VoLTE S1 handover completion (3/3)

After the UE confirms the completion of the S1 handover in a new cell, the target eNB initiates the completion procedure by sending towards the network. The completion procedure involves the SGW and PGW replacing the downlink S1 bearer from the source eNB to the source eNB releasing the UE associated resources and lastly, the release of Indirect Forwarding Tunnels by the SGW.

Figure 1. S1 handover procedure - complete

[12] The target eNB sends the Handover Notify to the MME notifying that the UE has been identified in the target cell and S1 handover has been completed.

• E-UTRAN CGI: globally identifies a cell and composed of PLMN identity and Cell identity.
• TAI: uniquely identify a Tracking Area and composed of PLMN identity and TAC (Tracking Area Code). 

Figure 2. Handover Notify

[13] The MME sends the Modify Bearer Request to the SGW requesting to switch the media path from the source eNB to the target eNB. The switch of media path from the SGW is done by updating the S1 bearer tunneling end point.

• Indication: includes various application flags. Refer to section 7.2.7 “Modify Bearer Request” in [TS 29.274].
• Bearer Context: identifies the bearer to be modified. It is composed of EPS Bearer ID and S1 eNB F-TEID sub elements.

In this practice, there are three bearers, i.e., bearer id = 5,6,7. The bearer 5 and 6 belongs to the internet APN, meanwhile the bearer 7 belongs to the IMS APN. The MME sends multiple Modify Bearer Request for each APN.

Note that in the following message, the value of F-TEID is set to 0x018088ed. This value is the one that was sent by the target eNB in step [4] Handover Request Acknowledge where it represents the tunneling end point of the target eNB for the downlink S1 bearer. Now, the MME deliver this value to the S-GW so the S-GW can use this value as a new target end point for the media path. As such, the downlink media path on S1 bearer is switched from the source eNB to the target eNB.

Figure 3. Modify Bearer Request for bearer 7

The following message shows the Modify Bearer Request for the bearer 5 and 6. In the same as described in the above, the value of F-TEID is set to 0x018008ed and 0x018048ed for bearer 5 and 6, respective. They are the same values as were sent by the target eNB in step [4] Handover Request Acknowledge where they represent tunneling end points of the target eNB for the downlink S1 bearer.

Figure 4. Modify Bearer Request for bearer 5 and 6

Note that the Modify Bearer Request is also sent by the SGW to the PGW which is not depicted here. The SGW and PGW are logically separated but in many cases, they are deployed on the same platform.

[14] The Modify Bearer Response sent from the PGW to the SGW and from the SGW to the MME.

• Cause: indicates the result of modifying the requested bearer.
• Bearer Context: identifies the bearer that has been modified. It is composed of EPS Bearer ID, Cause and S1 eNB F-TEID sub elements.

Note that in the following message, the value of F-TEID is set to 0x005b8422. This value is the one that was sent by the MME to the target eNB in step [3] Handover Request where it represents the tunneling end point of the SGW for the uplink S1 bearer. The tunneling end point values of SGW is created when the UE attaches the network at the very beginning. Upon S1 handover, the MME sends this value to the target eNB and the target eNB establishes uplink S1 bearer to the SGW using this value. At this moment, the source eNB also has uplink S1 bearer connection with the SGW with the same tunneling end points. As both source eNB and target eNB can send traffic to the SGW, there is no change that the data sent by the UE gets lost during handover procedure.

Now, the SGW sends this TEID to the MME for the downlink S1 bearer informing the tunneling end point for the downlink S1 bearer.

Figure 5. Modify Bearer Response for bearer 7

The same principal is applied to the following message which is for bearer 5 and 6.

Figure 6. Modify Bearer Response for bearer 5 and 6

[15] The MME requests the release of UE-associated S1 logical connections over S1 interface by sending UE Context Release Command. The source eNB releases its resources related to the given UE.

• UE S1AP ID Pair: contains the UE S1AP identifiers. It is composed of MME UE S1AP ID and eNB UE S1AP ID
• Cause: indicates the reason for this command. Refer to section 9.2.1.3 “Cause” in TS36.413 for the detail.

Figure 7. UE Context Release Command

[16] The source eNB confirms the release of UE-associated S1 logical connections over S1 interface by sending UE Context Release Complete.

• MME UE S1AP ID: uniquely identifies the UE association with the S1 interface within the MME.
• eNB UE S1AP ID: uniquely identifies the UE association with the S1 interface within the eNB.

Figure 8. UE Context Release Complete

[17] The MME requests the SGW to delete the Indirect Forwarding Tunnels by sending Delete Indirect Data Forwarding Tunnel Request.

