 6, 11, 16, 19 µs longer than the similar scenarios with the same conditions
CHAPTER 4.
PERFORMANCE ANALYSIS OF ZIGBEE NETWORK IN UNDERGROUND MINES
|
Curtain
|
72
created with the cluster-tree topology, respectively. For actual underground
operations, such a small increase in the end-to-end delay of the mesh topology would
not be a major. In addition, the greater energy consumed through the network will
not be as bad a negative aspect for the mesh topologies, as ZigBee nodes that are
currently in development will be able to switch between battery power and mine
power.
CONCLUSION
The selection of an appropriate network topology is crucial for the nodes
arrangement of the industrial wireless WSNs. In this chapter, the performance of
different network topologies for ZigBee-based WSNs are analysed for underground
mine applications. Then scenarios of the ZigBee mesh and cluster-tree topologies
under the IEEE 802.15.4 standard are investigated in the light of most important
network metrics. Throughput, packet delivery ratio, end-to-end delay, and energy
consumption are evaluated during simulations for varying nodes number including
12, 20, 30, 40 and 50-node scenarios.
In many sensitive industrial applications, the arrangement of wireless sensor nodes
mostly depends on achieving higher throughput, packet delivery ratio and network
security as well as lower latency data and energy consumption. While the cluster-tree
topology meets advantages of lower latency data and energy consumption, the
benefits of the mesh topology are higher throughput, packet delivery ratio and
network security, which are the most significant features for the underground ZigBee
node arrangements. The larger data latency and the slight increase in energy
consumption through the network are no major concerns for underground mines
projects, as the delay increases by only a few µs and future ZigBee nodes will be able
to switch power between battery and mine power. Thus, it is concluded that the
mesh topology enables ZigBee nodes to create an underground space wireless
network that is more secure and delivers a higher quality of service than cluster-tree
topology networks.
CHAPTER 4.
PERFORMANCE ANALYSIS OF ZIGBEE NETWORK IN UNDERGROUND MINES
|
Curtain
|
73
UNDERGROUND COMMUNICATION SYSTEM
INTEGRATION
INTRODUCTION
Underground mine safety and health remain challenging issues in the mining industry.
Death toll statistics in China’s coal mines have gradually reduced from 5798 to 2631
between 2000 and 2009 (Wu, 2011) but fatality still occurs. The number of
occupational mining fatalities in the United States’ underground metal mines has
fluctuated from 40 to 46 during the years 2001 to 2010. Most importantly, 33.8% of
the deaths have resulted from ignitions and explosions of gas or dust (CDC, 2012), in
underground mining. In April 2014, two men were killed when a wall collapsed in an
underground coal mine in New South Wales, Australia. Human errors were concluded
from reports as the most significant reasons for mining fatalities. Thus, safety is
always a significant concern in mining operation. Some studies have recently focused
on improving the health for underground miners. Laney and Attfield (2010) have
drawn attention to the fact that the prevalence of coal workers’ pneumoconiosis or
progressive massive fibrosis increased from 1990 to 2000 among United States
underground miners. Therefore, specific consideration of both safety and health
issues deserves priority in mine operation management and engineering designs to
provide and maintain a safe and healthy workplace. In response to these challenges,
mine automation by new technologies such as wireless sensor network (WSN)
assisted with geographic information system (GIS) has been widely utilised in
underground mines to enhance safety and health, productivity and reduce
operational costs (Bhattacharjee, 2012; Chehri, 2009).
To this end, an integration system is developed to mitigate underground safety and
health concerns. This system based on the development of ZigBee nodes is
introduced to sense the underground mine environment, to regulate ventilation
system and to communicate between surface offices and miners. Thus, reduced
power consumption, near real-time monitoring of the environment and bilateral
communicating between surface and underground personnel are achieved.
Experimental tests were carried out to verify network reliability and security of the
packet delivery in underground mines. The architecture of underground monitoring
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
74
and communication for the system integration is illustrated in Figure 5-1. Temporal
ZigBee data including messages, operation orders, and environmental attribute
readings such as temperature, humidity and gases concentration are transferred to
GIS management server in the surface control centre. The transmitted data is
received and stored by ZigBee program then provided for manipulation in the control
centre. Risk situations are immediately identified and responded through a logical
process of data analysis in the GIS management server before reaching dangerous
(unsafe) levels and accidents occurring. The ventilation system management is also
used for the workplace health and safety compliance and the optimisation of mine
site power usage.
Figure 5-1 Architecture of monitoring and communication system in underground
mines
The remainder of the chapter is organised as follows. The fundamental knowledge of
ZigBee technology and GIS are first described. Then, the implementation and
structure of system integration are demonstrated. Finally, the strategic process of
combining ZigBee data and map information through the GIS management server is
modelled for monitoring, communication and controlling the environmental
attributes in an underground mine. In this chapter, the applications and functions of
the underground mine monitoring and communication systems are considered based
on the capability of developed ZigBee nodes.
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
75
BACKGROUND
The underground WSNs consist of a few to several hundred nodes between a surface
gateway and specified sensor nodes at underground levels (Karl & Willig, 2005).
ZigBee based on IEEE 802.15.4 protocol is a new wireless sensor technology which
has more benefits than other WSNs for underground monitoring and communication
systems (Chen et al., 2012). Even though ZigBee technology provides only a low data
rate, its benefits are low power consumption, very cost-effective nodes, network
installation and maintenance (Shu-guang, 2011). It is also capable of providing
networking applications for data transmission between nodes (node to node relays)
with high performance based on many wireless hops. It does not require any access
point or central node to transmit data between clusters. Significance of ZigBee in
underground mines compared to other WSNs was evaluated in the recent publication
of authors (Moridi et al., 2014).
GIS is new technology used for spatial data analysis in order to capture, store, analyse,
manage, and present data that is linked to locations (ESRI, 2012). GIS allows users to
view, understand, question, interpret, and visualize data in many ways that reveal
relationships, patterns, and trends in the form of maps, globes, reports, and charts.
Web-GIS is an inevitable trend which helps solve the problems of spatial information
integration and sharing in technical aspect of web media (Ghorbani, 2012; Huang et
al., 2010). Recently, researchers have technically focused on the GIS supports for the
management of emergency and unsafe conditions (Kawamura, 2013; Salap, 2009;
Sharifzadeh et al., 2008).
USE OF GIS IN UNDERGROUND MINES
GIS is based on computer programs used for storage, modelling, retrieval, mapping
and analysis of geographic data. In this system, spatial features of a specified
environment are stored and manipulated in a coordinate system, which refers to a
specific place. GIS merges multi-layers of required geographic and spatial data for the
user evaluation, and helps determine the locations and times of possible incidents in
advance. Figure 5-2 illustrates a cycle of GIS to process data and layers for the
purpose of risk assessment in underground mine sites. GIS server is capable of
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
76
managing and processing data for a substantial number of attributes coming from
different sources. It also is able to distribute and share data between users based on
internet or intranet, and data could be saved, manipulated or informed by other users.
Therefore, GIS can decrease the time and cost of sharing geographic data and its
attributes.
Figure 5-2 GIS data process cycle and geographic layers in an underground mine
ZIGBEE AND GIS SYSTEM INTEGRATION
In the challenging environment and changing topology of a mine, reliable and
simplified communication is a high-stake issue with the objectives of safe and
efficient mining operations. Automation of remote and automatic systems has
improved workplace safety and health for miners, yielded cost-effectiveness,
management improvement of technical problems, energy savings, real-time
response to incidents. In response to these challenges, integration of technologies
has a significant role in underground mining automation. According to WSNs’ specific
features of high reliability and multi-hop networking, ZigBee can create an integrated
wireless network between nodes in the underground mine tunnels and the surface
gateway. In this study, ZigBee’s capability of monitoring underground environmental
attributes is combined with geographic information to provide potential applications
in communication, operational and environmental monitoring systems of
underground mining.
