hash
stringlengths 32
32
| doc_id
stringlengths 5
12
| section
stringlengths 4
595
| content
stringlengths 0
6.67M
|
|---|---|---|---|
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 84.11.3.2 PS deactivate channel request from MS | |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 84.11.3.2.1 Conformance requirement | 3GPP TS 44.318 subclause 8a.8.1:
If the MS needs to deactivate one or more transport channels for a particular domain, it shall send the GA-RRC DEACTIVATE CHANNEL REQUEST message to the GANC and start timer TU5002 for the domain. The MS shall include the IE "CN Domain Identity" and the IE "GA-RRC Cause". The GA-RRC Cause value shall be one of the following:
#0: normal release (e.g., due to inactivity timer timeout)
#115: unspecified failure
3GPP TS 44.318 subclause 8a.8.2:
When the GANC receives the GA-RRC DEACTIVATE CHANNEL REQUEST message, it shall request the selected CN domain to release the identified RABs associated with the MS. The GANC selects the CN domain based on the value of the received IE "CN Domain Identity". Note that the GANC may also request the selected CN domain to release the Iu connection for the MS in this case, based on local policy settings.
3GPP TS 44.318 subclause 8a.8.3:
The GANC normally initiates this procedure when it receives the RAB Assignment message from the CN indicating RAB release; however, the GANC may also initiate this procedure under certain failure conditions.
One or more circuit or packet transport channels may be deactivated using a single instance of the channel deactivation procedure; however, it is not possible to deactivate both circuit and packet transport channels using a single instance of the channel deactivation procedure.
The GA-RRC DEACTIVATE CHANNEL message includes the IE "GA-RRC Cause" with value as follows:
#0: normal event, e.g. deactivate due to RAB release request from CN
#115: unspecified failure
#10: relocation cancelled (e.g., the handover procedure is stopped because the call has been cleared)
3GPP TS 44.318 subclause 8a.8.4:
When the MS receives the GA-RRC DEACTIVATE CHANNEL message, it shall:
- deactivate the CTC(s) or PTC(s) identified in the IE "RAB ID List";
- send a GA-RRC DEACTIVATE CHANNEL COMPLETE message to the GANC.
3GPP TS 44.318 subclause 8a.8.5:
If timer TU5002 expires in the MS, the MS shall release the associated transport channel(s).
Reference(s)
3GPP TS 44.318 subclauses 8a.8.1, 8a.8.2, 8a.8.3, 8a.8.4, 8a.8.5 |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 84.11.3.2.2 Test purpose | To verify that MS is able to request deactivation of a PTC. |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 84.11.3.2.3 Method of test | Initial Conditions
System Simulator:
- 1 GAN cell, default parameters
Mobile Station:
- MS in GA-RRC-CONNECTED state (PS domain) in service of GAN cell, one PTC active
Foreseen Final State of the MS
MS in GA-RRC-IDLE state (PS domain)
Test Procedure
The MS is in GA-RRC-CONNECTED state (PS domain) in service of GAN cell, with one PTC active. MS sends GA-RRC DEACTIVATE CHANNEL REQUEST message to request deactivation of the PTC. SS sends GA-RRC DEACTIVATE CHANNEL message to deactivate the PTC. The MS deactivates the PTC and then replies by a GA-RRC DEACTIVATE CHANNEL COMPLETE message.
Specific test parameters
-
Maximum Duration of Test
1 min.
Expected Sequence
Step
Direction
Message
Comment
MS
SS
1
MS
MS in service with one PTC active on GAN cell
2
GA-RRC DEACTIVATE CHANNEL REQUEST
IE 'CN Domain Identity' indicates PS domain, IE ‘GA-RRC Cause’ = #0. For the (single) PTC specified in the IE 'RAB ID List': IE 'RAB ID'.
3
GA-RRC DEACTIVATE CHANNEL
IE 'CN Domain Identity' indicates PS domain, For the (single) PTC specified in the IE 'RAB ID List': IE 'RAB ID'
4
MS
MS deactivates the PTC
5
GA-RRC DEACTIVATE CHANNEL COMPLETE
IE 'CN Domain Identity' indicates PS domain.
6
MS
MS with GA-RRC connection (PS domain) but no active PTC
7
GA-RRC RELEASE
IE ‘GA-RRC Cause’ = #83
IE 'CN Domain Identity' indicates PS domain
8
GA-RRC RELEASE COMPLETE
IE 'CN Domain Identity' indicates PS domain
MS enters GA-RRC-IDLE state for PS domain |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 84.11.4 PS deactivate channel procedure / negative cases | |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 84.11.4.1 TU5002 timer expires | |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 84.11.4.1.1 Conformance requirement | 3GPP TS 44.318 subclause 8a.8.1:
If the MS needs to deactivate one or more transport channels for a particular domain, it shall send the GA-RRC DEACTIVATE CHANNEL REQUEST message to the GANC and start timer TU5002 for the domain. The MS shall include the IE "CN Domain Identity" and the IE "GA-RRC Cause". The GA-RRC Cause value shall be one of the following:
#0: normal release (e.g., due to inactivity timer timeout)
#115: unspecified failure
3GPP TS 44.318 subclause 8a.8.2:
When the GANC receives the GA-RRC DEACTIVATE CHANNEL REQUEST message, it shall request the selected CN domain to release the identified RABs associated with the MS. The GANC selects the CN domain based on the value of the received IE "CN Domain Identity". Note that the GANC may also request the selected CN domain to release the Iu connection for the MS in this case, based on local policy settings.
3GPP TS 44.318 subclause 8a.8.3:
The GANC normally initiates this procedure when it receives the RAB Assignment message from the CN indicating RAB release; however, the GANC may also initiate this procedure under certain failure conditions.
One or more circuit or packet transport channels may be deactivated using a single instance of the channel deactivation procedure; however, it is not possible to deactivate both circuit and packet transport channels using a single instance of the channel deactivation procedure.
The GA-RRC DEACTIVATE CHANNEL message includes the IE "GA-RRC Cause" with value as follows:
#0: normal event, e.g. deactivate due to RAB release request from CN
#115: unspecified failure
#10: relocation cancelled (e.g., the handover procedure is stopped because the call has been cleared)
3GPP TS 44.318 subclause 8a.8.4:
When the MS receives the GA-RRC DEACTIVATE CHANNEL message, it shall:
- deactivate the PTC(s) or PTC(s) identified in the IE "RAB ID List";
- send a GA-RRC DEACTIVATE CHANNEL COMPLETE message to the GANC.
3GPP TS 44.318 subclause 8a.8.5:
If timer TU5002 expires in the MS, the MS shall release the associated transport channel(s).
Reference(s)
3GPP TS 44.318 subclauses 8a.8.1, 8a.8.2, 8a.8.3, 8a.8.4, 8a.8.5 |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 84.11.4.1.2 Test purpose | To verify that the MS releases the PTC when the TU5002 timer (PS domain) expires. |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 84.11.4.1.3 Method of test | Initial Conditions
System Simulator:
- 1 GAN cell, default parameters
Mobile Station:
- MS in GA-RRC-CONNECTED state (PS domain) in service of GAN cell, one PTC active
Foreseen Final State of the MS
MS in GA-RRC-IDLE state (PS domain)
Test Procedure
The MS is in GA-RRC-CONNECTED state (PS domain) in service of GAN cell, with one PTC active. MS sends GA-RRC DEACTIVATE CHANNEL REQUEST message to request deactivation of the PTC and starts timer TU5002 (PS domain). SS does not respond and timer TU5002 (PS domain) expires. MS deactivates the PTC.
