4.3.2a Support for Dual Connectivity

23.4013GPPGeneral Packet Radio Service (GPRS) enhancements for Evolved Universal Terrestrial Radio Access Network (E-UTRAN) accessRelease 18TS

Dual Connectivity involves two RAN nodes, i.e. Master and Secondary RAN nodes (see TS 36.300 [5] for the definitions), in providing radio resources to a given UE (with active radio bearers), while a single S1-MME termination point exists for an UE between a MME and the E-UTRAN. The E-UTRAN architecture and related functions to support Dual Connectivity with E-UTRAN is further described in TS 36.300 [5]. Dual Connectivity with E-UTRAN as Master RAN node and NR as Secondary RAN node is further described in TS 37.340 [85].

Dual connectivity defines "Master Cell Group (MCG) bearer" and "Secondary Cell Group (SCG) bearer" alternatives (see TS 36.300 [5]). For E-RABs configured as "MCG bearers" the U-plane termination points are maintained, whereas for E-RABs configured as "SCG bearers" it enables changing the U-plane termination point in the E-UTRAN by means of S1-MME signalling without changing the S1-MME termination point.

Dual Connectivity also defines a "split bearer" alternative TS 36.300 [5]. The "split bearer" in the E-UTRAN is transparent to the core network entities (e.g. MME, S-GW etc.) with the exception of the CSG membership verification by the MME when the Secondary eNodeB is a hybrid access eNodeB.

The E-UTRAN uses the per-UE information supplied by the MME and local E-UTRAN configuration data to determine whether or not to use Dual Connectivity for that UE, and, on a per EPS bearer basis the E-UTRAN decides whether to use an MCG bearer or SCG bearer, and, whether or not that bearer is a "split bearer".

NOTE 1: Typically, the MME and SGW cannot determine whether the RAN termination point(s) for the S1-U interface are located on a Master RAN node that has multiple IP addresses, or, on a Secondary RAN node.

If the UE has indicated support for Dual Connectivity with NR and MME has an Access Restriction for NR for a UE (either signalled from the HSS, or, locally generated by VPLMN policy in the MME) then the MME shall signal this to the E-UTRAN as part of Handover Restriction List and to the UE in Attach and TAU Accept as defined in clauses 5.5.2.2.3, 5.5.2.4.3, 5.3.2.1, 5.3.3.1, 5.3.3.2 and D.3.6 respectively. An eNodeB supporting Dual Connectivity with NR checks whether the UE is allowed to use NR. If the UE is not allowed to use NR, the eNodeB shall not establish Dual Connectivity with NR as a secondary RAT.

The MME uses "UE support for dual connectivity with NR" for SGW and PGW selection when the UE indicates support for NR and there is no Access Restriction for NR for the UE.

An E-UTRAN cell, based on operator configuration, broadcasts whether it is capable of supporting dual connectivity with locally available NR secondary cell(s).

At inter-RAT handover from NR or GERAN/UTRAN, the Access Restriction for NR is either already in the MME’s UE context, or, is obtained from the HSS during the subsequent Tracking Area Update procedure (i.e. not from the source AMF/SGSN or source RAN). In both inter-RAT handover cases, any NR Access Restriction is then signalled to the E-UTRAN.

NOTE 2: This signalling of the Access Restriction during the TAU after the inter-RAT handover procedure means that there is a small risk that NR resources are transiently allocated.

The eNodeB, at which the S1-MME terminates, performs all necessary S1-MME related functions (as specified for any serving eNodeB) such as mobility management, relaying of NAS signalling, E-RAB handling, etc. and manages the handling of user plane connection of S1-U.

Additional functional characteristics are:

– User location information reporting is based on the identity of the cell that is serving the UE and supported by the eNodeB terminating S1-MME. The cell identity of the Primary cell in the secondary RAN node may also be included, if available.

– Path update signalling for E-RABs configured as "SCG bearers" and Serving GW relocation cannot occur at the same time.

– During handover with dual connectivity, the requirement of forwarding "end marker" packets to target node is also applicable to secondary RAN node if it is the source node for S1-U bearer.

– After handover with data forwarding, the E-UTRAN initiated E-RAB modification procedure of clause 5.4.7 should not be initiated by the target eNodeB before "end marker" packet is received at the target RAN node or a timer in target eNodeB expires.

– Relaying function is not supported.

– CSG function may be supported if the Secondary eNodeB is a hybrid access eNodeB (see more details in clause 5.4.7 and in TS 36.300 [5]).

NOTE 3: A HeNB cannot be the Master eNodeB, i.e. a HeNB cannot initiate the Secondary eNodeB Addition procedure.

NOTE 4: A HeNB is not allowed to be the Secondary eNodeB if the HeNB is a closed access eNodeB.

– When the Secondary eNodeB is a hybrid access eNodeB, the Master eNodeB may ask CSG membership verification to the MME using E-RAB Modification Indication message (for SCG bearers) or UE Context Modification Indication (for split bearers) message. The MME shall determine the CSG membership based on the CSG Membership Information as specified in TS 36.300 [5] and shall respond to the Master eNodeB using respectively a E-RAB Modification Confirm or a UE Context Modification Confirm, but shall not update the User CSG Information in the Core Network.

– The LIPA function may be supported for the SCG bearer alternative, in the case that the Secondary eNodeB is a HeNB with a collocated L-GW (see more details in TS 36.300 [5]).

– "SIPTO at the Local Network with L-GW function collocated with the (H)eNB" function may be supported (see more details in TS 36.300 [5]):

– For the MCG and split bearer alternatives, if the Master eNodeB is collocated with a L-GW; and/or

– For the SCG bearer alternative, if the Secondary eNodeB is a (H)eNB with a collocated L-GW.

NOTE 5: LIPA or SIPTO at the Local Network PDN connection can be established if the SeNodeB has already been added before the UE requests establishment of the LIPA or SIPTO at the Local Network PDN connection.

NOTE 6: LIPA or SIPTO at the Local Network PDN connection can be established if the UE is in the coverage of the candidate SeNodeB when the UE requests establishment of the LIPA or SIPTO at the Local Network PDN connection, but the SeNodeB has not yet been added. In this case, there is a time gap between the moment when the PDN connection establishment is completed and the moment when the SeNodeB Addition procedure is completed.

– "SIPTO at the Local Network with stand-alone GW" function may be supported for the MCG, SCG, and split bearer alternatives if the Master and Secondary eNodeBs belong to the same LHN (see more detail in TS 36.300 [5]).