4.3.20 Relaying function

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

4.3.20.1 General

The relaying function enables an operator to improve and extend the coverage area by having a Relay Node (RN) wirelessly connected to an eNodeB serving the RN, called Donor eNodeB (DeNB), via a modified version of the E-UTRA radio interface called the Un interface as specified in TS 36.300 [5].

The relaying function and use of RN/DeNB entities in a network is transparent to the operations of the UEs connected to it and associated core network entities (e.g. MME, S/P-GW, PCRF etc.) for the UEs.

The relaying architecture is shown in figure 4.3.20.1-1.

Figure 4.3.20.1-1: Relaying Architecture

NOTE 1: Impact to core network elements from the introduction of RNs and DeNB is minimized by reusing the existing nodes and protocols when interacting with the core network.

NOTE 2: Functions of the MME for the RN and MME for the UE may be collocated in a single MME.

The RN supports the eNodeB functionality like termination of the radio protocols of the E-UTRA radio interface and the S1 and X2 interfaces. The RN also supports a subset of the UE functionality and protocols to wirelessly connect to the DeNB.

In addition to supporting eNodeB functionality, the DeNB also embeds and provides the S-GW/P-GW-like functions needed for the RN operation. This includes creating a session for the RN and managing EPS bearers for the RN as shown in clause 4.3.20.3, as well as terminating the S1-AP and S11 interfaces towards the MME serving the RN. Due to the proxy functionality, the DeNB appears as an MME (for S1), an eNodeB (for X2) and an S-GW to the RN.

The RN and DeNB also perform mapping of signalling and data packets onto EPS bearers that are setup for the RN. The mapping is based on existing QoS mechanisms defined for the UE and the P-GW and are described in TS 36.300 [5].

4.3.20.2 RN startup and attach procedure

4.3.20.2.1 General

The startup procedure for the Relay Node is based on the normal UE attach procedure and consists of the following two phases:

– Phase I: Attach for RN preconfiguration.

– Phase II: Attach for RN operation (MME of the RN).

NOTE: When the certificate-based solution is used, the RN uses USIM-INI in Phase I and USIM-RN in Phase II with necessarily different IMSIs. When pre-shared key is used, there is only need for one USIM and the RN uses the same IMSI during Phase I and Phase II. The MME does not treat certificate-based and pre-shared key-based solution differently. The use of the certificate-based and pre-shared key solutions is specified in Annex D of TS 33.401 [41].

4.3.20.2.2 Attach for RN preconfiguration

The RN attaches to the E-UTRAN/EPC as a UE at power-up (i.e. the RN shall not include the RN indication in the RRC Connection establishment signalling). The eNodeB treats the RN as a normal UE when performing MME selection.

Because the eNodeB does not indicate that this is a RN in the S1 interface Initial UE message, the MME does not perform any further RN specific actions (e.g. it ignores any indication from the HSS that "this subscription includes a permission to operate as a RN").

The authentication of the "RN acting as an UE" is performed by the MME during this attach procedure, using the information obtained from the HSS.

The MME performs the S-GW and P-GW selection as for a normal UE.

NOTE: It is the responsibility of the HSS operator to ensure that the RN subscription includes an APN configuration that ensures that the RN subscription cannot be used for other purposes, e.g. only a single APN is configured for the use of RNs in phase I, and, that this APN is reserved for RNs only.

The RN retrieves initial RN configuration parameters as user plane traffic, across the SGi reference point, from RN OAM (e.g. list of DeNB cells and selected PLMN).

After this operation is complete, the RN detaches from the network using the normal UE initiated detach procedure, see clause 5.3.8.2.1 and the RN triggers Phase II.

4.3.20.2.3 Attach for RN operation

To start relay operations, the normal attach procedure, with the following exceptions, is applied:

– The RN and the USIM-RN perform local security operations (e.g. establishment of a secure channel between them) as specified in TS 33.401 [41];

– The RN selects a cell from the list acquired during Phase I;

– The RN establishes an RRC connection with the DeNB, indicating that the connection is for a RN;

– The DeNB is aware of the MMEs that support RN functionality. In all cases when the RN indication is received, the DeNB shall ensure that the current or (re)selected MME supports RN functionality;

NOTE 1: The RN follows normal UE behaviour, e.g. the RN’s NAS may use either an IMSI or a GUTI. Also, the RN’s NAS may or may not provide an S-TMSI to the RN’s AS, and hence, the RRCConnectionRequest may either contain an S-TMSI or a random value.

– In the S1 interface Initial UE Message, the DeNB sends the RN indication to the MME. This message also carries the IP address of the S‑GW/P-GW function embedded in the DeNB;

– The subscription data supplied to the MME by the HSS for USIM-RN includes an indication that the subscription is permitted to be used by a RN.

– If the S1 interface Initial UE Message indicates that this is a RN, but the subscription data does not indicate that the subscription includes a permission to operate as a RN, then the MME shall reject the NAS procedure (e.g. Attach Request, Tracking Area Update Request, Service Request, etc) with an appropriate cause value (e.g. one that avoids retries on this PLMN yet does not harm a RN that has unexpectedly performed PLMN reselection).

NOTE 2: It is anticipated that the MME checks that the HPLMN of the USIM-RN is authorised to attach RNs to this MME.

– The MME and RN perform the normal EPS Authentication procedures.

– MME (RN) selects the S-GW/P-GW in DeNB for the RN based on the IP address included in the Initial UE Message (i.e. all GW selection and APN related procedures are bypassed during this phase). The MME performs S11 interface signalling with the S-GW/P-GW located in the DeNB;

– The MME accepts the attach procedure and sets up an S1 context with the DeNB.

When relay function is enabled, MMEs in a pool should all have the same relaying function capability in order to have consistent support for functions such as redundancy, load balancing.

Figure 4.3.20.2-1: RN attach procedure

The detach procedure for the RN is the same as the normal UE detach procedure, though the RN should ensure that no UE is connected to the RN cells before detaching. It is up to RN implementation how it ensures no UE is connected.

4.3.20.3 DeNB E-RAB activation/modification

This procedure is used by the DeNB to change the EPS bearer allocation for the RN. The procedure is the same as the normal network-initiated bearer activation/modification procedure with the exception that the S-GW/P‑GW functionality (steps 1 and 6) is performed by the DeNB.

Figure 4.3.20.3-1: DeNB-initiated bearer activation/modification procedure

NOTE: It is up to implementation if and when the DeNB sets up and modifies EPS bearers for the RN, in addition to the initial bearer set up procedures at Attach.