5.3 Registration and Connection Management
23.5013GPPRelease 18System architecture for the 5G System (5GS)TS
5.3.1 General
The Registration Management is used to register or deregister a UE/user with the network, and establish the user context in the network. The Connection Management is used to establish and release the signalling connection between the UE and the AMF.
5.3.2 Registration Management
5.3.2.1 General
A UE/user needs to register with the network to receive services that requires registration. Once registered and if applicable the UE updates its registration with the network (see TS 23.502 [3]):
– periodically, in order to remain reachable (Periodic Registration Update); or
– upon mobility (Mobility Registration Update); or
– to update its capabilities or re-negotiate protocol parameters (Mobility Registration Update).
The Initial Registration procedure involves execution of Network Access Control functions as defined in clause 5.2 (i.e. user authentication and access authorization based on subscription profiles in UDM). As result of the Registration procedure, the identifier of the serving AMF serving the UE in the access through which the UE has registered will be registered in UDM.
The registration management procedures are applicable over both 3GPP access and Non-3GPP access. The 3GPP and Non-3GPP RM states are independent of each other, see clause 5.3.2.4.
5.3.2.2 5GS Registration Management states
5.3.2.2.1 General
Two RM states are used in the UE and the AMF that reflect the registration status of the UE in the selected PLMN:
– RM-DEREGISTERED.
– RM-REGISTERED.
5.3.2.2.2 RM-DEREGISTERED state
In the RM‑DEREGISTERED state, the UE is not registered with the network. The UE context in AMF holds no valid location or routing information for the UE so the UE is not reachable by the AMF. However, some parts of UE context may still be stored in the UE and the AMF e.g. to avoid running an authentication procedure during every Registration procedure.
In the RM-DEREGISTERED state, the UE shall:
– attempt to register with the selected PLMN using the Initial Registration procedure if it needs to receive service that requires registration (see clause 4.2.2.2 of TS 23.502 [3]).
– remain in RM-DEREGISTERED state if receiving a Registration Reject upon Initial Registration (see clause 4.2.2.2 of TS 23.502 [3]).
– enter RM-REGISTERED state upon receiving a Registration Accept (see clause 4.2.2.2 of TS 23.502 [3]).
When the UE RM state in the AMF is RM-DEREGISTERED, the AMF shall:
– when applicable, accept the Initial Registration of a UE by sending a Registration Accept to this UE and enter RM-REGISTERED state for the UE (see clause 4.2.2.2 of TS 23.502 [3]); or
– when applicable, reject the Initial Registration of a UE by sending a Registration Reject to this UE (see clause 4.2.2.2 of TS 23.502 [3]).
5.3.2.2.3 RM-REGISTERED state
In the RM‑REGISTERED state, the UE is registered with the network. In the RM-REGISTERED state, the UE can receive services that require registration with the network.
In the RM-REGISTERED state, the UE shall:
– perform Mobility Registration Update procedure if the current TAI of the serving cell (see TS 37.340 [31]) is not in the list of TAIs that the UE has received from the network in order to maintain the registration and enable the AMF to page the UE;
NOTE: Additional considerations for Mobility Registration Update in case of NR satellite access are provided in clause 5.4.11.6.
– perform Periodic Registration Update procedure triggered by expiration of the periodic update timer to notify the network that the UE is still active.
– perform a Mobility Registration Update procedure to update its capability information or to re-negotiate protocol parameters with the network;
– perform Deregistration procedure (see clause 4.2.2.3.1 of TS 23.502 [3]), and enter RM-DEREGISTERED state, when the UE needs to be no longer registered with the PLMN. The UE may decide to deregister from the network at any time.
– enter RM-DEREGISTERED state when receiving a Registration Reject message or a Deregistration message. The actions of the UE depend upon the cause value’ in the Registration Reject or Deregistration message. See clause 4.2.2 of TS 23.502 [3].
When the UE RM state in the AMF is RM-REGISTERED, the AMF shall:
– perform Deregistration procedure (see clauses 4.2.2.3.2, 4.2.2.3.3 of TS 23.502 [3]), and enter RM-DEREGISTERED state for the UE, when the UE needs to be no longer registered with the PLMN. The network may decide to deregister the UE at any time;
– perform Implicit Deregistration at any time after the Implicit Deregistration timer expires. The AMF shall enter RM-DEREGISTERED state for the UE after Implicit Deregistration;
– when applicable, accept or reject Registration Requests or Service Requests from the UE.
5.3.2.2.4 5GS Registration Management State models
Figure 5.3.2.2.4-1: RM state model in UE
Figure 5.3.2.2.4-2: RM state model in AMF
5.3.2.3 Registration Area management
Registration Area management comprises the functions to allocate and reallocate a Registration area to a UE. Registration area is managed per access type i.e. 3GPP access or Non-3GPP access.
When a UE registers with the network over the 3GPP access, the AMF allocates a set of tracking areas in TAI List to the UE. When the AMF allocates registration area, i.e. the set of tracking areas in TAI List, to the UE it may take into account various information (e.g. Mobility Pattern and Allowed/Non-Allowed Area (refer to clause 5.3.4.1)). An AMF which has the whole PLMN as serving area may alternatively allocate the whole PLMN ("all PLMN") as registration area to a UE in MICO mode (refer to clause 5.4.1.3). When AMF allocates registration area for UE registered for Disaster Roaming service as specified in clause 5.40.4, AMF shall only consider TAIs covering the area with the Disaster Condition.
The 5G System shall support allocating a Registration Area using a single TAI List which includes tracking areas of any NG-RAN nodes in the Registration Area for a UE.
In the case of SNPN, the TAI list allocated by AMF does not support Tracking Areas belonging to different SNPNs.
TAI used for non-3GPP access shall be dedicated to non-3GPP access. TAI(s) dedicated to Non-3GPP access may be defined in a PLMN and apply within this PLMN. Each N3IWF, TNGF, TWIF and W-AGF is locally configured with one TAI value. Each N3IWF, TNGF, TWIF and W-AGF may be configured with a different TAI value or with the same TAI value as other N3IWFs, TNGFs, TWIFs and/or W-AGFs. The TAI is provided to the AMF during N2 interface setup and as part of the User Location Information in UE associated messages as described in TS 38.413 [34].
When a UE registers with the network over a Non-3GPP access, the AMF allocates to the UE a registration area that only includes the TAI received from the serving N3IWF, TNGF, TWIF or W-AGF.
NOTE 1: For example, two W-AGFs can each correspond to a different TAI (one TAI per W-AGF) and thus support different sets of S-NSSAI(s).
When generating the TAI list, the AMF shall include only TAIs that are applicable on the access type (i.e. 3GPP access or Non-3GPP access) where the TAI list is sent.
NOTE 2: To prevent extra signalling load resulting from Mobility Registration Update occurring at every RAT change, it is preferable to avoid generating a RAT-specific TAI list for a UE supporting more than one RAT.
NOTE 3: For a UE registered on N3GPP access the TAI(s) provided to the UE as part of the Registration Area is expected to enable the support of the slices that are intended to be provided for this UE over this specific Non-3GPP access. In addition, the Registration Area provided to the UE for non-3GPP access will never change until the UE deregisters from non-3GPP access (either explicit deregistration or implicit deregistration due to Deregistration timer expiring due to UE entering CM_IDLE state).
For all 3GPP Access RATs in NG-RAN and for Non-3GPP Access, the 5G System supports the TAI format as specified in TS 23.003 [19] consisting of MCC, MNC and a 3-byte TAC only.
The additional aspects for registration management when a UE is registered over one access type while the UE is already registered over the other access type is further described in clause 5.3.2.4.