In order to achieve such smart underground mine system, integrating maps
information and spatio-temporal data from ZigBee nodes into a database at a control
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
78
The network demanded in an underground mine must be capable of providing
bilateral communications between the surface control centre and all underground
wireless nodes interactively. According to the threshold limit values for the different
variable parameters (V1, V2, …, Vn) of underground mine environment, the
conditions of safe, transient and unsafe were set. Thus, the remote or automatic
countermeasures in a GIS management server were arranged in order to control
ventilation fans and send alert or alarm messages to relevant authorities. Additionally,
immediate texting messages are bilaterally communicated between underground
personnel and the surface operator in emergency conditions.
Based on this system, near real-time monitoring data, remote and automatic controls,
and communication by texting messages have achieved the required safety and
health outcomes and improving underground mining operations. Such achievements
are more efficient for emergency management when system configuration enables
control, monitoring and communication between users in various places connected
by internet medium access.
SYSTEM STRUCTURE
Wireless network setup
The entire system of the tested underground WSN is composed of different ZigBee
nodes such as coordinator, routers and end devices. These products were developed
in collaboration with Tokyo Cosmos Electric Co., Ltd. The JN5148-EK010 kit (Jennic)
stacks were employed to create ZigBee network. The wireless network initially is
created by coordinator (gateway) to join other nodes. A ZigBee coordinator
connected to laptop (PC) using in the experiments is illustrated in Figure 5-4. Bilateral
communication was provided between the coordinator and end devices to send and
receive messages and readings instantaneously taken by their sensors. Routers with
the ability of sensing the environment were employed to relay communication
through the network. In addition, sending and receiving messages and remote
control of ventilation fans are enabled by the surface coordinator based on the
designed software (Figure 5-4).
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
79
Figure 5-4 ZigBee coordinator connected to laptop (PC)
To setup WSNs, power consumption and high reliability of packet delivery are the
most concerns. For the former case, ZigBee nodes are configured to transmit data in
longer periods when the mine is in safe and transient conditions which it is caused to
extend the life of batteries. In latter case, different time intervals are considered for
data delivery of environment sensing to avoid network congestion and possibility of
packets loss. The power usage of direct and alternating currents (DC/AC) for the
ZigBee nodes (except the coordinator) were designed to operate under battery and
mine site power supply, respectively. Thus, alternating currents power usage is
resulted in the extension of battery life, and ZigBee nodes are enabled to continue
long-time data telemetry during power outages at any accident. The ZigBee nodes
can last a few days to several months depends on their data rate and applications.
Sensing environment
The safety and health of coal and metal/non-metal mining operations were raised
considerably as the result of the wireless monitoring of environment attributes.
Digital temperature-humidity compound sensor on-board of each JN5148 with
advanced sensitivity and long-term stability for mine sites is utilised in the system.
Methane, Oxygen, CO2, CO, NOX and SO2 concentration sensors (readers) are easily
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
80
connected to ZigBee nodes to sense the environment. The sensors were configured
the single-line communication to transmit real-time data to the nodes. The
measurement of CO2 concentration in this study was considered to manage safety
and health risks nearby coal strata in coal mines or fumes-filled spaces in metal/non-
metal mines.
Text messaging operators
Developed ZigBee nodes are enabled to connect with laptops and mobile phones for
sending and receiving text messages. Figure 5-5 illustrates portable radio stations to
connect laptop (Tablet) which are designed to be placed in an underground refuge
chamber and mobile phones for emergency purposes. The radio station is getting a
significant role for wireless communication with surface operator during accidents
particularly when cable damage or power outage occurs. Even though, its primary
role is the remote control of ventilation fans. ZigBee nodes were placed in the boxes
to minimise environmental effects on their operation.
Figure 5-5 Portable ZigBee radio stations to communicate between laptop and
mobile phones
Ventilation control
Air ventilation deficiency in underground mines is a critical issue to the occupational
safety and health of mine personnel. Moreover, optimization of the fans power
consumption to supply underground fresh air is considered on ventilation system
design. Therefore, adding auxiliary fans to the ventilation system is economically
required to improve air quality during hot seasons, blasting, any gas leakages and
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
82
Figure 5-7 Designed computer interface to switch ON/OFF the (auxiliary) fans and
receiving/sending messages
Input data
The first step of our designed management server is to communicate with the outside
world to receive required information. Figure 5-8 illustrates data flow sheet and the
variety of input data for the GIS management server. Input datasets in the database
are comprised of map information, ZigBee nodes data, ZigBee text messages, ZigBee
node positions, threshold limit values and contact details. Map query is primary
process of map information to merge and display required features in GIS server to
represent the fundamental layers of underground tunnels, geographically. These
layers are revised according to the progress of underground mining activities. Then,
other input data is analysed and located on the layers for further manipulation.
The quality of input dataset is considered to process and analyse at any particular
database. Consequently, the quality of input data in our designed GIS management
server is divided between long-term and short-term datasets. Maps, ZigBee node
positions, threshold limit values and contact details are determined to be long-term
input data into the database which may be periodically updated. These data are
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
83
stored in attribute tables that are associated with ArcGIS geo-processing models.
ZigBee node data which measured environmental properties of mine tunnels is
derived as short-term (temporal) data. In this case, the datasets of environmental
phenomena such as temperature, humidity and gas concentrations change from time
to time or remain relatively continuous. Therefore, spatio-temporal data models,
which show both spatial and temporal characteristics of environment, are considered
as input data in the GIS management server. The spatio-temporal data is stored and
manipulated in the ArcGIS geo-processing based on the related or joined table
command to digital tables of data collection by ZigBee gateway software.
Figure 5-8 Data flow sheet of integrated system in GIS server
Process strategy
Real-time process strategy for safe working environments involves the combination
of data models and programs in GIS management server to monitor and
communicate underground mine automatically and remotely. A pattern of decision
making in managing spatio-temporal data was modelled as a procedure to monitor
the environment attributes of underground mine tunnels (Figure 5-3). To this end,
near real-time and flexible scheduling strategy was planned to apply the performance
of ZigBee network in an emergency status. An experiment was simulated on real
maps of underground mine with developed ZigBee nodes to control ventilation fans
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
84
(ON/OFF) and text emergency messages from surface control office. A section view
of an underground mine and ZigBee node positions in ArcGIS screen are illustrated in
Figure 5-9. In this model a gateway was located in the surface control office to receive
and transmit data through the underground network. The network is extended by
ZigBee routers between the surface gateway and underground end devices based on
optimised communication ranges. ZigBee End devices were divided to three groups
in this experiment. One is connected to the auxiliary fans to switch them on or off
automatically or remotely. Another is attached to a radio station which enables to
write and read messages. The radio station can be portable or located in underground
refuge room. Lastly, sensor nodes are mounted in working area which sense
environment attributes such as temperature, humidity and gas concentration.
Figure 5-9 A thematic map of an underground mine and ZigBee node positions in
ArcGIS
The transmitted data firstly is stored in the GIS management server located at control
centre. The ability of the map visualisation on GIS (ArcGIS) allows the position and
component of the attributes in underground mine environment to be visually
displayed on the screen. Then, the spatio-temporal data tables stored by ZigBee
software in the database were joined or related to the attribute tables of node
geographic positions in the geo-processing services of GIS management server. A
joining table of spatio-temporal data and geographic node position created in ArcGIS
(ArcMap) is illustrated in Table 5-1. In other words, in this joining table each node
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
85
position is connected to the related and measured variable parameters including
temperature, humidity and gas concentration.