Specific test parameters
-
Maximum Duration of Test
1 min.
Expected Sequence
Step
Direction
Message
Comment
MS
SS
1
MS
MS in service with one PTC active on GAN cell
2
GA-RRC DEACTIVATE CHANNEL REQUEST
IE 'CN Domain Identity' indicates PS domain, IE ‘GA-RRC Cause’ = #0. For the (single) PTC specified in the IE 'RAB ID List': IE 'RAB ID'. MS starts TU5002 (PS domain)
3
SS
SS waits for period longer than TU5002
4
MS
TU5002 (PS domain) expires. MS releases the PTC resources
5
MS
MS with GA-RRC connection (PS domain) but no active PTC
6
GA-RRC RELEASE
IE ‘GA-RRC Cause’ = #83
IE 'CN Domain Identity' indicates PS domain
7
GA-RRC RELEASE COMPLETE
IE 'CN Domain Identity' indicates PS domain
MS enters GA-RRC-IDLE state for PS domain |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90 Text Telephony (TTY) Services | This subclause contains test cases for Text Telephony (TTY) services. |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1 Transmission of CTM Bearer Code | |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.1 Mobile Originated TTY Call | |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.1.1 Conformance requirement | 1) When establishing a mobile originated call with TTY mode enabled in the MS, bit 6 of Octet 3a in the Bearer Capability Information Element shall be '1'.
2) When establishing a mobile originated call with TTY mode disabled in the TTY-compatible MS, bit 6 of Octet 3a in the Bearer Capability Information Element shall be '0'.
Reference(s):
For conformance requirement 1 and 2:
3GPP TS 04.08 / TS 24.008, subclause 10.5.4.5 |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.1.2 Test purpose | 1) To verify that a TTY compatible MS, with TTY mode enabled, correctly sets bit 6 of Octet 3a in the Bearer Capability Information Element to 1 when made to originate a call.
2) To verify that a TTY compatible MS, with TTY mode disabled, correctly sets bit 6 of Octet 3a in the Bearer Capability Information Element to 0 when made to originate a call. |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.1.3 Method of test | 90.1.1.3.1 void |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.1.3.2 Initial conditions | System Simulator:
- 1 cell, default parameters.
Mobile Station:
- The MS is in MM-state “idle, updated” with valid TMSI and CKSN. |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.1.3.3 Final foreseen state of the MS | U0, null. |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.1.3.4 Test Procedure | a) The MS is set to TTY mode using the normal MMI. A mobile originated call is established following the generic call set-up procedure for mobile originating speech calls.
b) After receipt of the SETUP message from the MS, the SS shall disconnect the call.
c) TTY mode is disabled in the MS using the normal MMI. A mobile originated call is established following the generic call set-up procedure for mobile originating speech calls.
d) After receipt of the SETUP message from the MS, the SS shall disconnect the call. |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.1.4 Test requirement | 1) In step a), the MS shall send a SETUP message where bit 6 of Octet 3a of the Bearer Capability Information Element is set to 1.
2) In step c), the MS shall send a SETUP message where bit 6 of Octet 3a of the Bearer Capability Information Element is set to 0. |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.2 Mobile Terminated TTY Call | |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.2.1 Conformance requirement | 1) When establishing a mobile terminated call with TTY mode enabled in the MS, bit 6 of Octet 3a in the Bearer Capability Information Element shall be '1'.
2) When establishing a mobile terminated call with TTY mode disabled in the TTY-compatible MS, bit 6 of Octet 3a in the Bearer Capability Information Element shall be '0'.
Reference(s):
For conformance requirement 1 and 2:
3GPP TS 04.08 / TS 24.008, subclause 10.5.4.5 |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.2.2 Test purpose | 3) verify that a TTY compatible MS, with TTY mode enabled, correctly sets bit 6 of Octet 3a in the Bearer Capability Information Element to 1 when receiving a mobile terminated call.
4) To verify that a TTY compatible MS, with TTY mode disabled, correctly sets bit 6 of Octet 3a in the Bearer Capability Information Element to 0 when receiving a mobile terminated call. |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.2.3 Method of test | 90.1.2.3.1 void |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.2.3.2 Initial conditions | System Simulator:
- 1 cell, default parameters.
Mobile Station:
- The MS is in MM-state “idle, updated” with valid TMSI and CKSN. |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.2.3.3 Final foreseen state of the MS | U0, null. |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.2.3.4 Test Procedure | a) The MS is set to TTY mode using the normal MMI. A mobile terminated call is established following the generic call set-up procedure for mobile terminating speech calls.
b) After receipt of the CALL CONFIRMED message from the MS, the SS shall disconnect the call.
c) TTY mode is disabled in the MS using the normal MMI. A mobile terminated call is established following the generic call set-up procedure for mobile terminating speech calls.
d) After receipt of the CALL CONFIRMED message from the MS, the SS shall disconnect the call. |
683b5b8a98f7b1390ddd5516ea9247a2 | 51.010-1 | 90.1.2.4 Test requirement | 1) In step a), the MS shall send a CALL CONFIRMED message where bit 6 of Octet 3a of the Bearer Capability Information Element is set to 1.
2) In step c), the MS shall send a CALL CONFIRMED message where bit 6 of Octet 3a of the Bearer Capability Information Element is set to 0. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 1 Scope | The present document contains the results from the study of improvements for Machine-type Communications in GERAN.
The following items are in the scope of the study:
- GERAN enhancements for Smart metering
- Enhancements which enable or improve efficient use of RAN resources and/or which lower complexity when a large number of MTC devices are served.
- GERAN enhancements for overload and congestion control on the radio, A and Gb interface
- GERAN enhancements regarding identifiers used for MTC devices in the radio access network |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 2 References | The following documents contain provisions which, through reference in this text, constitute provisions of the present document.
- References are either specific (identified by date of publication, edition number, version number, etc.) or non‑specific.
- For a specific reference, subsequent revisions do not apply.
- For a non-specific reference, the latest version applies. In the case of a reference to a 3GPP document (including a GSM document), a non-specific reference implicitly refers to the latest version of that document in the same Release as the present document.
[1] 3GPP TR 21.905: "Vocabulary for 3GPP Specifications".
[2] 3GPP TS 22.368: "Service requirements for Machine-Type Communications (MTC); Stage 1".
[3] SP-100224: "Liaison Statement: Prioritization of NIMTC functions in Rel-10".
[4] 3GPP TS 44.018: "Mobile radio interface layer 3 specification; Radio Resource Control (RRC) protocol".
[5] 3GPP TS 45.005: "Radio transmission and reception".