To ensure a UE initiates a Mobility Registration procedure when performing inter-RAT mobility to or from NB-IoT, a Tracking Area shall not contain both NB-IoT and other RATs cells (e.g. WB-E-UTRA, NR), and the AMF shall not allocate a TAI list that contains both NB-IoT and other RATs Tracking Areas.
For 3GPP access the AMF determines the RAT type the UE is camping on based on the Global RAN Node IDs associated with the N2 interface and additionally the Tracking Area indicated by NG-RAN. When the UE is accessing NR using unlicensed bands, as defined in clause 5.4.8, an indication is provided in N2 interface as defined in TS 38.413 [34].
The AMF may also determine more precise RAT Type information based on further information received from NG-RAN:
– The AMF may determine the RAT Type to be LTE-M as defined in clause 5.31.20; or
– The AMF may determine the RAT Type to be NR using unlicensed bands, as defined in clause 5.4.8.
– The AMF may determine the RAT Type to be one of the RAT types for satellite access, as defined in clause 5.4.10.
– The AMF may determine the RAT Type to be NR RedCap as defined in clause 5.41.
For Non-3GPP accesses the AMF determines the RAT type the UE is camping based on the 5G-AN node associated with N2 interface as follows:
– The RAT type is Untrusted Non-3GPP if the 5G-AN node has a Global N3IWF Node ID;
– The RAT type is Trusted Non-3GPP if the 5G-AN node has a Global TNGF Node ID or a Global TWIF Node ID; and
– The RAT type is Wireline -BBF if the 5G-AN node has a Global W-AGF Node ID corresponding to a W-AGF supporting the Wireline BBF Access Network. The RAT type is Wireline-Cable if the 5G-AN node has a Global W-AGF Node ID corresponding to a W-AGF supporting the Wireline Cable Access Network. If not possible to distinguish between the two, the RAT type is Wireline.
NOTE 4: How to differentiate between W-AGF supporting either Wireline BBF Access Network or the Wireline (e.g. different Global W-AGF Node ID IE or the Global W-AGF Node ID including a field to distinguish between them) is left to Stage 3 definition.
NOTE 5: If an operator supports only one kind of Wireline Access Network (either Wireline BBF Access Network or a Wireline Cable Access Network) the AMF may be configured to use RAT type Wireline or the specific one.
For Non-3GPP access the AMF may also use the User Location Information provided at N2 connection setup to determine a more precise RAT Type, e.g. identifying IEEE 802.11 access, Wireline-Cable access, Wireline-BBF access.
When the 5G-AN node has either a Global N3IWF Node ID, or a Global TNGF Node ID, or a Global TWIF Node ID, or a Global W-AGF Node ID, the Access Type is Non-3GPP Access.
5.3.2.4 Support of a UE registered over both 3GPP and Non-3GPP access
This clause applies to Non-3GPP access network corresponding to the Untrusted Non-3GPP access network, to the Trusted Non-3GPP and to the W-5GAN. In the case of W-5GAN the UE mentioned in this clause corresponds to the 5G-RG.
For a given serving PLMN there is one RM context for a UE for each access, e.g. when the UE is consecutively or simultaneously served by a 3GPP access and by a non-3GPP access (i.e. via an N3IWF, TNGF and W-AGF) of the same PLMN. UDM manages separate/independent UE Registration procedures for each access.
When served by the same PLMN for 3GPP and non-3GPP accesses, an UE is served by the same AMF except in the temporary situation described in clause 5.17 i.e. after a mobility from EPS while the UE has PDU Sessions associated with non-3GPP access.
The 5G NSWO authentication as defined in Annex S of TS 33.501 [29] does not impact the RM state.
An AMF associates multiple access-specific RM contexts for an UE with:
– a 5G-GUTI that is common to both 3GPP and Non-3GPP accesses. This 5G-GUTI is globally unique.
– a Registration state per access type (3GPP / Non-3GPP)
– a Registration Area per access type: one Registration Area for 3GPP access and another Registration Area for non 3GPP access. Registration Areas for the 3GPP access and the Non-3GPP access are independent.
– timers for 3GPP access:
– a Periodic Registration timer; and
– a Mobile Reachable timer and an Implicit Deregistration timer.
– timers for non-3GPP access:
– a UE Non-3GPP Deregistration timer; and
– a Network Non-3GPP Implicit Deregistration timer.
The AMF shall not provide a Periodic Registration Timer for the UE over a Non-3GPP access. Consequently, the UE need not perform Periodic Registration Update procedure over Non-3GPP access. Instead, during the Initial Registration procedure and Re-registration, the UE is provided by the network with a UE Non-3GPP Deregistration timer that starts when the UE enters non-3GPP CM-IDLE state.
When the 3GPP access and the non-3GPP access for the same UE are served by the same PLMN, the AMF assigns the same 5G-GUTI for use over both accesses. Such a 5G-GUTI may be assigned or re-assigned over any of the 3GPP and Non-3GPP accesses. The 5G-GUTI is assigned upon a successful registration of the UE, and is valid over both 3GPP and Non-3GPP access to the same PLMN for the UE. Upon performing an initial access over the Non-3GPP access or over the 3GPP access while the UE is already registered with the 5G System over another access of the same PLMN, the UE provides the native 5G-GUTI for the other access. This enables the AN to select an AMF that maintains the UE context created at the previous Registration procedure via the GUAMI derived from the 5G-GUTI, and enables the AMF to correlate the UE request to the existing UE context via the 5G-GUTI.
If the UE is performing registration over one access and intends to perform registration over the other access in the same PLMN (e.g. the 3GPP access and the selected N3IWF, TNGF or W-AGF are located in the same PLMN), the UE shall not initiate the registration over the other access until the Registration procedure over first access is completed.
NOTE: To which access the UE performs registration first is up to UE implementation.
When the UE is successfully registered to an access (3GPP access or Non-3GPP access respectively) and the UE registers via the other access:
– if the second access is located in the same PLMN (e.g. the UE is registered via a 3GPP access and selects a N3IWF, TNGF or W-AGF located in the same PLMN), the UE shall use for the registration to the PLMN associated with the new access the 5G-GUTI that the UE has been provided with at the previous registration or UE configuration update procedure for the first access in the same PLMN. Upon successful completion of the registration to the second access, if the network included a 5G-GUTI in the Registration Accept, the UE shall use the 5G-GUTI received in the Registration Accept for both registrations. If no 5G-GUTI is included in the Registration Accept, then the UE uses the 5G-GUTI assigned for the existing registration also for the new registration.
– if the second access is located in a PLMN different from the registered PLMN of the first access (i.e. not the registered PLMN), (e.g. the UE is registered to a 3GPP access and selects a N3IWF, TNGF or W-AGF located in a PLMN different from the PLMN of the 3GPP access, or the UE is registered over Non-3GPP and registers to a 3GPP access in a PLMN different from the PLMN of the N3IWF, TNGF or W-AGF), the UE shall use for the registration to the PLMN associated with the new access a 5G-GUTI only if it has got one previously received from a PLMN that is not the same as the PLMN the UE is already registered with. If the UE does not include a 5G-GUTI, the SUCI shall be used for the new registration. Upon successful completion of the registration to the second access, the UE has the two 5G-GUTIs (one per PLMN).
A UE supporting registration over both 3GPP and Non-3GPP access to two PLMNs shall be able to handle two separate registrations, including two 5G-GUTIs, one per PLMN, and two associated equivalent PLMN lists.