Table 5-1 Storage of transmitted data by ZigBee gateway
Following this, the spatio-temporal data was analysed, modelled and retrieved in the
GIS management server. A geo-processing model based on Python (ArcPy) was
designed to track and control the environmental attributes in different conditions.
Normal and threshold limit values to assess environmental attributes according to
underground mining standards were then derived. Normal and threshold limit values
for the discrete conditions of safe and unsafe statues are presented in Table 5-2.
According to the normal and threshold limit values, the status of working
environment in underground mine were assessed in three conditions of safe (green),
transient (orange) and unsafe (red).
Table 5-2 Threshold limit values for working environments in underground mine
Event procedure conditions
Variables (V i) Transient Unsafe
Safe (Green)
(Orange) (Red)
Temperature (T , T , …, T ),
1 2 n
Ti ≤28 28 < Ti < 40 Ti ≥40
Humidity (H , H , …, H ), Hi ≤75 75 < Hi < 85 Hi ≥85
℃ 1 2 n
Gas concentration for Co
2%
Gi ≤2000 2000 <Gi< 5000 Gi ≥5000
(G , G , …, G ),
1 2 n
𝐩𝐩𝐩𝐩𝐩𝐩
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
86
Finally, a loop of conditional procedures and trigger actions were set. The measured
parameters (spatio-temporal data) were stored while these data are less than or
equal normal limit values (safe condition). The loop was periodically retrieved each
30 minutes in order to consume less power and to extend the battery life of ZigBee
nodes and reduce congestion through the network. Otherwise, a trigger plan was set
for the values mounted in the range of between normal and threshold limits
(transient condition) or greater than threshold limit (unsafe condition). The trigger
action plan applied in the GIS management server to respond the deviation of values
from normality is presented in Table 5-3. In the transient (orange) condition, the
auxiliary fans which had designed for emergency ventilation system would be
automatically or remotely turned on. In this state, the model was also setup to send
alert messages to shift supervisors. The periodic time of data reading in orange state
is reduced to 15 min to ensure the safe and health conditions of underground
environment in the shorter time possible. Emergency (alarm) messages in the event
of unsafe (red) condition would be texted to surface authorise and to underground
personnel for immediate evacuee from the hazardous places. The cycle time of data
acquisition is minimised to 5 minutes in this situation.
Table 5-3 Trigger action response plan
Counter measure implements
Variables (V ) Safe Transient Unsafe
i
(Green) (Orange) (Red)
Reading time interval (min) 30 15 5
• Turn the
Temperature (T 1, T 2, …, T n) • Text message
auxiliary fan(s)
to all to
on
Humidity (H 1, H 2, …, H n) • Next evacuate from
• Text message to
reading unsafe
the shift
Gas concentration (G 1, G 2, place(s)
supervisors
…, G n) • Next reading
• Next reading
Output
Mine safety and health were improved by intelligent maps supporting spatio–
temporal data and coordinate of ZigBee nodes in this experiment. The schematic
representation of integrated system outputs for underground monitoring and
communication is illustrated in Figure 5-10. The final outputs of GIS management
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
87
server are comprised of 3D visualization monitoring of underground mine tunnels
and messages texting for alert and alarm conditions. The web-GIS is another
application supporting the GIS management server to promote the underground
monitoring and communication system.
Figure 5-10 Schematic representation of integrated system outputs for
underground monitoring and communication
Data storage
Data storage and management in the central data repository of server is an essential
part of the integrated system. In fact, all geographic and spatial data are stored and
managed in ArcMap’s geodatabase which accesses to the database at any time over
the long-term. In the geodatabase, organisational structure for storing datasets and
creating relationships between datasets were also provided for further analysis and
interpretation. In addition, a multi-user access is enabled to work and command
orders from different mine site offices.
An integrated data management and documentation to generate geospatial
metadata was another approach of geodatabase automation. Metadata can create
geospatial data document to investigate any genuine or non-genuine claims.
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
88
CONCLUSION
An integrated system based on the WSNs and GIS was introduced to automate
underground mine monitoring and communication. The proposed system enhances
safety and health, operational management and reduces capital costs. Considering
the capability of ZigBee network and ArcGIS, the applications of real-time
underground monitoring (temperature, humidity and gas concentration), ventilation
system control and communication in emergency conditions by surface user would
be achievable. The system is equipped with automatic or remote triggers action plans
for measured environmental attributes. The measured data were classified to three
categorises consisting of normal (green), transient (orange) and unsafe (red)
conditions based on their values compared to normal and threshold limit values. At
normal (green) condition, the measured attributes are below the normal value limits.
The mining operation is continuing as it was and readings are recorded with 30
minute intervals. At the transient (orange) condition, the measurements are between
normal and threshold value limits. In this state, trigger actions are become
automatically active to switch the auxiliary fan on and texting message to shift
supervisors. In addition, reading’s intervals are reduced to 15 minutes in this situation.
At unsafe (red) condition, the measurements are getting greater than threshold value
limits and the system texts messages to all underground personnel for immediate
evacuee from the hazardous places. Reading’s intervals are reduced to 5 minutes.
Furthermore, the system provides multi-users surface operation and 3D visualization
for realistic understanding of underground environment and miners’ conditions, and
it could be a useful approach for high-tech underground mining.
CHAPTER 5.
UNDERGROUND COMMUNICATION SYSTEM INTEGRATION
|
Curtain
|
89
ZIGBEE NETWORK MODEL
GENERALISATION FOR UNDERGROUND MINES
INTRODUCTION
The development of a generalised ZigBee network model is hugely beneficial for the
design of a wireless underground mine monitoring and communication systems. This
is owing to the large variety of networking variables, the rapid technological
advancement of ZigBee nodes, and considerable changes in environmental
parameters from one mine site to another one.
Thus, the recognition of entire variables is a key component for the evaluation of the
reliability of the ZigBee functions and applications in an underground mine. In fact, a
system design and a model of ZigBee network are proposed for the verification of the
reliability of required underground functions and applications. Ben Maissa et al.
(2013) emphasised the necessity of investing WSNs’ performance, based on model
analysis and validation, before handling critical functions by such systems. Stanley-
Marbell et al. (2008) observed the influences of the WSNs’ operation considering the
variables of the hardware, software and physical limits. They focused on the
importance of the recognition of the uncontrollable parameters of the environment
and run-time parameters alike to develop a more realistic model and evaluate the
performance of WSNs under a system model. These works attempted to provide
models of WSNs to predict system properties and challenges associated with cost and
time effectiveness on a real project. This chapter will demonstrate that practical
investigations to confirm and calibrate the results of such system models, in order to
sensibly evaluate controllable and uncontrollable parameters, are crucial in a
heterogeneous environment such as underground excavations.
Having selected and simulated ZigBee networks for monitoring and communication
systems in underground environments, Moridi, Kawamura, Sharifzadeh, Chanda, and
Jang (2014) concluded that recognising and assessing the effective parameters is
crucial in designing a ZigBee network. The efforts of (Zarei et al., 2013) posit a method
for assessing the principal parameters of tunnels water inflow. Accordingly, the
CHAPTER 6.
ZIGBEE NETWORK MODEL GENERALISATION FOR UNDERGROUND MINES
|
Curtain
|
90
controllable and uncontrollable parameters of a ZigBee network and the surrounding
environment are illustrated in Figure 6-1.