[6] 3GPP TR 25.942: "Radio Frequency (RF) system scenarios". |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 3 Definitions, symbols and abbreviations | |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 3.1 Definitions | For the purposes of the present document, the terms and definitions given in TR 21.905 [1] and the following apply. A term defined in the present document takes precedence over the definition of the same term, if any, in TR 21.905 [1].
MTC Device: A MTC Device is a UE equipped for Machine Type Communication, which communicates through a PLMN with MTC Server(s) and/or other MTC Device(s).
NOTE: A MTC Device might also communicate locally (wirelessly, possibly through a PAN, or hardwired) with other entities which provide the MTC Device “raw data” for processing and communication to the MTC Server(s) and/or other MTC Device(s). Local communication between MTC Device(s) and other entities is out of scope of this technical specification.
MTC Feature: MTC Features are network functions to optimise the network for use by M2M applications.
example: text used to clarify abstract rules by applying them literally. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 3.2 Symbols | Void. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 3.3 Abbreviations | For the purposes of the present document, the abbreviations given in TR 21.905 [1] and the following apply. An abbreviation defined in the present document takes precedence over the definition of the same abbreviation, if any, in TR 21.905 [1].)
CCCH Common Control Channel
GERANIMTC GERAN Improvements for Machine Type Communications
IP Internet Protocol
KPI Key Performance Indicator
MS Mobile Station
MTC Machine Type Communications
PDCH Packet Data Channel |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 4 Areas for study to effectively support MTC in GERAN | |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 4.1 General | Sub-clause 4 contains the outcome of the study of GERAN enhancements driven by the prioritized general MTC functions as defined in [3] that are considered applicable to GERAN specifications. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 4.2 Overload control | |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 4.2.1 General | Overload Control refers to use cases Radio Network Congestion, Signalling Network and Core Network Congestion as described in [2] Annex A. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 4.2.2 Description and Analysis | |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 4.2.2.1 CCCH Overload Control | |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 4.2.2.1.1 Description and Analysis | The large amount of access attempts that can be generated from mobile stations used for MTC is believed to increase the load and cause congestion on the common control channel (CCCH) and therefore may negatively impact legacy services.
The legacy pre-release 10 RR connection establishment procedure is not sufficient for the network to avoid CCCH congestion that can be caused by mobile stations used for MTC. However the implicit reject procedure specified in release 10 in 3GPP TS 44.018 can effectively protect the legacy services from CCCH congestion that can be caused by mobile stations configured with low access priority. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 4.2.2.1.2 Result | By using the implicit reject procedure, the network can effectively protect the CCCH from being overloaded by mobile stations configured with low access priority.
The objective of CCCH overload control in MTC study has been met with the implicit reject procedure with respect to preventing overload of CCCH hence minimising impact to legacy services from devices configured for low access priority. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 5 Void | |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6 Common assumptions | |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.1 Traffic model | |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.1.1 General | The traffic model is assumed to be mobile originated, meaning that the MTC server will not poll/request reports from the MTC devices. Hence, the MTC devices will require access to the network rather autonomously and thus the network need not page the MTC devices.
6.1.2 CCCH Signalling
In order to capture different network access behaviors the investigated scenarios are divided in both synchronized and non-synchronized access.
Three different traffic models are used as listed in Table 1.
Table 1: Traffic models.
Traffic model
Description
T1
MTC devices accessing the network in an uncoordinated/non-synchronized manner
T2
MTC devices accessing the network in a coordinated/synchronized manner with a certain distribution
T3
Legacy devices accessing the network in an uncoordinated/non-synchronized manner
Table 2: CCCH Traffic Scenarios
Scenario
T1
T2
T3
Number of devices
λ / (Reporting interval) (see note 1)
X
λ / (Reporting interval) (see note 1)
Arrival process
Poisson
Arrival intensity: λ [arrivals/second]
Time limited deterministic event distribution. See subclause 6.1.2.1.
The time-spread of the distribution is controlled by parameter T [s], which should include T=1.
Poisson
Arrival intensity: λ [arrivals/second]
Case 1: λ = 5 for CS traffic
Case 2: λ = 5 for CS traffic and
λ = 15 for PS traffic
Reporting interval
• 5 seconds
• 15 minutes
• 1 hour
• 1 day
see note 2
N/A
Report Sizes
• 10 byte
• 200 byte
• 1000 byte
• 10 byte
• 200 byte
• 1000 byte
N/A
NOTE 1: This assumption is roughly true as long as the data session duration is shorter than the reporting interval.
NOTE 2: With this traffic model reporting interval is not defined since the number of devices are fixed and the access need to be finished by all devices before the following access can take place.
Scenario T1 can be considered to be quite realistic, since for a large amount of users the overall arrival process can be modelled as a Poisson arrival process regardless of the individual arrival process.
Scenario T2 models the behavior when e.g. multitude of ill-configured power meters are set to deliver their measurements at the same time or when the meters starts reporting after e.g. a power outage. The MTC devices are here assumed to be synchronized within an interval of T seconds.
Scenario T3 models the behavior of CS and PS legacy devices where the overall arrival processes (separate for CS and PS) can each be modelled as a Poisson arrival process as the devices are assumed to be initiated independently of each other. Scenario T3 should be regarded as the reference case when evaluating impacts on legacy mobiles and the ASR for CS services simulated in Scenario T3 should be over 98%.
The overall objective of the T3 scenario is to be used in conjunction with either the T1 or T2 scenario, respectively, to evaluate the impact of the MTC traffic on the legacy traffic.
In the simulations, the network should not use pre-emptive retransmissions of messages on the CCCH/D. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.1.2.1 Time limited deterministic event distribution | Following considerations are made:
Assuming that all events take place between t=0 and t=T , the intensity is described by the distribution p(t) and the total number of devices in the cell is X, then the number of arrivals in the i:th TDMA frame is given by:
Equation 1: Number of arrivals in a given TDMA-frame
Any distribution should preserve the total number of access attempts when time duration T is changed, and should be limited in time:
.
The distribution used in this feasibility study is the so called Beta distribution, please see clause 6.1.2.1.1. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.1.2.1.1 Beta distribution | The benefit of this model is:
• This deterministic traffic model simplifies simulation (by virtue of being deterministic). It may be considered to approximate the traffic load generated by multiple devices accessing the network quasi-simultaneously (the selection of a time window of 1 second is arbitrary).
, where is the Beta function.
Figure 1: The Beta distribution with α=3, β=4 when T=1
The values of α=3 and β=4 for traffic model T2 are used, which gives the PDF that is depicted in Figure 1 above for the case when T=1.
6.1.3 Traffic model on PDCH
It is assumed that traces from CCCH Signalling simulations as defined in clause 6.1.2 are used to model the traffic for the PDCH simulations. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2 Methodology | |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.1 Simulator methodology | A single cell evaluation of possible congestion of the CCCH and PDCH is used.
Either a single cell simulator (sometimes also referred to as a protocol level simulator) or system level type simulator can be used where the basic difference is in that the system level simulator models dynamic interference from neighbouring cells while the single cell simulator uses network traces (see clause 6.2.1.1) to generate external interference.