When a UE 5G-GUTI assigned during a Registration procedure over 3GPP (e.g. the UE registers first over a 3GPP access) is location-dependent, the same UE 5G-GUTI can be re-used over the Non-3GPP access when the selected N3IWF, TNGF or W-AGF function is in the same PLMN as the 3GPP access. When an UE 5G-GUTI is assigned during a Registration procedure performed over a Non 3GPP access (e.g. the UE registers first over a non-3GPP access), the UE 5G-GUTI may not be location-dependent, so that the UE 5G-GUTI may not be valid for NAS procedures over the 3GPP access and, in this case, a new AMF is allocated during the Registration procedure over the 3GPP access.
When the UE is registered first via 3GPP access, if the UE registers to the same PLMN via Non-3GPP access, the UE shall send the GUAMI obtained via 3GPP access to the N3IWF, TNGF or W-AGF, which uses the received GUAMI to select the same AMF as the 3GPP access.
The Deregistration Request message indicates whether it applies to the 3GPP access the Non-3GPP access, or both.
If the UE is registered on both 3GPP and Non-3GPP accesses and it is in CM-IDLE over Non-3GPP access, then the UE or AMF may initiate a Deregistration procedure over the 3GPP access to deregister the UE only on the Non-3GPP access, in which case all the PDU Sessions which are associated with the Non-3GPP access shall be released.
If the UE is registered on both 3GPP and non-3GPP accesses and it is in CM-IDLE over 3GPP access and in CM-CONNECTED over non-3GPP access, then the UE may initiate a Deregistration procedure over the non-3GPP access to deregister the UE only on the 3GPP access, in which case all the PDU Sessions which are associated with the 3GPP access shall be released.
Registration Management over Non-3GPP access is further defined in clause 5.5.1.
5.3.3 Connection Management
5.3.3.1 General
Connection management comprises the functions of establishing and releasing a NAS signalling connection between a UE and the AMF over N1. This NAS signalling connection is used to enable NAS signalling exchange between the UE and the core network. It comprises both the AN signalling connection between the UE and the AN (RRC Connection over 3GPP access or UE-N3IWF connection over untrusted N3GPP access or UE-TNGF connection over trusted N3GPP access) and the N2 connection for this UE between the AN and the AMF.
5.3.3.2 5GS Connection Management states
5.3.3.2.1 General
Two CM states are used to reflect the NAS signalling Connection of the UE with the AMF:
– CM-IDLE
– CM-CONNECTED
The CM state for 3GPP access and Non-3GPP access are independent of each other, i.e. one can be in CM-IDLE state at the same time when the other is in CM-CONNECTED state.
5.3.3.2.2 CM-IDLE state
A UE in CM-IDLE state has no NAS signalling connection established with the AMF over N1. The UE performs cell selection/cell reselection according to TS 38.304 [50] and PLMN selection according to TS 23.122 [17].
There are no AN signalling connection, N2 connection and N3 connections for the UE in the CM-IDLE state.
If the UE is both in CM-IDLE state and in RM-REGISTERED state, the UE shall, unless otherwise specified in clause 5.3.4.1:
– Respond to paging by performing a Service Request procedure (see clause 4.2.3.2 of TS 23.502 [3]), unless the UE is in MICO mode (see clause 5.4.1.3);
– perform a Service Request procedure when the UE has uplink signalling or user data to be sent (see clause 4.2.3.2 of TS 23.502 [3]). Specific conditions apply for LADN, see clause 5.6.5.
When the UE state in the AMF is RM-REGISTERED, UE information required for initiating communication with the UE shall be stored. The AMF shall be able to retrieve stored information required for initiating communication with the UE using the 5G-GUTI.
NOTE: In 5GS there is no need for paging using the SUPI/SUCI of the UE.
The UE provides 5G-S-TMSI as part of AN parameters during AN signalling connection establishment as specified in TS 38.331 [28] and TS 36.331 [51]. The UE shall enter CM-CONNECTED state whenever an AN signalling connection is established between the UE and the AN (entering RRC Connected state over 3GPP access, or at the establishment of the UE-N3IWF connectivity over untrusted non-3GPP access or the UE-TNGF connectivity over trusted non-3GPP access). The transmission of an Initial NAS message (Registration Request, Service Request or Deregistration Request) initiates the transition from CM-IDLE to CM-CONNECTED state.
When the UE states in the AMF are CM-IDLE and RM-REGISTERED, the AMF shall:
– perform a network triggered Service Request procedure when it has signalling or mobile-terminated data to be sent to this UE, by sending a Paging Request to this UE (see clause 4.2.3.3 of TS 23.502 [3]), if a UE is not prevented from responding e.g. due to MICO mode or Mobility Restrictions.
The AMF shall enter CM-CONNECTED state for the UE whenever an N2 connection is established for this UE between the AN and the AMF. The reception of initial N2 message (e.g. N2 INITIAL UE MESSAGE) initiates the transition of AMF from CM-IDLE to CM-CONNECTED state.
The UE and the AMF may optimize the power efficiency and signalling efficiency of the UE when in CM-IDLE state e.g. by activating MICO mode (see clause 5.4.1.3).
5.3.3.2.3 CM-CONNECTED state
A UE in CM-CONNECTED state has a NAS signalling connection with the AMF over N1. A NAS signalling connection uses an RRC Connection between the UE and the NG-RAN and an NGAP UE association between the AN and the AMF for 3GPP access. A UE can be in CM-CONNECTED state with an NGAP UE association that is not bound to any TNLA between the AN and the AMF. See clause 5.21.1.2 for details on the state of NGAP UE association for an UE in CM-CONNECTED state. Upon completion of a NAS signalling procedure, the AMF may decide to release the NAS signalling connection with the UE.
In the CM-CONNECTED state, the UE shall:
– enter CM-IDLE state whenever the AN signalling connection is released (entering RRC Idle state over 3GPP access or when the release of the UE-N3IWF connectivity over untrusted non-3GPP access or the UE-TNGF connectivity over trusted non-3GPP access is detected by the UE), see TS 38.331 [28] for 3GPP access.
When the UE CM state in the AMF is CM-CONNECTED, the AMF shall:
– enter CM-IDLE state for the UE whenever the logical NGAP signalling connection and the N3 user plane connection for this UE are released upon completion of the AN Release procedure as specified in TS 23.502 [3].
The AMF may keep a UE CM state in the AMF in CM-CONNECTED state until the UE de-registers from the core network.
A UE in CM-CONNECTED state can be in RRC Inactive state, see TS 38.300 [27]. When the UE is in RRC Inactive state the following applies:
– UE reachability is managed by the RAN, with assistance information from core network;
– UE paging is managed by the RAN.
– UE monitors for paging with UE’s CN (5G S-TMSI) and RAN identifier.
5.3.3.2.4 5GS Connection Management State models
Figure 5.3.3.2.4-1: CM state transition in UE
Figure 5.3.3.2.4-2: CM state transition in AMF
When a UE enters CM-IDLE state, the UP connection of the PDU Sessions that were active on this access are deactivated.
NOTE: The activation of UP connection of PDU Sessions is documented in clause 5.6.8.
5.3.3.2.5 CM-CONNECTED with RRC Inactive state
RRC Inactive state applies to NG-RAN. UE support for RRC Inactive state is defined in TS 38.306 [69] for NR and TS 36.306 [70] for E-UTRA connected to 5GC. RRC Inactive is not supported by NB-IoT connected to 5GC.
The AMF shall provide assistance information to the NG-RAN, to assist the NG-RAN’s decision whether the UE can be sent to RRC Inactive state except due to some exceptional cases such as:
– PLMN (or AMF set) does not support RRC Inactive;
– The UE needs to be kept in CM-CONNECTED State (e.g. for tracking).