There are a considerable number of and localization of nodes and the metrics of the
network, are adjustable for better data telemetry in underground mines. The
uncontrollable parameters are the number of hops, network congestion and
infrequent failures in the reception of data packets. It might be possible to render
these parameters controllable within confined spaces. There are also environmental
variables of tunnels that are uncontrollable in ZigBee network design as opposed to
the known or controllable parameters of tunnel geometry, layout and employed
system support. Such uncontrollable parameters include the rate of water inflow
fluctuation, the degree of wall surface distortion and roughness, the radio frequency
interferences of operating and communication systems, obstacles like dump trucks,
boggers, and air compressors, as well as the variation rate of permeability, dielectric
constants, and conductivity in the surrounding rock mass along openings.
Therefore, a ZigBee network can efficiently be established after determining the
underground effective parameters influencing ZigBee communication signals, and
finding the maximum reliable communication distance between nodes in different
underground openings with all variables. Thus, quantifying all of the above
parameters is a prerequisite for the design of a reliable ZigBee network for
underground openings.
The purpose of this chapter is to generalise a ZigBee network model with a more
comprehensive and realistic representation of a communication and monitoring
system in underground mines. First, the procedure for the establishment of a ZigBee
network in an underground mine is described. Then, a system design and model is
developed based on the classification of results from an experiment undertaken at
an underground mine in Western Australia (an analytical study). After that, another
experiment was designed to physically verify the reliability of the proposed ZigBee
network model in the underground mine.
CHAPTER 6.
ZIGBEE NETWORK MODEL GENERALISATION FOR UNDERGROUND MINES
|
Curtain
|
92
This was be done by the testing system’s functions and applications, for example,
messages texting and controlling ventilation fan operations as a model testing. Finally,
the results of the analytical study and verified experiments are discussed with a
subsequent.
ZIGBEE NETWORK MODELLING IN UNDERGROUND MINES
In order to implement an underground monitoring and communication system,
building a model, considering the determination of required functions and
applications and the recognition of the variables of network metrics and
environmental variables, is necessary for the assessment of technical and economic
evaluations.
System design and modelling
In order to design
Input, process and output for an underground monitoring and communication
system design and model are illustrated in Figure 6-2. Access to basic information
including utilising ZigBee nodes technology, desired applications (such as
environmental monitoring, ventilation management, or type of communication) as
well as mine site details is mandatory for system design and modelling. Normally, for
the verification of a system design, system modelling is utilised (Robinson, 2012).
Therefore, a pilot experiment must be conducted to operationalise the principles
governing the system design and modelling of an underground ZigBee network. The
output of the model will lead to a reliable outcome for the required functions and
applications.
CHAPTER 6.
ZIGBEE NETWORK MODEL GENERALISATION FOR UNDERGROUND MINES
|
Curtain
|
93
Figure 6-2 Diagram of system design and model of an underground mine monitoring
and communication system
Generalising a model
The aim of generalising the ZigBee model in underground mines is to be able to
implement a systematic feasibility study of technical and economic evaluations,
based on the system design and modelled results. This procedure is illustrated in
Figure 6-3. Investigations of the ZigBee network model applied in underground
spaces are empirically verified. Numerous runs and adjustment ZigBee functions and
applications may be required before an adequate and reliable system design is
achieved. Results documentation of the process would undoubtedly be valuable for
the investment justification and to convince mine managers of the benefits of such
an innovative system.
AN UNDERGROUND ENVIRONMENT EXPERIMENT FOR SYSTEM DESIGN
An experiment is designed to investigate the reliability of a wireless underground
mine monitoring and communication system. To this end, the maximum distance of
radio communication is evaluated in different conditions of underground
environments and ZigBee nodes arrangement and location. After analysis of the
measurements and classification of results, the system design for a reliable ZigBee
network can be developed for an underground concept.
CHAPTER 6.
ZIGBEE NETWORK MODEL GENERALISATION FOR UNDERGROUND MINES
|
Curtain
|
96
Experiment setup
The underground attenuation of ZigBee nodes was considered in this experiment.
The measurement of WSN’s communication distance is attained based on the
acceptable radio signal strength of a received data packet in terms of the LQI value.
Ha et al. (2013) held that although RSSI provides a traditional metric for radio
transceivers, LQI is an effective metric which has become more common in the latest
ZigBee transceivers such as Chipcon’s CC2420. LQI values indicate that they are more
reliable for link quality estimation and have a higher correlation with the distance
between two ZigBee nodes compared with RSSI values. The LQI is an integer in the
range 0-255 where 255 represents the strongest signal. The relationship between the
LQI value and the detected power, P, in dBm for the ZigBee node in this experiment
(JN5148), is approximately given by Eq. 6-1.
Eq. 6-1
7×𝐿𝐿𝐿𝐿𝐿𝐿 – 1970
𝑃𝑃 = 20
Eq. 6-1 is valid for 0 < LQI < 255. According to the preceding approach described in
3.5.2, reliable LQI for certain data transmission between ZigBee nodes is assumed to
be greater than 50 (-80 dBm). Therefore, the recorded information in this experiment
would be analysed on this basis.
Experiment methodology
The experiment investigated the attenuation tendency of radio wave intensity. This
includes the estimation method of the maximum distance between ZigBee nodes in
the different underground conditions based on appropriate LQI for being cognisant
of identifiable variables. To this purpose, the test lines were designed where the
distance between the two ZigBee nodes increases continuously at certain intervals
until the LQI drops lower than the specified limit.
The procedure for LQI measurements in Tunnel 11 of the nickel mine are illustrated
in Figure 6-5. In this procedure, the coordinator was connected to the laptop (PC) for
recording data while the ZigBee nodes were mounted on tripods to gauge signal
strength. For consistency of the results, the measurements were repeated at least 5
times for each interval.
CHAPTER 6.
ZIGBEE NETWORK MODEL GENERALISATION FOR UNDERGROUND MINES
|
Curtain
|
103
system in every underground environment considering the unique nature and
circumstance of each.
Table 6-2 Experiment results - Summary of the maximum communication distance
between ZigBee nodes under different conditions
Descriptions
Communication Tunnel plan view
distance (m) Test line
Test line Node positions
Straight line in the tunnel axis 380
Straight line on the tunnel wall 180
Diagonal on the opposite
240
straight line tunnel walls
on the floor of
Straight line 120
tunnel
in the middle of
Curved line 120
curved tunnel
The results are analysed and classified to develop a more rigorous system design of a
ZigBee network taking into account relevant factors in the underground openings.
The effect of tunnel curvature on communication distance is indicated in Table 6-3.
Communication distance leads to a dramatic drop from 380m in a line of sight of radio
propagation compared with 120m in a non-line of sight under the similar conditions
with respect to support system, test line position from the walls and the floor and the
level of the ZigBee nodes. However, tunnel water inflow caused greater attenuation
in the line of sight test.
Table 6-3 Classification of results based on the passageway effect
Radio Commun
Test line propagat Tunnel Support Test line Node Water ication
view ion layout system position level conditions distance
regions (m)
water
line of mesh in the
1.5m inflow
sight straight shotcrete tunnel 380
height through a
(LOS) rock bolt axis
ditch
non-line curved mesh in the
1.5m
of sight and shotcrete tunnel dry 120
height
(NLOS) inclined rock bolt axis
CHAPTER 6.
ZIGBEE NETWORK MODEL GENERALISATION FOR UNDERGROUND MINES
|
Curtain
|
104
Classification of results based on the tunnel walls effect is indicated in Table 6-4. The
communication distance between two ZigBee nodes in a straight tunnel falls rapidly
from 380 m to 180 m where the test line changes the location from the middle of the
tunnel to nearby the walls. This occurred when the tests were performed where there
were similar conditions of tunnel layout, support system, ZigBee node height, and
water inflow through a ditch.