Irrespective of simulator level used network traces, as described in clause 6.2.1.1, should be presented for easier comparison of results from different companies. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.1.1 Network trace | In order to get a simplified distribution of the interfering signal that network level simulations are run to collect the signal distributions of the interferer.
Further on, the derived interference distribution is presented in tabulated format to allow for easier comparison and verification of contributions from different companies.
Note that the collection of signal interferers might be different depending on the traffic scenario investigated, i.e. CCCH or PDCH congestion. E.g. the CCCH distributions will be based on Iext, as defined in clause 6.2.2.6.1, while the distributions used for the PDCH evaluation is left vendor specific, see clause 6.2.2.6.2.
An example of a tabulated distribution of external interference for the RACH simulations is given in Figure 2.
Iext
Signal level [dBm]
CDF value
-110
0
-109
0.02
-108
0.03
-107
0.05
.
.
.
.
.
.
.
.
-29
0.98
-30
1
Figure 2: Example of RACH interferer distribution |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.1.2 Network load | The resource allocation from the background traffic in neighbouring cells is assumed to be fully allocated (constant transmission), transmitted at full power (no power control) using 8PSK modulation (for assumptions on power back-off see Table 3). |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.1.3 Cell under investigation | For the cell under investigation all traffic is assumed to be MTC devices while the background noise is assumed to be best effort PS traffic modelled as described in clause 6.1. This should be seen as a worst case scenario in terms of network access attempts.
6.2.1.4 Service coverage
Full service coverage of stationary MTC devices should be assumed, i.e. no service outage is accepted. This is ensured by allowing only minimum signal levels of -104 dBm – 3 for each MTC device, where an additional gain of 3 dB is assumed for a dual antenna MRC type BTS architecture. This would guarantee GMSK coverage. The minimum signal level should include fast fading, since TU0 is used (see clause 6.2.2.3). |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.2 Simulation assumptions | |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.2.1 General | This clause defines the parameters required for the simulations which may be required to conduct the study. The parameters are referenced where appropriate.
Table 3: Network level simulator parameters
Parameter
Value
Unit
Comment
Sectors per site
3
Sector antenna pattern
65º deg H-plane,
max TX gain 15
dBi
18 dBi antennas in 900‑band are large and not considered to be common in urban areas.
Path loss model
Per 30.03,
Hb = 5 m,
dB
In urban areas, 5 m over average roof height is considered more typical than the default value of 15 m in 30.03.
Minimum coupling loss
64
dB
1800: TR 25.942 2 GHz.
900: assumed 6 dB lower
Building penetration loss
15 / 20
dB
Indoor 1 / Indoor 2
Indoor 1/Outdoor devices
90 / 10
%
Scenario 1
Indoor 1/Indoor 2 devices
50 / 50
%
Scenario 2
Interference model
Neighbouring cells BCCH
The neighbouring cells according to the BCCH frequency reuse pattern are modelled as if they have full traffic.
Log-normal fading
Standard deviation
8
dB
Correlation distance
110
m
See note
Channel propagation
See table 6
Output power
- MS
- BTS
33
43
dBm
Excluding backoff
Backoff
- MS
6
dB
- BTS
4
dB
8PSK modulation assumed.
Noise figure
- MS
10
dB
- BTS
8
dB
Inter-site log-normal correlation coefficient
0
Low correlation in urban scenarios.
NOTE: For the cell under investigation it is not essential to model stationary devices (TU0), thus a correlation distance of 0 m can be used. Fewer cell realizations needed if this value is zero.
Table 4: Network scenario
Parameter
Value
Unit
Comment
Frequency band
900
MHz
Cell radius
500
m
Bandwidth
2.4
MHz
Number of channels
12
BCCH frequency reuse
4/12
BCCH or TCH under interest
BCCH
Table 5: Protocol level parameters
Parameter
Value
Comment
CCCH assumptions
- Tx-integer
- S
- Max. retrans (M)
- T3142
- T3146
20
109
4
5 sec.
(Tx+2S)/217=1.1 sec.
These default values should be included among those evalutated.
See 3GPP TS 44.018 for implementation details
BCCH configuration
Non-combined
# PDCHs
4
Number of PDCHS availabale data traffic
# AGCHs per 51-multiframe
6
PDCH Resource Assignment
1 TS UL + 1 TS DL (BTTI)
Link adaptation
Enabled
Service type
1. EGPRS
2. GPRS
RLC mode of operation
Acknowledged Mode (AM)
Table 6: Link specific settings.
Parameter
Value
Comment
Channel profile [MTC]
TU3
TU0
For PS users to derive network level trace on UL
1. For MTC devices in protocol level simulations
2. For PS users to derive network level trace on DL
Receiver type UL
MRC
Incremental redundancy
Enabled (only for EGPRS)
See Table 5 |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.2.2 Path loss | It is assumed that the gain (path loss + shadow fading + antenna gain) from a given MS to its serving BTS is the same in UL and DL. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.2.3 Channel propagation | It is assumed that the external interferers experience a TU3-channel while the MTC devices are assumed to be stationary and subject to TU0-channel propagation. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.2.4 External interference | It is assumed that the external interference levels are uncorrelated between the DL and UL, i.e. that uncorrelated samples are used from the respective distributions. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.2.5 Application protocol | It is assumed that the MTC application is using UDP as a transport protocol with acknowledgments on the application layer from the MTC server to the MTC client will be transmitted, i.e. there will both be PUANs and data blocks (containing application Acks) transmitted in the DL for the PDCH evaluation. Details are left FFS.
During a simulation session the application performs a single access attempt, i.e. there should be no re-attempts triggered by the application. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.2.5.1 IP version | The IP version to use for the evaluation is left FFS. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.2.6 Link model | |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.2.6.1 CCCH | A simplistic link-to-system interface is assumed.
It is assumed that only a total co-channel interference level needs to be assumed for each burst. Adjacent channel suppression is assumed to be 18 dB. To capture the correct combined channel propagation behavior of the total interfering signal, impacts on fast fading is proposed to be included in the signal distribution of the interferer. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.2.6.1.1 RACH (CCCH/U) | For possible reception of an access burst, CRACH/(IRACH + ITOT) needs to be greater than 9 – 3 = 6 dB. RACH reference interference ratio is specified at 9 dB (Channel propagation TU3, 3GPP TS 45.005) and an additional gain of 3 dB is assumed for a dual antenna MRC type BTS architecture.
On top of this an error rate of 15% is added (RACH reference interference performance, TU3, 45.005).
This should be seen as a worst case scenario since no errors could be expected above a certain CRACH/(IRACH + ITOT) threshold.
NOTE: The figures above are being investigated. It could be considered instead to leave this unspecified and document it when results are displayed. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.2.6.1.2 AGCH (CCCH/D) | For possible reception of an access grant, CAGCH/ITOT needs to be greater than 9 dB. AGCH reference interference ratio is specified at 9 dB (Channel propagation TU3, 3GPP TS 45.005).
On top of this an error rate of 22% is added (AGCH reference interference performance, TU3, TS 45.005).
NOTE: The figures above are being investigated. It could be considered instead to leave this unspecified and document it instead when results are displayed. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.2.6.2 PDCH | Vendor specific L2S mapping methodology is to be used that can be verified against a set of pre-defined interferer scenarios.