The "RRC Inactive Assistance Information" includes:
– UE specific DRX values;
– UE specific extended idle mode DRX values (cycle length and Paging Time Window length);
– The Registration Area provided to the UE;
– Periodic Registration Update timer;
– If the AMF has enabled MICO mode for the UE, an indication that the UE is in MICO mode;
– Information from the UE identifier, as defined in TS 38.304 [50] for NR and TS 36.304 [52] for E-UTRA connected to 5GC, that allows the RAN to calculate the UE’s RAN paging occasions;
– An indication that Paging Cause Indication for Voice Service is supported;
– AMF PEIPS Assistance Information (see clause 5.4.12.2) for paging a UE in CM-CONNECTED with RRC Inactive state over NR as defined in TS 38.300 [27].
The RRC Inactive Assistance Information mentioned above is provided by the AMF during N2 activation with the (new) serving NG-RAN node (i.e. during Registration, Service Request, Handover) to assist the NG RAN’s decision whether the UE can be sent to RRC Inactive state. If the AMF allocates a new Registration Area to the UE, the AMF should update the NG-RAN with the new Registration Area by sending the RRC Inactive Assistance Information accordingly. The Paging Cause Indication for Voice Service is used to assist NG RAN to perform RAN based paging.
RRC Inactive state is part of RRC state machine, and it is up to the RAN to determine the conditions to enter RRC Inactive state. If any of the parameters included in the RRC Inactive Assistance Information changes as the result of NAS procedure, the AMF shall update the RRC Inactive Assistance Information to the NG-RAN node.
When the UE is in CM-CONNECTED state, if the AMF has provided RRC Inactive assistance information, the RAN node may decide to move a UE to CM-CONNECTED with RRC Inactive state.
The state and "endpoints" (in the case of Dual Connectivity configuration) of the N2 and N3 reference points are not changed by the UE entering CM-CONNECTED with RRC Inactive state. A UE in RRC inactive state is aware of the RAN Notification area and periodic RAN Notification Area Update timer.
The 5GC network is not aware of the UE transitions between CM-CONNECTED with RRC Connected and CM-CONNECTED with RRC Inactive state, unless the 5GC network is notified via N2 notification procedure in clause 4.8.3 of TS 23.502 [3].
At transition into CM-CONNECTED with RRC Inactive state, the NG-RAN configures the UE with a periodic RAN Notification Area Update timer taking into account the value of the Periodic Registration Update timer value indicated in the RRC Inactive Assistance Information, and uses a guard timer with a value longer than the RAN Notification Area Update timer value provided to the UE.
If the periodic RAN Notification Area Update guard timer expires in NG-RAN, the NG-RAN shall initiate AN Release procedure as specified in clause 4.2.6 of TS 23.502 [3].
When the UE is in CM-CONNECTED with RRC Inactive state, the UE performs PLMN selection procedures as defined in TS 23.122 [17] and TS 24.501 [47].
When the UE is CM-CONNECTED with RRC Inactive state, the UE may resume the RRC Connection due to:
– Uplink data pending;
– Mobile initiated NAS signalling procedure;
– As a response to RAN paging;
– Notifying the network that it has left the RAN Notification Area;
– Upon periodic RAN Notification Area Update timer expiration.
If the UE resumes the connection in a different NG-RAN node within the same PLMN or equivalent PLMN or within the same SNPN or equivalent SNPN, the UE AS context is retrieved from the old NG-RAN node and a procedure is triggered towards the CN (see clause 4.8.2 of TS 23.502 [3]).
NOTE 1: With Dual Connectivity configuration if the UE resumes the RRC connection in the Master RAN node, the Secondary RAN node configuration is defined in TS 38.300 [27].
If the RAN paging procedure, as defined in TS 38.300 [27], is not successful in establishing contact with the UE the procedure shall be handled by the network as follows:
– If NG-RAN has at least one pending NAS PDU for transmission, the RAN node shall initiate the AN Release procedure (see clause 4.2.6 of TS 23.502 [3]) to move the UE CM state in the AMF to CM-IDLE state and indicate to the AMF the NAS non-delivery.
– If NG RAN has only pending user plane data for transmission, the NG-RAN node may keep the N2 connection active or initiate the AN Release procedure (see clause 4.2.6 of TS 23.502 [3]) based on local configuration in NG-RAN.
NOTE 2: The user plane data which triggers the RAN paging can be lost, e.g. in the case of RAN paging failure.
If a UE in CM-CONNECTED with RRC Inactive state performs cell selection to GERAN/UTRAN/E-UTRAN, it shall follow idle mode procedures of the selected RAT as specified in clause 5.17.
In addition, a UE in CM-CONNECTED state with RRC Inactive state shall enter CM-IDLE state and initiates the NAS signalling recovery (see TS 24.501 [47]) in the following cases:
– If RRC resume procedure fails,
If the UE receives Core Network paging,
– If the periodic RAN Notification Area Update timer expires and the UE cannot successfully resume the RRC Connection,
– In any other failure scenario that cannot be resolved in RRC Inactive state and requires the UE to move to CM-IDLE state.
When a UE is in CM-CONNECTED with RRC Inactive state, and a trigger to change the UE’s NG-RAN or E‑UTRAN UE Radio Capability information happens, the UE shall move to CM-IDLE state and initiate the procedure for updating UE Radio Capability defined in clause 5.4.4.1. (For specific requirements for a UE operating in dual-registration mode see clause 5.17.2.1)
When UE is in CM-CONNECTED with RRC Inactive state, if RAN has received Location Reporting Control message from AMF with the Reporting Type indicating single stand-alone report or continuously reporting whenever the UE changes the cell, the RAN shall perform .location reporting as specified in clause 4.10 of TS 23.502 [3].
When the UE is CM-CONNECTED with RRC Inactive state. If the AMF receives Nudm_UECM_DeregistrationNotification from UDM, the AMF shall initiate AN Release procedure as specified in clause 4.2.6 of TS 23.502 [3].
When UE is in CM-CONNECTED with RRC Inactive state, if RAN has received Location Reporting Control message from AMF with the Reporting Type of the Area Of Interest based reporting, the RAN shall send a Location Report message to AMF including UE presence in the Area Of Interest (i.e. IN, OUT, or UNKNOWN) and the UE’s last known location with time stamp.
When the UE is in CM-CONNECTED with RRC Inactive state, if the old NG-RAN node that sents the UE into RRC Inactive state receives the downlink N2 signalling, it initiates the RAN paging as defined in TS 38.300 [27]. If the UE resumes the RRC Connection towards a different NG-RAN node, the old NG-RAN node includes the "UE Context Transfer" indication into a response container to the NF (e.g. AMF or SMF) that generates such N2 downlink signalling. Then the NF shall reattempt the same procedure when the path switch from the old NG-RAN node to the new NG-RAN node is complete.
5.3.3.3 NAS signalling connection management
5.3.3.3.1 General
NAS signalling connection management includes the functions of establishing and releasing a NAS signalling connection.
5.3.3.3.2 NAS signalling connection establishment
NAS signalling connection establishment function is provided by the UE and the AMF to establish a NAS signalling connection for a UE in CM-IDLE state. The AMF shall provide the list of recommended cells/ TAs / NG-RAN node identifiers for paging, if the NG-RAN had provided that information in an earlier AN Release Procedure in the AN (see clause 4.2.6 of TS 23.502 [3]).
When the UE in CM-IDLE state needs to transmit an NAS message, the UE shall initiate a Service Request, a Registration or a Deregistration procedure to establish a NAS signalling connection to the AMF as specified in clauses 4.2.2 and 4.2.3 of TS 23.502 [3]. If the NAS signalling connection is to be established via an NG-RAN node, but the AMF detects that this UE has already established a NAS signalling connection via old NG-RAN node, the AMF shall release the old established NAS signalling connection by triggering AN Release Procedure.