Table 6-4 Classification of results based on the tunnel walls effect
Test line Test line Tunnel Support Node Water Communication
view position layout system level conditions distance (m)
water
in the mesh
straight 1.5m inflow
tunnel shotcrete 380
tunnel height through a
axis rock bolt
ditch
on the mesh water flow
straight 1.5m
tunnel shotcrete through a 180
tunnel height
wall rock bolt ditch
Results classified according to the evaluation of ZigBee nodes level- height effect is
indicated in Table 6-5. ZigBee communication distance in this test reduced
significantly from 380m to 120m mainly because of nodes placed in close proximity
to the floor. In addition to the effect of floor unevenness (increases attenuation), the
tunnel floor has an inclination of 0.3%. For this reason, any line of sight tests where
the distance between ZigBee nodes was greater than 100 m may have been
considered as a non-line of sight test. This explains why the height of nodes
placement becomes an important factor in the system design of a ZigBee network in
underground excavations.
Table 6-5 Classification of results based on the ZigBee nodes level - height effect
Test line Node Test line Tunnel Support Water Communication
view level position layout system conditions distance (m)
water
1.5m in the mesh
straight inflow
heigh tunnel shotcrete 380
tunnel through a
t axis rock bolt
ditch
on in the mesh water flow
straight
the tunnel shotcrete through a 120
tunnel
floor axis rock bolt ditch
Results classified from the evaluation of ZigBee nodes arrangement is indicated in
Table 6-6. The communication distance increases from 180m to 240 m provided that
CHAPTER 6.
ZIGBEE NETWORK MODEL GENERALISATION FOR UNDERGROUND MINES
|
Curtain
|
105
the test line changes from a straight to a diagonal one. This derives from conducting
tests in similar conditions and ZigBee node levels. Thus, the relative positioning of
ZigBee nodes to each other could certainly be a definitive item in optimising an
underground system design in terms of cost and energy efficiency.
Table 6-6 Classification of results based on the nodes arrangement effect
Commun
Nodes
Test line Tunnel Test line Support Water ication
arrang node level
view layout position system conditions distance
ement
(m)
on the mesh water flow
straight straight 1.5m
tunnel shotcrete through a 180
line tunnel height
wall rock bolt ditch
on the Water
mesh
diagon straight opposite 1.5m inflow
shotcrete 240
al line tunnel tunnel height through a
rock bolt
walls ditch
Therefore, it is confirmed that factors such as the passageway, the walls and the floor
of a tunnel and the level and arrangement of ZigBee nodes have major impacts on
radio wave attenuation and consequently on the distance of communication. On the
basis of such experiment results, the design of a ZigBee network becomes more
sensible taking into account a variety of parameters in an underground context.
AN UNDERGROUND ENVIRONMENT EXPERIMENT FOR VERIFYING THE
SYSTEM DESIGN
The purpose of this experiment is to verify the system design of a ZigBee network
that has been based on the classified results obtained from the communication
distance experiment. This experiment also includes an investigation into the
reliability of ZigBee functions and applications, specifically those involving the ZigBee
nodes developed by our research group for bilateral underground mine
communication (via message texting) as well as remote control of ventilation system.
The ZigBee nodes applicable to this experiment are illustrated in Figure 6-16.
Experiment apparatus
The system tested is composed entirely of different ZigBee nodes such as coordinator,
routers and end nodes. These products were developed in collaboration with Tokyo
Cosmos Electric Co., Ltd. The JN5148-EK010 kit (Jennic) stacks were employed to
CHAPTER 6.
ZIGBEE NETWORK MODEL GENERALISATION FOR UNDERGROUND MINES
|
Curtain
|
106
create the ZigBee network. The wireless network is initially generated by the
coordinator (gateway) which invites other nodes to join the network. A ZigBee
coordinator (gateway) connected to the laptop (PC) used in the experiments is
illustrated in Figure 6-16. The coordinator would normally be located in the surface
office to allow users to monitor the underground mine, but was located in the tunnel
for this experiment. Sending and receiving messages and remote control of
ventilation fans are also normally enabled by the surface coordinator.
Figure 6-16 Applicable ZigBee nodes for underground environments
Bilateral communication provides wireless connections between the coordinator and
a radio station (end device) to send and receive messages and data readings taken
and delivered by sensors. It is advantageous to locate the radio station including a
ZigBee node and a tablet in a refugee chamber in the event of an emergency,
particularly where there is a failure of primary communication systems such as
telephones or leaky feeders. The extended capability of the ZigBee node connection
to a cell phone is another communication support option between miners and
refugee chambers or surface operators.
Ventilation fan control is also provided by the ZigBee node using the ability to
wirelessly connect to the fans. A screenshot of the designed program showing on PCs
(laptop and tablet) with ON/OFF switches and receiving/sending messages is
CHAPTER 6.
ZIGBEE NETWORK MODEL GENERALISATION FOR UNDERGROUND MINES
|
Curtain
|
107
illustrated in Figure 6-18. There are separate command icons for each auxiliary fan in
the program. Routers are manufactured with the ability of real-time sensing of the
environment as well as relaying communication signals throughout the network. A
digital temperature, humidity and luminance compound sensor on board of each
JN5148, with advanced sensitivity and continued stability, were utilised for the
experiment. ZigBee nodes were placed in the boxes to minimise adverse
environmental effects on their operation.
Alternating current (AC) power was required for the ZigBee coordinator and radio
station (end-device). Therefore, the test line to establish the ZigBee network selected
based on the availability of power points on level 11. Routers were used which were
compatible with direct current (DC) power between 9 and 32 volts. In this experiment,
a 12-volt battery was used.
Experiment setup
To set up the experiment, the coordinator was first turned on and connected to the
laptop to save and monitor data and also to establish an automatic wireless network
to join routers and end devices to the network. The applicable ZigBee nodes were
arranged at level 11 based on the classification of results of the last experiment to
verify the design of the system. This is illustrated in Figure 6-17. Also, the coordinator
was located in refugee chamber connecting to the power point which was supplied
with 220 volts. The first router’s preferred location is in a line of sight from the
coordinator at the specified maximum distance. Otherwise, an imbalance in the
number of nodes in the system design will affect network performance negatively.
For example, if a ZigBee network with a high density of nodes is used, the ability to
control traffic congestion will be affected as well as impact overall economic costs.
On the other hand, if there are fewer nodes over a greater communication distance,
the reliability of communication would be the main concern for network
performance. As a result, an accurate appraisal of optimally arranged nodes for the
ZigBee network, as was done during the system design experiment, must include
consideration of the controllable and uncontrollable parameters. The ZigBee network
was constructed in accordance with the outcomes that appraisal the total length of
CHAPTER 6.
ZIGBEE NETWORK MODEL GENERALISATION FOR UNDERGROUND MINES
|
Curtain
|
108
the ZigBee network was 365 m and had a total of 5 nodes through the level 11 as
shown in Figure 6-17.
Figure 6-17 Arrangement of ZigBee nodes for the verification of the system design
Experiment results
The radio station was mounted at the end of the test line to communicate with the
coordinator through the other nodes. Successful communication between
coordinator and radio station is illustrated in Figure 6-18 where the messages sent
(as red) and received (as blue) were transmitted from the coordinator to the radio
station. As shown in Figure 6-18, the ControlTerm program was configured both on
the laptop which was connected to the coordinator as well as the tablet connected
to the radio station.