Common assumptions for the UL receiver include:
• Dual antenna base station
• MRC receiver algorithm.
Common assumptions for the DL receiver include:
• Single antenna mobile station |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.2.2.7 Number of CCCHs | The CCCH performance is evaluated using a single CCCH. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.3 Output | |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.3.1 General | All results should be presented as per indicated below. It should be noted that this list is not exhaustive and that outputs not currently listed cannot be precluded that could affect the conclusions of this work.
Upon evaluation of different proposals, the KPIs of services with a higher priority should be seen to take precedence over the KPIs of services with lower priority. The Access success rate for legacy CS services should be considered more crucial than the Access success rate of a MTC device configured for low-access-priority. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.3.2 Overall MTC simulation and evaluation output | MTC success rate = Number of successfully received reports (i.e. all application level payload associated with this report) sent from the device to the network divided by the total number of arrivals.
MTC delay = The time it takes for a MTC device to successfully transfer its application level payload, as from when it makes its first application initiated access [50/95/99 percentile].
MTC coverage outage = Percentage of MTC devices that are initially placed out of coverage. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.3.3 CCCH signalling output | Access success rate = Number of successful Immediate Assignment procedures, see sub-clause 3.3.1.1 in [4] divided by total number of Immediate Assignment procedures, inclusive of both RACH and AGCH.
Access attempts needed = Number of access attempts per successfully completed Immediate Assignment procedures, inclusive of both RACH and AGCH [histogram].
Access time = Time from when an Immediate Assignment procedure is initiated by higher layers until successful completion of the said Immediate Assignment procedure, inclusive of both RACH and AGCH [50/95 percentile].
CCCH Capacity Used = Percentage of CCCH capacity used. To be evaluated for both RACH and AGCH.
The impact on the legacy traffic should be evaluated for both T1 and the T2 scenario as described below.
For the T1 scenario in conjunction with the legacy traffic modelled by T3, the evaluation of the Access success rate for the legacy traffic should be conducted with a time-window starting at a period in time such that all initialization effects from different random access procedures are excluded.
Furthermore, when the T2 scenario is evaluated in conjunction with the legacy traffic modelled by T3, a windowed evaluation should be performed of the Access success rate, evaluating all legacy devices initiating their random access procedure within consecutive 10 second time-windows. The T2 peak traffic should be initiated when the traffic load modelled by T3 has reached a stable level.
Figure 3: Periodic evaluation of random access procedure
The statement above is clarified in Figure 3, where [i] denotes where device i initiates its Immediate Assignment procedure and the dashed line for how long period the current Immediate Assignment procedure is active. The access success rate for the first period (0 – 10 s) should be calculated for users 1, 2 and 3, even though the end of the Immediate Assignment procedure for the user is in the subsequent evaluation period. The access success rate for the second period (10 – 20 s) should be calculated for users 4, 5, 6 and 7, and the access success rate for the third period (20 – 30 s) should be calculated for users 8, 9 and 10.
Upon the windowed evaluation of the Access success rate an overall measure of the access success rate should be provided. This measure should use a time-window large enough to cover all effects from the MTC devices accesses.
The Average access success rate for the legacy CS services when MTC traffic is added should not be significantly decreased as compared to the reference case of the T3 scenario (see sub-clause 6.1.2). The Average access success rate of legacy PS services may have some relaxation. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 6.3.4 PDCH traffic output | TBF Blocking Rate = Blocking rate due to insufficient resources (e.g. USF and TFI identifiers), which makes it impossible for the network to assign uplink PDCHs to the MTC devices. The output should be differentiated between different causes.
MTC payload transfer delay = The time it takes for a MTC device to successfully transfer its application level payload, as from when it received its TBF assignment [50/95/99 percentile]. |
686035b507b3b9278dc8a7741151e4bb | 43.868 | 7 Summary and conclusions | The impacts on GERAN specifications identified in sub-clause 4 should be used as a basis for additional normative specification work.
Annex A:
Change history
Change history
Date
TSG #
TSG Doc.
CR
Rev
Subject/Comment
Old
New
2012-11
56
GP-121309
Approved at TSG GERAN#56
2.0.0
12.0.0
2014-02
61
GP-140199
0001
1
Correction to Cross References
12.0.0
12.1.0 |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 1 Scope | The present document contains the result of the study on the impacts on signalling between the UE and core network when energy saving measures are applied to network entities.
The study aims, within the defined CT1 work areas, at:
- analysing UE idle mode procedures and signalling between the UE and core network resulting from switch on/off of radio equipment in all types of 3GPP accesses, including home cell deployment and I-WLAN, as well as power adaptation of radio equipment (where applicable);
- performing a corresponding analysis for connected mode UEs;
- analysing similar impacts from activation status of non-3GPP access networks;
- documenting limitations, weaknesses and inefficiencies in these procedures, with emphasis on mass effects in the signalling between the UE and core network; and
- studying potential optimizations and enhancements to these procedures.
The study also evaluates potential enhancements to 3GPP specifications under CT1 responsibility.
This study takes into account decisions made by other 3GPP working groups in their related work. |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 2 References | The following documents contain provisions which, through reference in this text, constitute provisions of the present document.
- References are either specific (identified by date of publication, edition number, version number, etc.) or non‑specific.
- For a specific reference, subsequent revisions do not apply.
- For a non-specific reference, the latest version applies. In the case of a reference to a 3GPP document (including a GSM document), a non-specific reference implicitly refers to the latest version of that document in the same Release as the present document.
[1] 3GPP TR 21.905: "Vocabulary for 3GPP Specifications".
[2] 3GPP TR 32.826: "Study on Energy Savings Management (ESM)".
[3] 3GPP TS 23.234: "3GPP system to Wireless Local Area Network (WLAN) interworking; System description".
[4] 3GPP TS 24.234: "3GPP System to Wireless Local Area Network (WLAN) interworking; WLAN User Equipment (WLAN UE) to network protocols".
[5] 3GPP TS 33.234: "3G security; Wireless Local Area Network (WLAN) interworking security".
[6] 3GPP TS 32.551: "Energy Saving Management (ESM); Concepts and requirements".
[7] 3GPP TS 24.302: "Access to the 3GPP Evolved Packet Core (EPC) via non-3GPP access networks".
[8] 3GPP TR 23.888: "System Improvements for Machine-Type Communications".
[9] 3GPP TS 29.118: "Mobility Management Entity (MME) – Visitor Location Register (VLR) SGs interface specification".
[10] 3GPP TS 24.167: "3GPP IMS Management Object (MO); Stage 3".
[11] 3GPP TS 24.216: "Communication Continuity Management Object".
[12] 3GPP TS 24.301: "Non-Access-Stratum (NAS) protocol for Evolved Packet System (EPS)".
[13] 3GPP TS 24.237: "IP Multimedia Subsystem (IMS) Service Continuity; Stage 3".
[14] 3GPP TS 23.251: "Network Sharing; Architecture and functional description".
[15] 3GPP TS 23.401: "GPRS enhancements for E-UTRAN access".
[16] 3GPP TS 23.402: "Architecture enhancements for non-3GPP accesses".