Based on UE preferences, UE subscription, Mobility Pattern and network configuration, the AMF may keep the NAS signalling connection until the UE de-registers from the network.
5.3.3.3.3 NAS signalling connection Release
The procedure of releasing a NAS signalling connection is initiated by the AN node (either 5G (R)AN node or N3IWF) or the AMF. The NG-RAN node may include the list of recommended cells/ TAs / NG-RAN node identifiers for paging, during the AN Release Procedure in the AN (see clause 4.2.6 of TS 23.502 [3]). The AMF stores this information, if provided by the NG-RAN.
The UE considers the NAS signalling connection is released if it detects the AN signalling connection is released. The AMF considers the NAS signalling connection is released if it detects the N2 context is released.
5.3.3.4 Support of a UE connected over both 3GPP and Non-3GPP access
The AMF manages two CM states for an UE: a CM state for 3GPP access and a CM state for Non-3GPP access. An N2 interface can serve the UE for either 3GPP access or for Non 3GPP access. UE connected over both 3GPP and Non-3GPP has got two N2 interfaces, one for each access. A UE may be in any combination of the CM states between 3GPP and Non-3GPP access, e.g. a UE may be CM-IDLE for one access and CM-CONNECTED for the other access, CM-IDLE for both accesses or CM-CONNECTED for both accesses.
When the UE CM state in the AMF is CM-IDLE for 3GPP access and CM-CONNECTED for Non-3GPP access, the AMF shall perform a network triggered Service Request procedure, when it has downlink data to be sent to this UE for 3GPP access, by sending either the Paging Request via 3GPP access or the NAS notification via Non-3GPP access to this UE (see clause 4.2.3.3 of TS 23.502 [3]).
Connection Management over Non-3GPP access is further defined in clause 5.5.2.
5.3.4 UE Mobility
5.3.4.1 Mobility Restrictions
5.3.4.1.1 General
Mobility Restrictions restrict mobility handling or service access of a UE. The Mobility Restriction functionality is provided by the UE (only for mobility restriction categories provided to the UE), the radio access network and the core network.
Unless otherwise stated, Mobility Restrictions only apply to 3GPP access and wireline access, they do not apply to other non-3GPP accesses.
The UE and the network shall override Mobility restriction as specified in clause 5.16.4.3 when accessing the network for Emergency Services. For MPS and MCX, service area restriction does not apply, as specified in TS 24.501 [47].
For UE requesting Disaster Roaming service, the UE is only allowed to receive services in the area with Disaster Condition as specified in clause 5.40.4. The other areas within the PLMN shall be considered as forbidden area for the UE registered for Disaster Roaming service.
Service Area restrictions and handling of Forbidden Areas for CM-IDLE state and, for CM-CONNECTED state when in RRC Inactive state are executed by the UE based on information received from the core network. Mobility Restrictions for CM-CONNECTED state when in RRC-Connected state are executed by the radio access network and the core network.
In CM-CONNECTED state, the core network provides Mobility Restrictions to the radio access network within Mobility Restriction List.
Mobility Restrictions consists of RAT restriction, Forbidden Area, Service Area Restrictions, Core Network type restriction and Closed Access Group information as follows:
– RAT restriction:
Defines the 3GPP Radio Access Technology(ies), a UE is not allowed to access in a PLMN. In a restricted RAT a UE based on subscription is not permitted access to the network for this PLMN. For CM-CONNECTED state, when radio access network determines target RAT and target PLMN during Handover procedure, it should take per PLMN RAT restriction into consideration. The RAT restriction is enforced in the network, and not provided to the UE.
– Forbidden Area:
In a Forbidden Area, the UE, based on subscription, is not permitted to initiate any communication with the network for this PLMN. The UE behaviour in terms of cell selection, RAT selection and PLMN selection depends on the network response that informs the UE of Forbidden Area. A Forbidden Area applies either to 3GPP access or to non-3GPP access.
Further description on Forbidden Area when using wireline access is available in TS 23.316 [84].
Support for Forbidden Area with NR satellite access is described in clause 5.4.11.8.
Forbidden Areas should not be used for Untrusted or Trusted non-3GPP access.
NOTE 1: If a UE receives that the UE is accessing from a forbidden tracking area when registering over untrusted non-3GPP access or trusted non-3GPP access, the UE cannot determine the corresponding TAI and thus needs to consider that access to untrusted non-3GPP access and to trusted non-3GPP access in this PLMN is forbidden until the forbidden area list is removed as described in TS 24.501 [47].
NOTE 2: The UE reactions to specific network responses are described in TS 24.501 [47].
– Service Area Restriction:
Defines areas in which the UE may or may not initiate communication with the network as follows:
– Allowed Area:
In an Allowed Area, the UE is permitted to initiate communication with the network as allowed by the subscription.
– Non-Allowed Area:
In a Non-Allowed Area a UE is service area restricted based on subscription. The UE and the network are not allowed to initiate Service Request, or any connection requests for user plane data, control plane data, exception data reporting, or SM signalling (except for PS Data Off status change reporting) to obtain user services that are not related to mobility.
The UE shall not use the entering of a Non-Allowed Area as a criterion for Cell Reselection, a trigger for PLMN Selection or Domain selection for UE originating sessions or calls. The RRC procedures while the UE is in CM-CONNECTED with RRC Inactive state are unchanged compared to when the UE is in an Allowed Area. The RM procedures are unchanged compared to when the UE is in an Allowed Area. The UE in a Non-Allowed Area shall respond to core network paging or NAS Notification message from non-3GPP access with Service Request and RAN paging. The UE in a Non-Allowed Area may initiate MA PDU Session establishment or activation over a non-3GPP access other than wireline access, but the User Plane resources on the 3GPP access for the MA-PDU shall not be established or activated. The handling of Non-Allowed Area when using wireline access is described in TS 23.316 [84].
NOTE 3: When the services are restricted in 5GS due to Service Area Restriction, then it is assumed that the services will be also restricted in all RATs/Systems at the same location(s) using appropriate mechanisms available in the other RATs/Systems.
NOTE 4: Delivery of SOR transparent container, UE policy container, UE parameters update transparent container as defined in TS 24.501 [47], or removal of any stored Paging Restriction Information from network via Registration Request (see clause 5.38), is part of the mobility related service and is allowed in an area with service restriction.
NOTE 5: For a UE in CM-CONNECTED state then neither control plane data transmission nor, if user plane resources are already established, user plane data transmission are restricted by a non-allowed area.
– Core Network type restriction:
Defines whether UE is allowed to connect to 5GC only, EPC only, both 5GC and EPC for this PLMN. The Core Network type restriction when received applies in the PLMN either to both 3GPP and non-3GPP Access Types or to non-3GPP Access Type only.
NOTE 6: The Core Network type restriction can be used e.g. in network deployments where the E-UTRAN connects to both EPC and 5GC as described in clause 5.17. When the Core Network type restriction applies to non-3GPP Access Type, the UE is restricted from using any connectivity to an N3IWF.
– Closed Access Group information:
As defined in clause 5.30.3.
For a given UE, the core network determines the Mobility Restrictions based on UE subscription information, UE location and/or local policy (e.g. if the HPLMN has not deployed 5GC, HPLMN ID of the UE and the operator’s policy are used in the VPLMN for determining the Core Network type restriction). The Mobility Restriction may change due to e.g. UE’s subscription, location change and local policy. Optionally the Service Area Restrictions or the Non-Allowed Area may in addition be fine-tuned by the PCF e.g. based on UE location, PEI and network policies. Service Area Restrictions may be updated during a Registration procedure or UE Configuration Update procedure.