In summary, these experiments demonstrated that wireless sensor networks can
significantly improve the efficiency of underground monitoring regarding personnel,
plant, and equipment location, operational readings, and communications. They also
verified that ZigBee network performance of a carefully designed system is reliable
for underground wireless monitoring and communication systems.
The results show that stable communication distances for ZigBee nodes are
sustainable up to 360 m and up to 120 m in straight and curved tunnels, respectively.
Additionally, the following outcomes were successfully achieved: the real-time
monitoring of underground spaces in terms of temperature, humidity and
CHAPTER 6.
ZIGBEE NETWORK MODEL GENERALISATION FOR UNDERGROUND MINES
|
Curtain
|
111
CONCLUSIONS AND FUTURE WORK
The study set out to explore a reliable monitoring and communication system for
underground mines. The reasons were to make a contribution to mining automation
so as to improve mining safety and health as well as operational management in such
adverse environments. To this end, an integrated system considering new
technologies of wireless sensor networks (WSNs) and GIS was proposed. The main
findings of the thesis and connections across the chapters are illustrated in Figure 7-1.
These findings can be classified as theoretical and empirical discoveries. Theoretical
findings on WSNs in underground environments in the second and third chapters
were illustrated where the selection of an adequate networking through reviewing
the literature and investigating the radio wave propagation and attenuation in
underground openings were performed. The empirical findings of the performance
evaluation, system integration design, and generalising model for ZigBee networks
were examined and analysed in the fourth, fifth and sixth chapters respectively.
The aim of the second chapter of the study clearly sought to answer two questions.
First, what is the history of WSNs in underground mines? As ZigBee network is one
technology of WSNs, it is crucial to review past work utilising other technologies in
underground mines. The evaluation of sensors’ ability for underground mining in the
literature strengthened the importance of this study on the subject. The services of
WSNs were classified in monitoring environmental features, communication, and
target tracking which enable mitigating significant concerns in underground mines.
Second, how ZigBee technology was proposed and how it significantly benefits the
improvement of monitoring and communication systems for underground mines. It
also attempted to analyse the applications, stack, routeing protocols, topologies,
reliability, and the security of ZigBee network. Then, the strengths and weaknesses
of ZigBee network to establish such system in underground spaces were examined.
CHAPTER 7.
CONCLUSIONS AND FUTURE WORK
|
Curtain
|
113
some factors of underground environments affect the behaviour of radio waves and
what is the methodology and simulation of investigating a stable communication
distance in tunnels. Although certain ZigBee nodes that were used in the experiments
with a limited number of tunnels, they did show a certain distance of a stable
communication and the consistency of experimental and simulation results supports
the validity of radio wave equations for underground mine environments.
The evaluation of network performance and security is another attempt that has to
become a precedent in developing a monitoring and communication system in
underground mines. Chapter four covered the answers of which network topology
makes the ZigBee system more secure with appropriate network metrics for the
purpose of underground mine operations. The simulation results showed that mesh
topology is a better option for greater throughput, packet delivery ratio and network
security with lower latency data and energy consumption. The approaches represent
the optimal arrangement of nodes in terms of the economic and technical
assessments for underground mine monitoring and communication systems.
An information system for data collection and a process is required to quickly respond
to underground events when ZigBee network is transferring data. Thus, the ZigBee
network integrated with GIS was proposed in chapter five. It explains how the system
integration utilising ZigBee and GIS for underground mine monitoring and
communication operates. Additionally, automatic and remote trigger action plans for
measured environmental attributes to control mining operations are developed.
Multi-users surface operation and 3D visualisation for the realistic understanding of
the underground environment and miners’ conditions could also have other obvious
implications of this system integration.
To this point, all experimental and simulation results that propose an underground
mine monitoring and communication system were analysed based on the data
collection limited to the certain types of ZigBee nodes for the certain underground
mines. In other words, every underground mine site has its own environmental and
operational parameters which affect radio wave propagation. Also, the progress of
technology in ZigBee node’s features is significant. Therefore, chapter six generalises
a model for monitoring and communication systems at any underground
CHAPTER 7.
CONCLUSIONS AND FUTURE WORK
|
Curtain
|
114
environment using ZigBee networks. This model was then verified by conducting
experiments in an underground nickel mine in Western Australia. The first
experiment was carried out to find maximum and stable communication distance
between ZigBee nodes in the presence of the surrounding parameters. This was the
basis in designing the monitoring and communication system for that underground
mine. The second experiment was performed to evaluate specified ZigBee functions
and applications in the levels of that underground mine in order to verify the system
design in the first experiment. Finally, the generalisation of the model was approved
the reliability of establishment of ZigBee networks in underground mines through
achieving the successful experimental findings.
The future work for the establishment of a reliable monitoring and communication
system must interact between surface and underground operators so it covers all
mobile and fixed functions and applications based on the ZigBee technology. The
prospect of a proposed ZigBee communications system is illustrated in Figure 7-2.
Figure 7-2 Prospect of proposed communications system in underground mine site
This system enables for real-time monitoring and communication between surface
operators and the functions and applications of underground mines. That is, it is able
to join all stationary and mobile sensors with different duties such as data-reading of
CHAPTER 7.
CONCLUSIONS AND FUTURE WORK
|
Curtain
|
115
the geotechnical instrument, tracking of the miners and the plant machinery,
controlling the ventilation system, and managing the traffic signals from the surface.
It is an essential tool for underground mine automation to improve project
management in the era of safety, health, economic, cost and operations. These
approaches consider the sensor node’s abilities and the applications requirement,
and are generally classified as follows:
(a) Safety and health approaches
• Air quality and quantity measurements
• Determination of workers’ locations
• Emergency and safety communications
• Gas detector and fire alarm
• Geotechnical instrumentation
(b) Operations management and control
• Real-time monitoring of underground mining from a surface control centre
• Improving the underground operation cycles (scheduling)
• Traffic control (Signals)
As seen in Figure 7-2, some functions and applications which were not covered in this
study are marked for further investigation. Furthermore, to establish such a robust
and reliable monitoring and communication system on the basis of data telemetry
using ZigBee nodes (without access points), there are still more concepts to be tested
and verified. Particularly, there is a need to analyse the reliability and accuracy of the
ZigBee network for tracking mobile nodes.
CHAPTER 7.
CONCLUSIONS AND FUTURE WORK
|
Curtain
|
127
A simulation program of REMCOM was used to calculate the RSSI values based on the
received power of rays’ reflections on the walls’ surface of tunnels and the source.
Wireless InSite 2.7.0 (June 2013) was utilised as an electromagnetic modelling tool
for predicting the effects of buildings and terrain on the propagation of
electromagnetic waves. It predicts how the locations of the transmitters and
receivers within an urban area affect signal strength.
Wireless InSite is capable of modelling signal propagation for virtually any indoor
environment. Floor plans may be read into Wireless InSite from CAD files, such as DXF,
or they can be created from scratch using the Wireless InSite Floor Plan Editor. This
editor allows the user to create a custom indoor environment by specifying wall
locations, wall heights, ceilings, floors, windows, and doorways. The material
properties of each of these structures can be changed to accurately reflect the real
environment.
The calculations are made by shooting rays from the transmitters, and propagating
them through the defined environment. These rays interact with environmental
features and make their way to receivers. Interactions include reflections from
feature faces, diffractions around objects, and transmissions through features.
Wireless InSite uses advanced high-frequency electromagnetic methods to provide
accurate results over a frequency range from approximately 50 MHz to 100 GHz. The
effects of each interaction along a rays’ path to the receiver are evaluated to
determine the rays’ electric field. At each receiver location, contributions from
arriving ray paths are combined and evaluated to determine predicted quantities
such as electric and magnetic field strength, received power, interference measures,
path loss, delay spread, direction of arrival, impulse response, electric field vs. time,
electric field vs. frequency, and power delay profile.