[17] 3GPP TS 23.261: "IP flow mobility and seamless Wireless Local Area Network (WLAN) offload; Stage 2".
[18] 3GPP TS 24.303: "Mobility management based on Dual-Stack Mobile IPv6; Stage 3".
[19] 3GPP TR 36.927: "Evolved Universal Terrestrial Radio Access (E-UTRA); Potential solutions for energy saving for E-UTRAN".
[20] 3GPP TS 24.368: "Non-Access Stratum (NAS) configuration Management Object (MO)".
[21] 3GPP TS 24.008: "Mobile radio interface Layer 3 specification; Core network protocols; Stage 3". |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 3 Definitions, symbols and abbreviations | |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 3.1 Definitions | For the purposes of the present document, the terms and definitions given in 3GPP TR 21.905 [1] and the following apply. A term defined in the present document takes precedence over the definition of the same term, if any, in 3GPP TR 21.905 [1].
Macro cell: generic term used for all cell types under operator’s control (in contrast to "home cells" or "femto cells"); in this sense it includes also so-called "micro" and "pico" cells (used in the context of hierarchical cell structures). |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 3.2 Abbreviations | For the purposes of the present document, the abbreviations given in 3GPP TR 21.905 [1] and the following apply. An abbreviation defined in the present document takes precedence over the definition of the same abbreviation, if any, in 3GPP TR 21.905 [1]. |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 4 Overview | One main energy saving mechanism in the context of this Technical Report is realized by switch-off of radio equipment on the network side. As a consequence, a UE currently being served by the radio equipment subject to switch-off will have to find an alternative, either in the same RAT (if possible by coverage, this would naturally be preferred), or in another RAT. We call these cases "intra-RAT energy saving" and "inter-RAT energy saving", respectively. The fundamental assumption is that, due to overlapping radio coverages, in all practically relevant cases an alternative radio access can indeed be found.
In figure 4-1 two base coverage scenarios allowing energy saving by switch-off are illustrated (only one representative UE shown).
On the left hand side an overlay structure of one and the same RAT is shown; apart from capacity considerations every UE in coverage area 2 can also be served by the radio equipment in coverage area 1; thus radio equipment for coverage area 2 may be switched off (during suitable times) for energy saving purposes. This is an intra-RAT energy saving case. A variant for the non-overlaid intra-RAT configuration is described in subclause 5.1.3 and subclause 5.5.
On the right hand side, in contrast, coverage area 4 is not fully overlaid by coverage area 3, and they belong to different RATs. Consequently, radio equipment for coverage area 4 could only be switched off if still other coverage areas exist (not shown, but indicated by dots), or all UEs are located in the overlapping area. This case realizes inter-RAT energy saving.
More details on such energy saving scenarios are found in 3GPP TR 32.826 [2] and 3GPP TR 36.927 [19].
Figure 4-1: Two base coverage scenarios for energy saving by switch-off of radio equipment
In table 4-1 we use the terms "source radio" and "target radio" with the intended meaning that the considered UE transitions from on to the other radio access due to switch-off; we list all combinations in a matrix, for a UE attached for PS services. Regarding the UE behaviour following a loss of (source) radio access, depending on the availability of RATs as target radio access, we find either fully standardized and optimized behaviour within 3GPP RATs (GERAN, UTRAN, E-UTRAN), or non- or only partially standardized and optimized behaviour (if either source or target radio access is not a 3GPP access).
Table 4-1: Combinations of source and target radio for energy saving (for UE attached for PS services) and possible procedures after switch-off of source radio
The main diagonal of this matrix describes intra-RAT energy saving, and the non-diagonal elements constitute inter-RAT energy saving. A characteristic difference between the two directions of switching, i.e. between switching off and switching on an access network, is that with the latter the UE in case of non-optimized handover is not immediately forced to act, because it ends up in having more choices for access networks than before.
CS only mode encompasses solely GERAN and UTRAN radio access; however, RAN3 has defined enhancements for intra-RAT energy saving only for E-UTRAN. Aspects of CS/PS mode of operation are discussed in subclauses 5.1.8 and to some extent in subclause 5.1.6. Aspects of domain selection and relationship to applications are discussed in subclause 5.1.6. |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5 Analysis of signalling procedures between the UE and core network for energy saving scenarios | |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.1 Switch-off/on of 3GPP macro cells | |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.1.1 General description | The following main characteristics of 3GPP RATs GERAN, UTRAN and E-UTRAN need to be taken into account in the study of energy saving:
- clear definition of idle and active mode;
- definition of registration areas (LAs, RAs, TAs);
- fully optimized handovers; and
- fully (operator) planned deployment.
General descriptions and more technical details are found in other sources (e.g. 3GPP TR 32.826 [2] and specifically for E-UTRAN in 3GPP TR 36.927 [19]). Here the goal is to complete the enumeration of cases and also consider more the aspects relevant to this study (i.e. signalling between the UE and core network). |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.1.2 Overlaid intra-RAT | A capacity enhancing cell, subject to switch-off for energy saving, is placed into the (full) coverage realized by other cells (see figure 5.1.2-1). Such a cell can be either totally within the range of one cell or it can intersect with multiple cells of the base coverage. The relative position of a capacity-enhancing cell, regarding registration area boundaries, determines whether signalling between the UE and core network is to be expected as a result of switching the cell off/on.
NOTE: In 3GPP TR 36.927 [19] this is called "inter-eNB scenario 1"
Figure 5.1.2-1: Coverage and registration areas for overlaid intra-RAT energy saving with 3GPP macro cells |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.1.3 Non-overlaid intra-RAT | The cell that is subject to switch-off has no alternative coverage, thus some cell adaption has to take place, in order not to disrupt service for the UEs due to a coverage "hole" of this RAT (and e.g. with no capability for an alternative RAT). As shown in figure 5.1.3-1, registration area boundaries will be distorted with switch-off.
NOTE: In 3GPP TR 36.927 [19] this is called "inter-eNB scenario 2"
Figure 5.1.3-1: Coverage and registration areas for non-overlaid intra-RAT energy saving with 3GPP macro cells |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.1.4 Inter-RAT | This case is illustrated in figure 5.1.4-1.
Figure 5.1.4-1: Coverage and registration areas for inter-RAT energy saving with 3GPP macro cells |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.1.5 Key issue: Registration signalling resulting from switch-off/on of macro cells | |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.1.5.1 Overlaid intra-RAT case (E-UTRAN) | The situation is shown in figure 5.1.5-1 derived from the use case given in the annex B of 3GPP TS 32.551 [6] (it can be similarly applied to the other 3GPP RATs). Only idle mode UEs are considered here, as UEs in connected mode can be handled by the network via timely handovers to cells remaining permanently active.
Cell 5 is used for providing the peak capacity and is overlaid on cells 1 to 4 (it is called here a "capacity enhancing cell"). Incidentally, also a tracking area (TA) boundary is passing through, and cell 5 is (arbitrarily) allocated to TA1.
NOTE 1: The assumption that a capacity enhancing cell is never located on the border of a TA cannot be made, as it would be too restrictive.