NOTE 7: The subscription management ensures that for MPS service subscriber the Mobility Restrictions is not included.
If the network sends Service Area Restrictions to the UE, the network sends only either an Allowed Area, or a Non-Allowed Area, but not both at the same time, to the UE. If the UE has received an Allowed Area from the network, any TA not part of the Allowed Area is considered by the UE as non-allowed. If the UE has received a Non-Allowed Area from the network, any TA not part of the Non-Allowed Area is considered by the UE as allowed. If the UE has not received any Service Area Restrictions, any TA in the PLMN is considered as allowed.
If the UE has overlapping areas between Forbidden Areas, Service Area Restrictions, or any combination of them, the UE shall proceed in the following precedence order:
– The evaluation of Forbidden Areas shall take precedence over the evaluation of Service Area Restrictions.
The UDM shall provide to the AMF the information defined in TS 23.008 [119] about the subscriber’s NR or E-UTRA access restriction set by the operator determined e.g. by subscription scenario and roaming scenario:
– For NR:
– NR not allowed as primary access.
– NR not allowed as secondary access.
– NR in unlicensed bands not allowed as primary access.
– NR in unlicensed bands not allowed as secondary access.
– NR(LEO) satellite access not allowed as primary access.
– NR(MEO) satellite access not allowed as primary access.
– NR(GEO) satellite access not allowed as primary access.
– NR(OTHERSAT) satellite access not allowed as primary access.
– NR RedCap not allowed as primary access.
Editor’s note: The final decision on whether Dual Connectivity and Carrier Aggregation are supported is in RAN WG2 and RAN WG3. Subsequently SA WG2 will consider whether any system features are needed to support such DC or CA RAN functionality such as mobility restrictions for secondary access.
– For E-UTRA:
– E-UTRA not allowed as primary access.
– E-UTRA not allowed as secondary access.
– E-UTRA in unlicensed bands not allowed as secondary access.
– NB-IoT not allowed as primary access.
– LTE-M not allowed as primary access.
In order to enforce all primary access restrictions, the related access has to be deployed in different Tracking Area Codes and the subscriber shall not be allowed to access the network in TAs using the particular access.
With all secondary access restrictions, the subscriber shall not be allowed to use this access as secondary access.
5.3.4.1.2 Management of Service Area Restrictions
This clause describes Service Area Restrictions for 3GPP access. For Service Area Restrictions when using wireline access, see TS 23.316 [84].
A Service Area Restriction may contain one or more (e.g. up to 16) entire Tracking Areas each or the Service Area Restriction may be set as unlimited (i.e. contain all Tracking Areas of the PLMN). The UE’s subscription data in the UDM includes a Service Area Restriction which may contain either Allowed or Non-Allowed Areas–specified by using explicit Tracking Area identities and/or other geographical information (e.g. longitude/latitude, zip code, etc). The geographical information used to specify Allowed or Non-Allowed Area is only managed in the network, and the network will map it to a list of TAs before sending Service Area Restriction information to the PCF, NG-RAN and UE.
When the AMF assigns a limited allowed area to the UE, the AMF shall provide the UE with Service Area Restrictions which consist of either Allowed Areas or Non-Allowed Areas. The Allowed Areas included in the Service Area Restrictions can be pre-configured and/or dynamically assigned by the AMF.
The Allowed Area may alternatively be configured as unlimited i.e. it may contain all Tracking Areas of the PLMN. The Registration Area of a UE in the Non-Allowed Area should consist of a set of TAs which belongs to a Non-Allowed Area of the UE. The Registration Area of a UE in the Allowed Area should consist of a set of TAs which belongs to an Allowed Area of the UE. The AMF provides the Service Area Restriction in the form of TA(s), which may be a subset of full list stored in UE’s subscription data or provided by the PCF, to the UE during the Registration procedure.
NOTE: As the finest granularity for Service Area Restrictions are at TA level, subscriptions with limited geographical extent, like subscriptions for Fixed Wireless Access, will be allocated one or a few TAs and will consequently be allowed to access services in a larger area than in e.g. a FWA system.
The limited allowed area may also be limited by the AMF by a maximum allowed number of Tracking Areas, even though this limitation is not sent to the UE. If maximum allowed number of Tracking Areas is used in combination with Allowed Area, the maximum allowed number of Tracking Areas indicates (to the AMF) the maximum number of TAs allowed in limited allowed area inside the Allowed Area. If maximum allowed number of Tracking Areas is used in combination with Non-Allowed Area, the maximum allowed number of Tracking Areas indicates (to the AMF) the maximum number of TAs allowed in limited allowed area outside of the Non-Allowed Area.
The UDM stores the Service Area Restrictions of a UE as part of the UE’s subscription data. The PCF in the serving network may (e.g. due to varying conditions such as UE’s location, application in use, time and date) further adjust Service Area Restrictions of a UE, either by expanding an Allowed Area or by reducing a Non-Allowed Area or by increasing the maximum allowed number of Tracking Areas. If NWDAF is deployed, the PCF may use analytics (i.e. statistics or predictions) on UE mobility from NWDAF (see TS 23.288 [86]) to adjust Service Area Restrictions. The UDM and the PCF may update the Service Area Restrictions of a UE at any time. For the UE in CM-CONNECTED state the AMF updates the UE and RAN immediately. For UE in CM-IDLE state the AMF may page the UE immediately or store the updated service area restriction and update the UE upon next signalling interaction with the UE, as defined in TS 24.501 [47].
During registration, if the Service Area Restrictions of the UE is not present in the AMF, the AMF fetches from the UDM the Service Area Restrictions of the UE that may be further adjusted by the PCF. The serving AMF shall enforce the Service Area Restrictions of a UE. A limited allowed area given by a maximum allowed number of Tracking Areas, may be dynamically assigned by the AMF adding any not yet visited (by the UE) Tracking Areas to the limited allowed area until the maximum allowed number of Tracking Areas is reached (i.e. the AMF adds new TAs to the limited allowed area until the number of TAs is equal to the maximum allowed number of Tracking Areas). The AMF deletes the list of TAs that have been used up under the maximum allowed number of Tracking Areas quota at every Initial Registration.
For a UE in CM-CONNECTED state the AMF shall indicate the Service Area Restrictions of this UE to the RAN, using a Mobility Restriction List.
The UE shall store the received Service Area Restrictions and, if there is previously stored Service Area Restrictions, replace them with the newly received information. If the Service Area Restrictions include a limited allowed area, the Service Area Restrictions are applicable for the Tracking areas indicated in Service Area Restrictions. If the Service Area Restrictions included an unlimited allowed area, the received Service Area Restrictions are either applicable for the registered PLMN and its equivalent PLMN(s) that are available in the Registration Area, or the registered SNPN that is available in the Registration Area. The RAN uses the Service Area Restrictions for target cell selection in Xn and N2 based handover.
Upon change of serving AMF due to mobility, the old AMF may provide the new AMF with the Service Area Restrictions of the UE that may be further adjusted by the PCF.
The network may perform paging for a UE to update Service Area Restrictions with Generic UE Configuration Update procedure (see clause 4.2.4 of TS 23.502 [3]).
In the case of roaming, the Service Area Restrictions are transferred from the UDM via the serving AMF to the serving PCF in the visited network. The serving PCF in the visited network may further adjust the Service Area Restrictions.
Support for Service Area Restrictions with NR satellite access is described in clause 5.4.11.8.