Figure 9. Delete Indirect Data Forwarding Tunnel Request

[18] The SGW responds with Delete Indirect Data Forwarding Tunnel Response to the MME.

Figure 10. Delete Indirect Data Forwarding Tunnel Response

Red Mouse

REFERENCES

[1] 3GPP TS23.401, “General Packet Radio Service (GPRS) enhancements for Evolved Universal Terrestrial Radio Access Network (E-UTRAN) access”, v12.4.0, Mar 2014

[2] 3GPP TS29.274, “Evolved General Packet Radion Service (GPRS) Tunneling Protocol for Control Plane (GTPv2-C); Stage3”, v13.2.0, Jun 2015

[3] 3GPP TS36.411, "Evolved Universal Terrestrial Radio Access (E-UTRA); S1 Application Protocol (S1AP)", v13.0.0, Jun 2015

VoLTE S1 handover execution (2/3)

S1 handover preparation procedure includes the decision of S1 handover by the source eNB, allocation of network resources to establish Indirect Data Forwarding Tunnel among two eNBs and common S-GW and establishment of uplink S1 bearer from the target eNB to the S-GW. Once the S1 handover preparation procedure is completed, the MME initiates the S1 handover by sending Handover Command to the UE via the source eNB. The UE executes the handover by detaching from the source eNB and attaching to the target eNB. Meanwhile, the downlink data is forwarded to the target eNB and buffered while the UE handover is in progress. Lastly, the UE informs the target eNB of the fact that the handover has been successfully completed and thereafter, the buffered data and downlink data is forwarded the UE through the target eNB. 

Figure 1. S1 handover procedure - execution

[8] The Handover Command received from the MME is wrapped by the source eNB within the RRC Connection Reconfiguration and sent to the UE. The RRC Connection Reconfiguration is the message to perform logical, transport and physical channel configurations. In this case, it is used to send NAS signaling to the UE to reduce the latency. Upon receiving the Handover Command, the UE detaches from the source eNB and performs handover to the target eNB.

[9] The source eNB stops assigning PDCP-SNs to downlink packets and sends the eNB Status Transfer to the target eNB via MME that contains uplink and downlink PDCP-SN and HFN (Hyper Frame Number) for each respective E-RAB. This procedure is initiated by the source eNB at the moment when it considers the transmitter/receiver status to be frozen. The use of PDCP-SN and HFN is part of overflow control mechanism for radio. The PDCP-SN is the serial number of PDCP packets increasing up to MAX-PDCN-SN. If the number reaches the MAX-PDCP-SN, the HFN is incremented by one. 

• Subject to Transfer Items: contains uplink/downlink PDCP-SN and HFN for each respective E-RAB.
• E-RAB ID: Identifies a radio access bearer for a particular UE. This value remains the same after S1-handover.
• uL-/dL-Count value: contains the PDCP-SN and HFN values
• Received Status of UL PDCP SDUs: indicates the missing and the received uplink SDUs (Service Data Units) for each bearer for which the source eNB has accepted the request from the target eNB for uplink forwarding.

  

Figure 2. eNB Status Transfer

[10] The MME forwards the received PDCP-SN and HFN information to the target eNB by sending MME Status Transfer. Upon receiving the MME Status Transfer, the target eNB does not deliver any uplink packet whose PDCP-SN is lower than the value received in the PDCP-SN in the uL-COUNT value. The target eNB uses the received PDCP-SN in the dL-COUNT value for the first downlink packet for which no PDCP-SN is assigned yet. The downlink traffic received by the source eNB is routed to the target eNB following the Indirect Data Forwarding Tunnel.

[11] After the UE successfully synchronized with the target cell, it sends a Handover Confirm to the target eNB. The Handover Confirm is contained in the RRC Connection Reconfiguration Complete message on RRC. Please note that, at this moment, there is no direct S1 bearer established yet between S-GW and the target eNB. Therefore, the downlink data is sent to the source eNB and forwarded to the target eNB following the Indirect Data Forwarding Tunnel. The uplink data from the UE will be forwarded to the S-GW following the direct S1 interface.

Red Mouse

REFERENCES

[1] 3GPP TS25.331, "Radio Resource Network (RRC); Protocol specification", v12.3.0, Sep 2014

[2] 3GPP TS23.401, “General Packet Radio Service (GPRS) enhancements for Evolved Universal Terrestrial Radio Access Network (E-UTRAN) access”, v12.4.0, Mar 2014

[3] 3GPP TS36.331, "Evolved Universal Terrestrial Radio Access (E-UTRA); Radio Resource Control (RRC); Protocol specification", v12.3.0, Sep 2009

[4] Blog, “How LTE Stuff Works?: RRC Connection Reconfiguration”, Oct 2011