VoLTE: Tracking Area Update and combined attach

A telecommunication network provides the way of identifying and tracking the location of the UE to maintain the UE mobility. The UE registers its location first time when it attaches the LTE network. Once the UE location is registered in the network, the UE may either periodically or per event initiate the tracking area update procedure. In case the tracking area update procedure has failed, the UE retries the same procedure based on the retry scheme. If the UE loses the connection and/or falls into the idle state, the UE needs to re-register its location as soon as it comes back to connected state. Without location report, the network may assume that the UE stays at the last visited tracking area and if not, the network will have a strategy to get to know the UE location based on the paging scheme.

NOTE Refer to the section 5.3.3 "Tracking Area Update procedures" of 3GPP TS23.401 for the detailed cases where the UE triggers the Tracking Area Update (TAU) procedure.

I. Identifying UE location 

The location of the UE is identified by the combination of several identifiers and it is subject to the configuration of the physical and logical network topology. The following shows the conceptual diagram of the radio network.

Fig 1. UE location and Tracking Area

• Cell-ID : locally unique identifier of a cell with the eNB
• eNB-ID : locally unique identifier of an eNB within the PLMN
• E-UTRAN Cell ID(ECI): locally unique identifier of a cell within a PLMN [eNB ID + Cell ID]
• E-UTRAN CGI (ECGI): globally unique identifier of a cell [PLMN ID + ECI]
• Tracking Area Code (TAC): locally unique identifier of the tracking area within a PLMN
• Tracking Area Id (TAI): globally unique identifier of a tracking area [PLMN ID + TAC]
• PLMN ID: globally unique identifier of a PLMN [MCC + MNC]

II. Tracking Area Update procedure and combined attach

The tracking area update procedure is initiated by the UE to register its own location to the network and it can occur when the UE is either in an idle state (i.e., ECM-IDLE) or in an active state (i.e., ECM-CONNECT bearer before release it sends TAU request. The UE uses the LAI and the ECGI to represent its own location. The UE may request the combined attach for both EPS services and non-EPS services. In this case, the EPS will need to interwork with the legacy network. The failure of such interworking will lead to retrials of tracking area update procedure. The following flow shows a normal procedure of the UE registering its location and updating its tracking area.

Fig 2. Initial attach and Tracking Area Update procedure

[1] After RRC connection has been completed, the UE attaches to LTE network by sending Attach Request to MME. The S1AP InitialUEMessage that delivers the Attach Request contains TAI and ECGI to inform the network of its location. Upon receiving the Attach Request, the MME performs the LTE authentication and Security measurement with the UE.

In this example, the UE requested the combined attach with its EPS attach type set to "Combined EPS/IMSI attach", which indicates that the UE wants to attach for both EPS and non-EPS services. The voice domain preference and UE usage has been set to "IMS PS voice preferred, CS voice as secondary". Given these, the UE may be able to attempt the VoLTE and if it is not available it will perform CS fallback (CSFB) procedure.

[2-3] The MME update the location of the UE to the HSS by sending Update Location Request that contains the visited PLMN identifier (i.e., MCC and MNC).

[4-7] The MME creates a GTP-C session with the SGW/PGW by sending the Create Session Request and the PGW triggers the Credit-Control-Request (CCR) to the PCRF. The Create Session Request contains the Serving-Network IE for the PLMN ID and the User Location Info IE for TAI and ECGI. The CCR contains the 3GPP-User-Location-Info AVP to accommodate the TAI and ECGI.

[8] Upon receiving the Create Session Response from the SGW/PGW, the MME responds with the Attach Accept towards the UE. The Attach Request contains the Tracking Area List (TAL), which is the list of TAIs to which UE is implicitly registered by the network. Henceforth, the UE does not need to update its location when it moves within the given TAL, thereby the intention is to reduce the traffic regarding the TAU.

In this example, the combined attach has been accepted with its attach result set to "EPS only" and supports the IMS Voice over PS session in S1 mode. The "EPS only" indicates that the EPS service has been successful but attach for non-EPS has failed[3].

[9-11] The UE establishes the E-RAB with the eNB and sends the Accept Complete to the MME. The MME requests the SGW/PGW to establish the S1 bearer by sending Modify EPS bearer request.

[12-13] The UE may trigger the TAU procedure by sending the TAU request following the criteria as defined in TS23.401. The TAU contains the UE's current TAI and the ECGI. The voice domain preference has been set to "IMS PS voice preferred, CS voice as secondary" indicating that the VoLTE is preferred but if it's not available, the UE will perform CS fallback. The EPS update type has been set to "Combined TA/LA updating with IMSI attach" indicating that the UE wants to perform an attach for non-EPS services while it is attached for EPS services. The last visited TAI is included if the UE has a valid TAI of the last visited tracking area and used by the MME to make a good list of TAI (i.e., TAL) for the UE. The TAL is contained in the TAU Accept.

III. Periodic TAU and retrying scheme

The timer, T3402 and T3411, specifies the time interval after which the tracking area update procedure takes place. Upon expiry of T3402 or T3411, the UE sends TAU with its EPS update type set to "Combined TA/LA updating with IMSI attach". The timer value is set by the network and sent to the UE being contained in the Attach Accept and every TAU response.

When the UE receives the "Network Failure" in the TAU response for combined attach, the UE starts the T3411 and retries the TAU procedure up to 5 times. If it has already retried 5 times, the UE starts the T3420 for the next retry cycle. The T3411 is recommended to be set to 10 seconds and the T3420 to 12 minutes according to 3GPP TS24.301.

The following shows the TAU retries history when the UE receives the Network Failure error for combined update. There are only 4-times of retrial in the first cycle as the first request in the Attach Request has not been counted.

NOTE The TAU procedure can also take place when the UE receives the RRC connection release with its reason set to "load balancing TAU required" for MME offloading or when the UE moves into the LTE from 3G network. This will be discussed later under the different subject.

***

In summary, the UE triggers the TAU procedure periodically and whenever it looks the UE location has changed or needs to be updated at MME. The detailed cases have been specified in 3GPP TS24.301 though, the operators may need to optimize the TAU traffic in a way not to cause a traffic overload. The traffic can be possibly controlled by finding out appropriate timer values or by optimizing the list of TAIs in a TAL.

Red Mouse 

REFERENCES

[1] 3GPP TS23.003, "Numbering, Addressing and Identification", v11.0.0, Dec 2011
[2] 3GPP TS23.401, "General Packet Radio Service(GPRS) enhancements for Evolved Universal Traversal Radio Access Network(E-UTRAN) access", v12.4.0, Mar 2014

[3] 3GPP TS24.301, "Non-Access-Stratum(NAS) protocol for Evolved Packet System(EPS); stage3", v12.4.0, Mar 2014

[4] 3GPP TS36.300, "Evolved Universal Traversal Radio Access (E-UTRA) and Evolved Universal Traversal Radio Access Network(E-UTRAN); Overall description; stage2", v10.0.0, Jun 2010
[5] 3GPP TS36.331, "Evolved Universal Traversal Radio Access (E-UTRA); Radio Resource Control(RRC); Protocol specification", v12.6.0, Jun 2015
[6] Netmanias, "LTE Identification