5.3.4.2 Mobility Pattern
The Mobility Pattern is a concept that may be used by the AMF to characterise and optimise the UE mobility. The AMF determines and updates Mobility Pattern of the UE based on subscription of the UE, statistics of the UE mobility, network local policy, and the UE assisted information, or any combination of them. The statistics of the UE mobility can be historical or expected UE moving trajectory. If NWDAF is deployed, the statistics of the UE mobility can also be analytics (i.e. statistics or predictions) provided by the NWDAF (see TS 23.288 [86]).
The Mobility Pattern can be used by the AMF to optimize mobility support provided to the UE, for example, Registration area allocation.
5.3.4.3 Radio Resource Management functions
5.3.4.3.1 General
To support radio resource management in NG-RAN the AMF provides the parameter ‘Index to RAT/Frequency Selection Priority’ (RFSP Index) to NG-RAN across N2. The RFSP Index is mapped by the RAN to locally defined configuration in order to apply specific RRM strategies, taking into account any available information in RAN. The RFSP Index is UE specific and applies to all the Radio Bearers. Examples of how this parameter may be used by the RAN:
– to derive UE specific cell reselection priorities to control idle mode camping.
– to decide on redirecting active mode UEs to different frequency layers or RATs (e.g. see clause 5.3.4.3.2).
The HPLMN may set the RFSP Index taking into account the Subscribed S-NSSAIs. The AMF receives the subscribed RFSP Index from the UDM (e.g. during the Registration procedure). For non-roaming subscribers, the AMF chooses the RFSP Index in use according to one of the following procedures, depending on operator’s configuration:
– the RFSP Index in use is identical to the subscribed RFSP Index, or
– the AMF chooses the RFSP Index in use based on the subscribed RFSP Index, the locally configured operator’s policies, the Allowed NSSAI and the UE related context information available at the AMF, including UE’s usage setting, if received during Registration procedures (see clause TS 23.502 [3]).
NOTE 1: One example of how the AMF can use the "UE’s usage setting," is to select an RFSP value that enforces idle mode camping on E-UTRA for a UE acting in a "Voice centric" way, in the case voice over NR is not supported in the specific Registration Area and it contains NR cells.
The AMF may report to the PCF the subscribed RFSP Index received from the UDM for further evaluation as described in clause 6.1.2.1 of TS 23.503 [45]. When receiving the authorized RFSP Index from the PCF, the AMF shall apply the authorized RFSP Index instead of the subscribed RFSP Index for choosing the RFSP index in use (as described above). For roaming subscribers, the AMF may choose the RFSP Index in use based on the visited network policy, but can take input from the HPLMN into account (e.g. an RFSP Index value pre-configured per HPLMN, or a single RFSP Index value to be used for all roamers independent of the HPLMN).
NOTE 2: The PCF can provide validity time together with the authorized RFSP Index indicating a change in priority from 5G access to E-UTRAN access as specified in clause 5.17.2.2.
The RFSP Index in use is also forwarded from source to target NG-RAN node when Xn or N2 is used for intra-NG-RAN handover.
The AMF stores the subscribed RFSP Index value received and the RFSP Index value in use. During the Registration procedure, the AMF may update the RFSP Index value in use (e.g. the AMF may need to update the RFSP Index value in use if the UE related context information in the AMF has changed). When the RFSP Index value in use is changed, the AMF immediately provides the updated RFSP Index value in use to NG-RAN node by modifying an existing UE context or by establishing a new UE context in RAN or by being configured to include the updated RFSP Index value in use in the NGAP DOWNLINK NAS TRANSPORT message if the user plane establishment is not needed. During inter-AMF mobility procedures, the source AMF forwards both RFSP Index values to the target AMF. The target AMF may replace the received RFSP Index value in use with a new RFSP Index value in use that is based on the operator’s policies and the UE related context information available at the target AMF.
In order to enable UE idle mode mobility control and priority-based reselection mechanism considering availability of Network Slices at the network and the Network Slices allowed for a UE, an RFSP is derived as described in clause 5.3.4.3, considering also the Allowed NSSAI for the UE.
A UE supporting NSAG (see clause 5.15.14) may be configured, for some of the S-NSSAIs in the configured NSSAI, with NSAGs it can use as described in TS 38.300 [27], TS 38.304 [50], TS 38.331 [28], TS 38.321 [143], TS 24.501 [47] and as described in clause 5.3.4.3.4.
5.3.4.3.2 Preferred band(s) per data radio bearer(s)
The NG-RAN may prefer to use specific radio resources per data radio bearer(s), e.g. depending on the Network Slices associated to the data radio bearer used by the UE. The UE idle mode mobility control and priority-based reselection mechanism operates as described in clause 5.3.4.3.1, and when UP resources are activated e.g. for a specific S-NSSAI the NG-RAN can use local policies to decide on what specific radio resources to use for the associated data radio bearer(s). A UE may be served by a set of data radio bearers which may be served by cells in different bands, selected based on RRM policies.
5.3.4.3.3 Redirection to dedicated frequency band(s) for an S-NSSAI
If a Network Slice, S-NSSAI, is configured to be available only in TAs covering specific dedicated frequency band(s), then there may be a need to redirect the UE to the dedicated frequency band(s) when such S-NSSAI is requested. If the Requested NSSAI contains S-NSSAI(s) that are not available in the UE’s current TA, see clause 5.15.8, the AMF itself or by interacting with the NSSF as described in clause 5.15.5.2.1 may determine a Target NSSAI to be used by the NG-RAN, in addition to the information the AMF receives, such as the Allowed NSSAI and the RFSP for the Allowed NSSAI, to attempt to redirect the UE to a cell and TA in another frequency band and TA that supports the S-NSSAIs in the Target NSSAI. The Target NSSAI includes at least one S-NSSAI from the Requested NSSAI not available in the current TA, but available in another TA in different frequency band possibly overlapping with the current TA, and optionally additional S-NSSAIs from the Requested NSSAI that are configured to be available within the same TAs as the S-NSSAIs not available in the current TA. If the serving PLMN supports the subscription-based restrictions to simultaneous registration of network slices (see clause 5.15.12), and if the UE has NSSRG as part of the subscription information received from the HPLMN, the Target NSSAI includes only S-NSSAIs sharing at least one NSSRG.
The Target NSSAI may be excluding some of the Allowed NSSAIs and include some of the rejected S-NSSAIs due to lack of support in the TA where the UE is located based on network policies that are in line with customer and operator agreements. The Target NSSAI shall only include S-NSSAIs that can be provided in an Allowed NSSAI for the UE. The Target NSSAI may include e.g.:
– all or a subset of the Rejected S-NSSAIs for RA when none of the S-NSSAIs in the Requested S-NSSAI were available in the TA where the UE is;
– all the S-NSSAIs of the Allowed NSSAI and all or a subset of the Rejected S-NSSAIs for the RA;
– a subset of the S-NSSAIs in the Allowed NSSAI and all or a subset of the Rejected S-NSSAIs for the RA if the operator policy is to prefer this Target S-NSSAI to the Allowed NSSAI.
The AMF should retrieve an RFSP Index suitable for the Target NSSAI and includes the RFSP Index in the information sent to the NG-RAN. The AMF retrieves the RFSP Index from the PCF or, in case PCF is not deployed the AMF determines the RFSP Index according to local configuration. The RFSP index associated to the Target NSSAI is considered if the RAN succeeds to redirect the UE to a new TA outside the RA, otherwise the RFSP index of the Allowed NSSAI is considered.