Figure 5.1.5-1: Capacity enhancing cell being switched off (overlaid intra-RAT case)
When switching off cell 5, all UEs camping on it have to re-select other cells. For those within coverage of cells 1 and 2 the cell re-selection affects only the UE internal state. In contrast, UEs within coverage of cells 3 and 4, after selecting these for camping, the corresponding UEs have to perform a TAU. In total, depending on the number of cells being switched off (nearly) at the same time and the actual TA configuration, there is a danger of overload on MME(s).
The situation after switch-on of a previously switched off capacity enhancing cell is shown in figure 5.1.5-2. The re-selection of cells occurs only gradually, due to the fact that some threshold in the difference of radio levels must be measured.
Figure 5.1.5-2: Example of a capacity enhancing cell being switched on again
A problem or at least an inefficiency can occur, if many UEs in idle mode start with their transmission of data (by issuing service requests) still in the source cells; this could have the effect that many handovers (into the target cell) are necessary. Depending on the TA configuration, additionally TAUs could be necessary (before the active mode is entered). The scenario is likely e.g. in the morning, when formerly sleeping users are waking up and start checking their mails, browsing internet in a synchronized manner, e.g. just before or after their favourite morning news broadcast. It would be advantageous if idle mode UEs re-select target cells more quickly after they have been switched on, before they become active.
NOTE 2: A switch-off can also be performed by a decrease of power over time. This case is described in subclause 5.5 "Power adaptation for 3GPP macro cells". |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.1.5.2 Inter-RAT case (E-UTRAN target for switch-off) | For this subclause it is assumed that GERAN/UTRAN provides the overall coverage and is used as fallback after E-UTRAN has been switched off.
At this time a similar issue as described in subclause 5.1.5.1 arises more massively (independent of a cell location on a registration area boundary), because all idle mode UEs with ISR not activated have to perform RAU signalling, after selection of GERAN/UTRAN.
UEs in idle mode with ISR activated benefit from the double registration. After selecting GERAN/UTRAN no immediate registration signalling is necessary. Their behavior regarding ISR deactivation timer is specified in 3GPP TS 24.301 [12] (subclause 5.1.3) and shown in figure 5.1.5.2-1.
Figure 5.1.5.2-1: Behaviour of UEs with ISR activated when E-UTRAN is switched off
1) E-UTRAN is switched off for energy saving. UEs will subsequently fall into state EMM-REGISTERED.NO-CELL-AVAILABLE.
2) Affected UEs (those in the coverage area and currently registered with E-UTRAN), with ISR deactivated, have to perform RAU signalling, after selecting GERAN/UTRAN. This can lead to a massive peak in signalling, depending on synchronization of switch-off of eNBs (corresponding to the number of UEs affected).
3) Affected UEs with ISR activated have their periodic TAU timer running (unsynchronized, and in general with varying, individual durations). Only those are considered in the subsequent steps.
4) After expiry of the periodic TAU timer, affected UEs cannot perform the periodic TAU and start their individual timer for deactivation of ISR (T3423).
5) T3423 timers expire and UEs deactivate ISR.
6) E-UTRAN is switched on again.
7) UEs realize that E-UTRAN is again available and can re-select it. At this point they need to initiate TAU, which potentially leads to the re-activation of ISR. |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.1.6 Key issue: Cross layer aspects | In the current 3GPP system various control features exist which govern the selection of domains for delivery of services (e.g. voice, SMS). As an example, it may be an operator's preference to handle voice communication via IMS/PS only; if no HSPA is rolled out then effectively only E-UTRAN would remain for voice. This constitutes a cross layer coupling (IMS voice layer effectively requires a particular radio access), and there are related administrative settings (per UE) in NAS related specifications. The situation is illustrated for one UE in figure 5.1.6-1.
Figure 5.1.6-1: Cross layer dependency and relationship to inter-RAT energy saving options
The effect of switching off/on radio equipment for inter-RAT energy saving potentially counteracts above described domain selection principles. A hard setting for only one domain (i.e. no other domain admissible) could mean that inter-RAT energy saving is effectively not possible for certain cells, if and as long as such UE(s) need to be accommodated (under the assumption that service needs to be guaranteed for them).
Editor's note: It is FFS how the alignment between administrative settings per UE and inter-RAT energy saving procedures is achieved/maintained, and on which scale of dynamicity.
NOTE: A similar issue arises for UEs with support of only one particular RAT (e.g. LTE only UEs). |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.1.7 Key issue: Idle mode UEs with emergency bearer services | UEs in idle mode with emergency bearer services possibly await a callback from e.g. an emergency answering point or need to establish an outgoing emergency call; both can be deemed critical and in some cases life-saving.
The switch-off of cells for energy saving could result in a UE becoming temporarily unavailable for a callback, e.g. until the UE has successfully completed a location registration if the switched off cell is located near the boundary of a location registration area. (Of course this can also occur if the UE itself moves into a new location registrations area.)
Editor's note: it is FFS whether this needs to be considered, e.g. by exempting all cells potentially serving idle mode UEs with emergency bearer services from switch-off for energy saving.
Editor's note: it is FFS how the signalling between UE and core network can be optimized with respect to this issue. |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.1.8 Key issue: Impact of inter-RAT energy saving on signalling over SGs | For a UE configured to use CS fallback, according to figure 4.2.2.1 in 3GPP TS 29.118 [9] the SGs state has to be moved to SGs-NULL if a Location Update Request is received at VLR via A, Iu or Gs interface. That is the case some time after E-UTRAN radio access has been turned off, as the UE has no means to perform combined tracking area updates.
In the opposite direction, after E-UTRAN is switched on again, the SGs associations will be re-established by registration signalling of all affected UEs. There are two parts: EPS-NAS signalling and SGs signalling.
If any synchronization effect and consequently increased load for EPS-NAS signalling is found relevant (this is described as a key issue in subclause 5.1.5.2), it also applies over SGs for UEs configured to use CS fallback. |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.2 Switch-off/on of 3GPP home cells | |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.2.1 General description | Switch-off of radio equipment in the home cell deployment scenario seems to be an attractive target case for energy saving, due to the expected mass of equipment (seen in total, i.e. for the whole population). Key characteristics of this scenario are:
- in many cases the overlay by the macro network exists (except in the case where home cells are deployed for reasons of coverage extension);
- the users are aware of (or even configure) the switch-off/on times; and
- the number of UEs potentially affected by the switch-off of one single home cell is very small, compared to a macro cell.
For the enterprise scenario, the second and third bullets may not apply. |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.2.2 Key issue: Incidental synchronization of H(e)NB switch-off/switch-on | For marketing reasons and potentially also due to stronger legal requirements for energy saving, the support of the simplest energy saving feature, namely switch-off during periods of expected inactivity (e.g. according to a daily or weekly schedule), may become a standard feature. In this case configuration settings for the switch-off and switch-on times need to be provided. Potentially there could be sources of incidental, unwanted synchronization, e.g.:
- factory pre-settings for the energy saving mode, e.g. 24:00 for switch-off and 06:00 for switch-on, which may not be changed by a majority of users;
- even if configuration settings are changed by users, there may be habits to set them to almost identical times. As an example, many people know that they go to sleep immediately after their favourite TV broadcast and may set the switch-off time accordingly; or
- another problematic case would be if an implementation allowed configuration settings to be selected only from a few predefined times (e.g. only half-hourly).