APPENDIX A
|
Curtain
|
128
Radio wave propagation model in tunnel:
To simulate radio wave propagation in our cases (tunnels), Full 3D Model in REMCOM
was selected because two ray tracing methods are available with the FULL 3D model
including the Shooting and Bouncing Ray (SBR) method and the Eigenray method.
Ray tracing models based on a geometrical optics (GO) approach is investigated in
this study for simulation results. The excitation plane is used in the GO model based
on Sun (2010) study on the evaluation of EM field distribution. According to this
method, the summation of all rays’ powers receiving from all reflection points and
the source are computed.
Tunnel environment model:
The tunnel cross section size is defined with a height of 5.5 and a width of 5.5 ;
the tunnel wall, ceiling, and floor are made of the same material with electrical
𝑚𝑚 𝑚𝑚
parameters (permittivity) = 4 , (conductivity) = 0.01 / ; the tunnel interior is
0
filled with air ( = , = 0 / ). The transmitting power is 3 dBm with the central
𝜀𝜀 0 𝜀𝜀 𝜎𝜎 𝑆𝑆 𝑚𝑚
frequency of 2.4 GHz band. The transmitting and receiving antennas are horizontally
𝜀𝜀 𝜀𝜀 𝜎𝜎 𝑆𝑆 𝑚𝑚
polarized dipoles at the same height. Both antennas of the transmitter and receiver
in the model are defined approximately at the centre of the tunnel width, and the
walls of mine tunnels are presumed smooth.
Simulations:
A. Straight tunnel
Indoor design is used for radio wave investigation for tunnel channel in Wireless
InSite (REMCOM software). In the tunnel case, thickness of walls, floors and ceilings
are assumed 100 m.
2.4 GHz narrowband Sinusoid waveform and a Half-wave dipole antenna are created
in the project. Transmitter point and receiver routs are designed 1.5m above the floor.
APPENDIX A
|
Curtain
|
Abstract
Planners and managers tasked with strategic open pit scheduling in real indus-
trial situations are faced with two key dilemmas.
Firstly, findingoptimalanswersfortheOpenPitProductionSchedulingProb-
lem requires the solution of large combinatorial optimisation problems. Unfor-
tunately, and despite a lot of effort, progress has been slow in finding ways to
solve these large problems satisfactorily (or at all) in a reasonable amount of time
through mathematical programming approaches.
The second issue is that until recently, most analysts and planners have relied
on deterministic inputs to their models and have ignored parameter uncertainty.
In doing this they have assumed in effect that information related to geological
and grade variability, as well as the future behaviour of commodity prices and
other economic factors (amongst many others), are known at the time that the
decision is made. In reality these inputs are highly uncertain and depending on
which specific instances are selected, could lead to vastly different results in pit
designs and mining schedules.
With reference to the first issue, the size problem is commonly dealt with by
either reducing the size of the input data set or by implementing clever formula-
tion and solution strategies.
From the data perspective, a quick and easy strategy with drastic effects is
to remove such data from the model upfront which are superfluous from a geo-
metric point of view. In other cases, an attempt is made to lift the burden of
dealing with large models by re-blocking or aggregating blocks into data sets with
a smaller more manageable number of blocks. This makes it easier to solve the
problem but potentially dilutes the resolution of the input data and results.
ix
|
Curtain
|
PAGEx
Fromthesolutionangle, heuristicmethodsareoftenusedtosolvelargemodels
faster than exact mathematical methods but this sacrifices accuracy and the abil-
ity to prove optimality. Some encouraging advances have been made over the last
decade using mathematical programming methods such as mixed integer linear
programming(MILP)intheformulationandsolvingoftheproblemwithoutjeop-
ardising the resolution of the input data. Key improvements include intelligent
data pre-processing, removing obsolete variables with respect to earliest and lat-
est start times, the use of branching algorithms with relaxed integer constraints,
exploitingstrongbranchingcharacteristicsoftheproblem, theintroductionofim-
provedcuttingplanes, anddecompositionmethodssuchasLagrangianrelaxation.
Thesecondproblem(inputuncertainty)hasreceivedlessattention. Apromis-
ing simulation method (Conditional Simulation) has enabled decision makers in
theminingindustrytogeneratemultiplerealisationsrepresentingthespatialvari-
ability of geological features and attributes such as metal grades and densities of
a deposit. When such simulations are considered together they provide insights
into the estimation of the uncertainty inherent in the geological or grade model
due to limited information. Similarly, the stochastic behaviour of commodity
prices resulting in limitations in accurate forecasting has been captured by meth-
ods such as Monte Carlo simulation and Real Option Valuation.
Unfortunately many of these methods, although bestowing massive improve-
ments compared to historical deterministic models, still result in multiple out-
comes (or scenarios) which ultimately leave the decision maker with the dilemma
of having to select the best option.
In this thesis we first draw on some basic advancements made in the field
of mathematical programming to construct a simplistic MILP formulation for a
small instance of the Open Pit Production Scheduling Problem. We then in-
troduce multiple Conditional Simulation data sets which are successively solved
using the MILP formulation generating candidate solutions to the problem. We
alsointroduceaninterpretativeframeworkwhichcomparesaveragevaluesagainst
accompanying standard deviation. We combine these two metrics into a single
indicator, the coefficient of variation (CV), which we propose to be used to find a
suitable trade off between risk (standard deviation of values) and return (average
of values). This metric can then be used to make sense of candidate solutions and
OptimisedDecision-MakingunderGradeUncertaintyinSurfaceMining
|
Curtain
|
PAGExi
optimisation results derived from multiple grade instances based on Conditional
Simulations. This framework enables the identification of a single or a select
number of most ’attractive’ options by considering the trade off between risk and
return (expressed as the CV).
We then use a version of the MILP formulation to incorporate a Scenario Op-
timisation approach which was developed in the 90’s and applied to projects in
power generation and financial portfolio optimisation. The approach can be used
to solve a stochastic model based on a particular method for combining scenario
solutions into a single feasible and ’robust’ strategy. The results from this ap-
proach are then similarly tested for viability by using the interpretive framework.
Thisthesiscombinesadvancementsmadeinexactmathematicalmethodswith
a probabilistic scenario optimisation approach which incorporates and considers
grade uncertainty while arriving at a single ’best’ or most attractive solution (al-
though not necessarily highest value or lowest risk). The resultant methodology
is tested on a case study of 40 conditional simulations.
The three key contributions of this thesis are:
1. Scenario Optimisation
By combining an exact deterministic MILP optimisation approach based
on Caccetta and Hill (2003) with multiple conditional simulation inputs we
utilised a modified two-stage scenario optimisation approach to generate a
unique solution to an open pit scheduling case study.
This approach has an initial stage during which optimised solutions are
generated using a MILP objective function maximisation. This is then
followed in a second stage by finding a solution which minimises the sum of
the absolute differences between the original solutions derived in stage one
and the current solution applied to the equivalent simulation data set.
This two-stage scenario optimisation approach was modified from Dembo
(1992) who used it in financial portfolio optimisation and in hydroelectric
power generation. Our research indicates that this is the first time this
approach has been used in a mining context.
2. Interpretive Framework using Coefficient of Variation
OptimisedDecision-MakingunderGradeUncertaintyinSurfaceMining
|
Curtain
|
PAGExii
In order to incorporate both risk minimisation and value maximisation into
the decision making criteria when comparing various ‘competing’ candidate
solutions, we developed an interpretive framework which utilises the coef-
ficient of variation (CV) as a measure which includes both factors. The
framework based on the CV enables comparison of multiple optimisation
results and their associated statistics when compared against the underly-
ing data. Further, it provides an intuitive and easy way of identifying those
solutions that offer a favourable trade-off between risk and return.