If the Requested NSSAI contains S-NSSAI(s) which map to S-NSSAI(s) of the HPLMN subject to Network Slice-Specific Authentication and Authorization that are not available in the UE’s current TA, the AMF shall proceed with the Network Slice-Specific Authentication and Authorization procedure as described in clause 4.2.9 of TS 23.502 [3]. If the AMF determines a new Allowed NSSAI at the end of Network Slice-Specific Authentication and Authorization steps and some S-NSSAI is not available in the UE’s current TA, a Target NSSAI and corresponding RFSP index may be determined and provided to NG-RAN during UE Configuration Update procedure as described in clause 4.2.4.2 of TS 23.502 [3].
The NG-RAN shall attempt to find cells of TAs that can support all the S-NSSAIs in the Target S-NSSAIs, and if no such cell of a TA is available the RAN can attempt to select cells of TAs that best match the Target S-NSSAI. The NG-RAN shall attempt to ensure continuity of the PDU Sessions with activated User Plane associated with the S-NSSAIs in the Allowed NSSAI which are in the Target NSSAI. Also, the NG-RAN should attempt to ensure continuity of service for the S-NSSAIs of the Allowed NSSAI also available in the Target NSSAI, before prioritizing cells that are not supporting one or more of the S-NSSAI of the Allowed NSSAI also available in the Target NSSAI.
The NG-RAN attempts to determine target cell(s) supporting the Target NSSAI considering the UE Radio Capabilities (i.e. the AMF (if available in the UE context) shall provide the NG-RAN with the current UE Radio Capability Information or the RACS UE Radio Capability ID when a Target NSSAI is provided, if the NG-RAN had not yet received any of them, or, if the AMF cannot provide any of these, the UE Radio Capability Information may be retrieved by the NG-RAN from the UE).
Once the target cells are determined, the NG-RAN initiates RRC redirection procedure towards the target cells, or the NG-RAN initiates handover for the UE with active PDU Sessions associated with the S-NSSAIs which are in the Target NSSAI, if possible.
After a successful redirection or handover of the UE to a new TA outside the current RA, the UE shall perform a Mobility Registration Update procedure and the S-NSSAIs the new TA supports can be allowed if the UE requests them. In order to ensure that the UE is redirected to a TA outside the current RA, thus triggering a Mobility Registration Update procedure enabling the UE to request the S-NSSAI(s) that were rejected for the RA, the AMF shall set the RA so that the RA does not include TAs supporting the S-NSSAIs rejected for the RA included in the Target NSSAI when the AMF provides a Target NSSAI to the RAN.
5.3.4.3.4 Network Slice based cell reselection and Random Access
When one or more S-NSSAI(s) are associated with NSAG(s), the UE may perform Network Slice based cell reselection and Random Access as described in TS 38.300 [27], TS 38.304 [50], TS 38.331 [28], TS 38.321 [143] and TS 24.501 [47].
When providing NSAG Information to the UE, the AMF shall also provide the NSAG priority information for the NSAGs provided in the NSAG Information. The AMF determines the NSAG priority information based on configured local policy of the serving PLMN or SNPN.
NOTE 1: How the AMF assigns the NSAG priority information per UE is not specified but AMF can take into account information like e.g. UE MM capabilities, Subscribed S-NSSAIs and HPLMN.
NOTE 2: The AMF can assign same priority value for NSAGs provided in the NSAG Information.
If the UE has received NSAG Information from the AMF, the UE shall use the NSAG Information provided by the AMF for cell reselection and Random Access as described below. If the UE has not received any NSAG Information from the AMF, the UE shall not use Network Slice based cell reselection and Random Access at all.
The UE NAS provides to the UE AS the NSAG Information as received from the AMF and the S-NSSAIs in the Allowed NSSAI as input to cell reselection, except when the UE intends to register with a new set of S-NSSAIs with a Requested NSSAI different from the current Allowed NSSAI, in which case the UE NAS provides to the UE AS layer the NSAG Information as received from the AMF and the S-NSSAIs in the Requested NSSAI, and this may trigger a cell reselection, before sending the Registration Request including the new Requested NSSAI.
For Network Slice based Random Access, different Random Access resources may be assigned to different NSAG(s). The UE determines Random Access configuration among NSAGs that are published in SIB for Random Access and that are associated to the S-NSSAIs triggering the access. If the signalling transaction triggering the access attempt is related to more than one network slice, and the S-NSSAIs of these network slices are associated with more than one NSAG for Random Access, the NSAG with the highest priority is selected.
NOTE 3: How the UE NAS provides the NSAGs priorities to UE AS is based internal UE interface, and not specified.
5.3.4.4 UE mobility event notification
5G System supports the functionality of tracking and reporting UE mobility events.
The AMF provides the UE mobility related event reporting to NF that has been authorized to subscribe to the UE mobility event reporting service. Any NF service consumer such as SMF, NEF or NWDAF that wants to be reported on the UE location is able to subscribe to the UE mobility event notification service to the AMF with the following parameters:
– Event reporting type that specifies what to be reported on UE mobility (e.g. UE location, UE mobility on Area of Interest).
– Event filters indicating the:
– Area Of Interest that specifies a location area within 3GPP system. The Area Of Interest is represented by a list of Tracking Areas, list of cells or list of (R)AN node identifiers. In the case of LADN, the event consumer (e.g. SMF) provides the LADN DNN to refer the LADN service area as the Area Of Interest. In the case of PRA, the event consumer (e.g. SMF or PCF) may provide an identifier for Area Of Interest to refer predefined area as the Area Of Interest.
– S-NSSAI and optionally the NSI ID(s).
– Event Reporting Information: event reporting mode, number of reports, maximum duration of reporting, event reporting condition (e.g. when the target UE moved into a specified Area Of Interest, immediate reporting flag).
– Notification Endpoint of NF service consumer to be notified.
– The target of event reporting that indicates a specific UE, a group of UE(s) or any UE (i.e. all UEs). Further details on the information provided by the NF service consumer are provided in clause 4.15 of TS 23.502 [3].
If an NF service consumer subscribes to the UE mobility event notification service provided by AMF for reporting of UE presence in Area Of Interest, the AMF tracks UE’s location considering UE’s CM state and using NG-RAN procedures (if RRC Inactive state applies to NG-RAN) in order to determine the UE presence in the Area Of Interest, as described in clause 4.15.4.2 of TS 23.502 [3]. Upon detecting the change of the UE presence in the Area Of Interest, the AMF notifies the UE presence in the Area Of Interest and the new UE location to the subscribed NF consumer.
When the AMF is changed, the subscription of mobility event is transferred from the old AMF. The new AMF may decide not to notify the SMF with the current status related to the subscription of mobility event if the new AMF determines that, based on MM Context of the UE, the event is reported by the old AMF.
In the network deployment where a UE may leave or enter the Area Of Interest without any notification to the 5GC in CM-CONNECTED state (i.e. in the case that RRC Inactive state applies to the NG-RAN), the AMF may initiate the NG-RAN location reporting as described in clause 5.4.7 or N2 Notification as described in clause 4.8.3 of TS 23.502 [3] to track the UE presence in the Area Of Interest.
The AMF may provide UE mobility event reporting to PCF, using Policy Control Report Triggers defined in TS 23.503 [45].
5.3.5 Triggers for network analytics
Triggers for the AMF to request for or subscribe to the analytics information from the NWDAF are internal logic in the AMF and may include for example:
– UE access and mobility related event subscription by other NFs (e.g. SMF, NEF);
– locally detected events;
– analytics information received.
The trigger conditions may depend on operator and implementation policy in the AMF. When a trigger condition happens, the AMF may decide if any analytics information is needed and if so, request for or subscription to the analytics information from the NWDAF.
The AMF may, upon detection of certain local events, e.g. frequent mobility re-registration of one or more UEs, subscribe to mobility related abnormal behaviour analytics of the UE(s) as described in TS 23.288 [86] in order to trace UE mobility trend and take appropriate actions.