In all the above cases mass events regarding re-selection of cells and, depending on the structure of tracking areas, NAS signalling may result. |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.3 Switch-off/on of WLAN access networks in 3GPP I-WLAN | |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.3.1 General description | Switching off a WLAN access network has the consequence that it will no longer be detected in active or passive scanning (see subclause 4.3.1 in 3GPP TS 24.234 [4]), therefore it would drop out as a candidate for 3GPP I-WLAN initial procedures. On the other hand, if switch-off for energy saving is foreseen at all, detecting an active SSID in scanning does not guarantee that it will remain in operation. If a SSID is deactivated while the UE is connected to it, this produces for the UE essentially the same effect as if the UE lost coverage by movement; however, in the network the difference is that WLAN actions, e.g. the WLAN initiated disconnection procedure as defined in subclause 7.5 of 3GPP TS 23.234 [3], can no longer be performed. If I-WLAN coverage for the UE remains (i.e. it is an intra-RAT energy saving case), according to 3GPP TS 24.234 [4] both the WLAN UE 3GPP I-WLAN re-selection procedure and the WLAN UE PLMN re-selection procedures have to be performed.
For access network selection with 3GPP I-WLAN some configuration data in the UE are foreseen (see subclause 5.1 of 3GPP TS 24.234 [4]): "User Controlled WLAN Specific Identifier list" and "Operator Controlled WLAN Specific Identifier list", both including a priority for selection. These lists have currently a static character.
NOTE: the analysis of activation status of WLAN access networks inter-connected to EPC via ePDG is covered in subclause 5.4. |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.3.2 Key issue: Latency associated with WLAN access network re-selection after switch-off (WLAN 3GPP IP access) | A compact view of what happens (for one UE) when a WLAN access network is switched-off is illustrated in figure 5.3.2-1 (non-roaming case; for simplicity WAG is not shown, and it is assumed that the same PDG can be reached).
Figure 5.3.2-1: Compact view of procedures after switch-off of WLAN access network in 3GPP I-WLAN
Taking into account the number of signalling steps (e.g. counted from figure 4 and figure 7A in 3GPP TS 33.234 [5]) it becomes apparent that current procedures are not optimal for switch-off of a WLAN access network; the latency associated with the dashed box in figure 5.3.2-1 likely degrades the user experience. |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.3.3 Key issue: Mass effect resulting from WLAN access network re-selection after switch-off | WLANs can be considered a means to enhance capacity for mobile users in hotspots. If their switch-off is foreseen, the resulting mass effect has to be analysed; this can also be derived from figure 5.3.2-1, differentiating between procedures local to every UE and procedural parts running on the same network elements for the affected mass of UEs.
As an example case, the closing of a WLAN hot spot can be imagined: I-WLAN connectivity for e.g. ~100 users was provided via a WLAN access network 1, it is turned off and users have to be accomodated by alternative WLAN(s).
Analysing figure 5.3.2-1 regarding mass events in the network, the following is noticed:
Procedure A: local action per UE, therefore no mass event;
Procedure B: mass event on PDG, 3GPP AAA server / HSS;
Procedure C: mass event in WLAN access network 2;
Procedure D: mass event in WLAN access network 2 and 3GPP AAA server / HSS; and
Procedure E: mass event in WLAN access network 2, 3GPP AAA server / HSS and PDG.
NOTE: for an estimation of the scale of the mass effect it has to be taken into account that procedure D is optional and procedure E is only performed for WLAN 3GPP IP access.
Every mass event effectively consists of a peak in signalling. Further observations and assumptions can be made:
- (over)load in one network entity may lead to the broadening of the signalling peak in other nodes due to delay and queing; e.g. for procedure D, (over)load in the WLAN access network 2 leads to broadening of the signalling peak in 3GPP AAA server / HSS and PDG;
- load resulting from procedure B and procedure D adds up in 3GPP server / HSS, load resulting from procedure B and procedure E adds up in PDG;
- if there is a multiplicity of alternative WLANs or PDGs, the load problem is reduced by this factor for them. But as a consequence, the above mentioned broadening of peak signalling may be reduced, or may not happen at all, for the other nodes; and
- even if all load can be accomodated easily by the WLAN access network(s), there is one danger: if e.g. due to identical closing times of hotspot venues the switch-off of many WLAN access networks happens at approximately the same time, this is a drastic stress scenario for the backend 3GPP I-WLAN related infrastructure (3GPP AAA server / HSS and PDG). |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.4 Switch-off/on of non-3GPP access networks | |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.4.1 General description | The switch-off of a non-3GPP radio access can occur on a varying scale, from a single access point to a whole metro area network or a country-wide set of hotspots. As indicated in table 4.1, with current 3GPP specifications it can be handled as a (non-optimized) handover on IP layer. Handover procedures in 3GPP TS 23.402 [16] do not specify details of the trigger for handover initiation, but discovery of the availability of a target radio access is the prerequisite. Two cases exist:
1. If the non-availability of a non-3GPP radio access (due to switch-off) can be known sufficiently well in advance, then this can be used as a trigger and timely handovers are possible.
2. Another case is constituted when the UE is not aware of the imminent switch-off and just looses radio coverage of its currently selected RAT and access network. With current 3GPP specifications the UE behaviour is left implementation specific (only network selection and re-selection upon switch-on of the UE and recovery from lack of coverage, but not upon detection of lack of coverage is specified). This affects only the initial step of the handover procedure (trigger function, containing access network selection), the rest remains unchanged. Still, the likely result is added latency and thus a less seamless handover. |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.4.2 Key issue: Degradation of service due to switch-off of non-3GPP access networks | NOTE: the terms "planned" / "unplanned" handover are not formally defined in 3GPP and used here for the sake of concise description.
Non-3GPP accesses have been integrated into EPS with the aim to provide service continuity across accesses (currently by virtue of non-optimized, IP based handovers) and as seamless as possible. In figure 5.4.2-1 the general sequence of such a handover from non-3GPP access to E-UTRAN access is sketched, as deduced from clause 8 of 3GPP TS 23.402 [16]. The assumption of a "planned" handover is taken, i.e. that the procedure for establishing connectivity via the 3GPP access can essentially be run in parallel to the established connection via the non-3GPP access.
Figure 5.4.2-1: Compact view of procedures in (planned) handover from non-3GPP access to 3GPP access
When a switch-off of a non-3GPP access occurs, then without any further means for mitigation, the handover procedure can only be initiated after the UE has detected the loss of the non-3GPP access and becomes therefore unplanned.
Figure 5.4.2-2: Compact view of procedures in (unplanned) handover from non-3GPP to 3GPP access
As a consequence, the issue of service degradation will always occur, even for UEs which would support the planned handover. The gap in connectivity may become even larger in case of multiple PDN connectivity. |
2d016e763d82d2bbce817fafc33663a6 | 24.826 | 5.5 Power adaptation for 3GPP macro cells |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.