Although the coefficient of variation is a well known statistical metric for
normalising and comparing different scenarios and options with different
means and standard deviations, our research indicates that this is the first
time the CV has been used in such an interpretive framework and certainly
the first time it has been used in the mining context for comparing optimi-
sation solutions derived from conditional simulations.
3. Data Perturbations
We conducted early exploratory research with a concept that we call ‘data
perturbation’ and that we believe has a lot of merit for further research and
investigation.
Generally, a number of ways can be utilised to generate valid candidate
mining schedules. These can be as basic as starting from the top of a
depositandminingnaivelydownwardsbenchbybench,andassophisticated
as following a true holistically optimised solution.
As long as certain requirements are fulfilled, such as adherence with prede-
termined precedence constraints (or pit slope angles) and compliance with
minimum and maximum annual production capacity constraints, any such
schedule will qualify as a ‘valid’ solution. Of course it could be a partic-
ularly poor economic solution (it might even have a negative NPV) but it
would be permissible as a solution.
One way of generating such a permissible solution is to use a slightly per-
turbed version of the underlying conditional simulation data set to gener-
ate an optimised solution. This can for instance be achieved by applying
a small random modifying factor (for example a uniformly selected factor
that varies between -5 and +5 percent) on the true block value.
Although such a modification is not permitted in generating valid condi-
tional simulation data sets during geo-statistical estimation, it could never-
OptimisedDecision-MakingunderGradeUncertaintyinSurfaceMining
|
Curtain
|
1.1.OVERVIEW PAGE2
considerations. There are many areas within such operations where Operations
Researchingeneralandoptimisationtechniquesspecificallycanbesuccessfullyin-
troducedtogainthemaximumbenefit. Thesetechniquesaremostlyimplemented
on the development and exploitation stages of mining projects and operations.
Areas of application include ore-body modelling and ore reserve estimation, the
design and scheduling of optimum pits, determining optimum blends and cut off
policies, equipment maintenance and fleet optimisation studies to name but a
few.
This thesis however occupies itself with the Open Pit Mining Production
Scheduling Problem. A strategic solution to this particular problem involves the
decision on:
(a) which of the available blocks in a mining block model to extract by surface
mining methods (i.e. mine them or leave them in the ground),
(b) when (in which time period) should the extraction take place, and
(c) what to do with them (send a mined block to waste, to the processing plant
for treatment or to a stockpile for later consideration).
All these decisions need to be made with the objective of maximising the
economic value of such an exercise while satisfying safe wall-slope (precedence)
restrictions, production capacity, market factor and environmental constraints.
Problems such as these are routinely formulated as mathematical models
with pre-determined input values, decision variables and technical or capac-
ity constraints. Such problems are then solved as linear programming (LP) or
integer/mixed-integer programming (IP or MIP) formulations using commercial
software.
One of the key limitations of these models is the high level of uncertainty
in the input values (e.g. prices and ore/metal grades) and assumptions used to
construct the models originally as well as the extent to which this is taken into
account or ignored. In some cases this can be further exacerbated due to the
size of the models and the number of variables required to be solved for models
reaching anything close to realistic sizes.
OptimisedDecision-MakingunderGradeUncertaintyinSurfaceMining
|
Curtain
|
1.2.PROBLEMSTATEMENT PAGE3
1.2 Problem statement
The issue that we are trying to address in relation to the uncertainty problem
can be stated as follows:
Many of the current mine scheduling optimisation approaches ignore the un-
certainty around input data (e.g. grade values, density) thus failing to assess the
inherent risk or robustness of derived solutions. Often when they do incorporate
risk through methods like Conditional Simulation, they provide no method to
interpret results or isolate a single or few most suitable options.
1.3 Previous research
Over the last few decades the Open Pit Mine Production and Scheduling Problem
has been approached from many angles. Sequential methods divide the problem
into components by first generating a solution to the ultimate pit problem, fo-
llowed by generating periodic phases (also called ‘push backs’) and only then
finding schedule solutions to the individual subcomponents. Pioneering work
usingasequentialapproachandfocussingonsolvingtheultimatepitproblemwas
done by Lerchs and Grossmann (1964) in the 60’s using Graph Theory, Dynamic
Programming and Network Flow methods. Numerous researchers subsequently
builton2-Dand3-DversionsandmodificationsofLerchsandGrossman’soriginal
work. Original literature utilising a Network Flow method was proposed by
Picard (1976) and expanded on by various others. Heuristic methods have also
been used to solve the ultimate pit problem with one of the most well known
being the Moving (Floating) Cone Method. Early solutions using the Moving
Cone Method were proposed by David et al. (1974) and Lemieux (1979) and
modifications to the method has been proposed by many others.
More holistic or integrated methods considered the problem as a whole and
attempted to solve the problem without subdivision into parts. These meth-
ods include linear (LP) and mixed integer linear programming (MILP) methods
combined with solution strategies such as branch and bound and cut generation,
as well as decompositions methods such as Lagarangian relaxation. Key liter-
ature from within this genre include Caccetta and Hill (2003), Gaupp (2008),
Askari-Nasab et al. (2010), Darby-Dowman and Wilson (2002), Ramazan and
Dimitrakopoulos (2004), Gershon (1982) and Weintraub et al. (2008).
Compared to the volume of work done on deterministic mine planning and
schedule optimisation, research done on introducing uncertainty into mine plan-
OptimisedDecision-MakingunderGradeUncertaintyinSurfaceMining
|
Curtain
|
1.4.RESEARCHOBJECTIVESANDMOTIVATION PAGE4
ning has been relatively sparse. Since the early 90’s Dimitrakopoulos and various
contributors have been exploring the use of Conditional Simulation as a way
to incorporate uncertainty into the ultimate pit design and production schedul-
ing problem e.g. Journel and Huijbregts (1978), Dimitrakopoulos (1990), Dimi-
trakopoulos (1994) and Armstrong and Dowd (1994).
An interesting approach for incorporating uncertainty using Scenario Optimi-
sation was suggested by Ron Dembo (Dembo, 1992) in the 90’s but never applied
to the mining industry. In this thesis we build on his approach.
1.4 Research Objectives and Motivation
The primary objective of this study can be stated as:
To devise an effective method which enables the inclusion of geological (grade)
uncertainty into the problem formulation to derive a risk-adjusted ‘robust’ solu-
tion (i.e. a solution that involves and acceptable trade off between risk and
return) and to develop an interpretive framework to make sense of the results as
an aid to decision making.
The motivation for this research is that grade variability and uncertainty is
routinely ignored in the compilation of mine planning models. In the cases where
grade risk is introduced through conditional simulations, the exercise involves
the generation of multiple simulations with limited sense-making occurring in the
interpretation of the results often still leaving the decision maker with multiple
options without the capacity to select the ‘best’ or most attractive solution(s).
1.5 Research Methodology
The research methodology is as follows:
(a) Review literature and approaches used in the past to deal with the primary
problem under review.
(b) Forasuitablebasemetaldeposit, formulateadeterministicMILPmodelfor
a small to medium size block model (1,000 to 50,000 blocks) based on pre-
vious research and solve the problem using a commercial solver (CPLEX).
Performpre-processingoninputdatainordertoreducethesizeoftheblock
model and data inputs as much as possible.
OptimisedDecision-MakingunderGradeUncertaintyinSurfaceMining
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.