Content for TS 23.060 Word version: 18.0.0

1… 5… 5.3.8… 5.4… 5.4.2… 5.4.9… 5.6… 5.6.2 5.6.3… 5.6.3.7… 5.7… 6… 6.3… 6.5… 6.6… 6.8… 6.9… 6.9.1.3 6.9.2… 6.9.2.2… 6.9.2.2.2 6.9.2.2.3… 6.9.2.2.5… 6.9.3… 6.10… 6.12… 6.13… 6.13.1.2… 6.13.2… 6.13.2.2 6.14… 8… 8.2 9… 9.2.2… 9.2.2.2 9.2.2.3… 9.2.3… 9.2.3.2… 9.2.3.3… 9.2.4… 9.2.4.2… 9.2.5… 12… 12.5… 12.6… 12.7… 12.8… 13… 14… 15… 15.3… 16… 16.2… A… B…

5.1 GPRS Access Interfaces and Reference Points 5.2 Network Interworking 5.3 High-Level Functions 5.3.0 General 5.3.1 Network Access Control Functions 5.3.2 Packet Routeing and Transfer Functions 5.3.3 Mobility Management Functions 5.3.4 Logical Link Management Functions (A/Gb mode) 5.3.5 Radio Resource Management Functions 5.3.6 Network Management Functions 5.3.7 Selection functions
...

5 General GPRS Architecture and Transmission Mechanism p. 24

5.1 GPRS Access Interfaces and Reference Points p. 24

Each PLMN has two access points to GPRS services, the radio interface (labelled Um in A/Gb mode and Uu in Iu mode) used for mobile access and the R reference point used for origination or reception of messages. The R reference point for the MSs is defined in TS 27.060.

An interface differs from a reference point in that an interface is defined where specific information is exchanged and needs to be fully recognised.

There is an inter PLMN interface called Gp or S8, respectively that connects two independent GPRS packet domain networks for message exchange.

There is also a PLMN to packet data network reference point called Gi or SGi, respectively. Gi and SGi are defined in TS 29.061.

Reproduction of 3GPP TS 23.060, Fig. 1: GPRS Access Interfaces and Reference Points

Figure 1: GPRS Access Interfaces and Reference Points

There may be more than a single network interface to several different packet data networks. These networks may both differ in ownership as well as in communications protocol (e.g. TCP/IP etc.). The network operator defines and negotiates interconnection with each interconnected packet data network.

5.2 Network Interworking p. 25

Network interworking is required whenever a packet domain PLMN and any other network are involved in the execution of a service request. With reference to Figure 1, interworking takes place through the Gi or SGi reference point and the Gp or S8 interface.

The internal mechanism for conveying the PDP PDU through the PLMN is managed by the PLMN network operator and is not apparent to the data user. The use of the GPRS service may have an impact on and increase the transfer time normally found for a message when communicated through a fixed packet data network.

5.2.1 Internet (IP) Interworking p. 25

GPRS shall support interworking with networks based on the Internet protocol (IP). IP is defined in RFC 791. The packet domain may provide compression of the TCP/IP header when an IP datagram is used within the context of a TCP connection.

Mobile terminals offered service by a service provider may be globally addressable through the network operator's addressing scheme.

5.3 High-Level Functions p. 25

5.3.0 General |R8| p. 25

The following list gives the logical functions performed within the packet domain network for GPRS with GERAN or UTRAN accesses. Several functional groupings (meta functions) are defined and each encompasses a number of individual functions:

Network Access Control Functions.
Packet Routeing and Transfer Functions.
Mobility Management Functions.
Logical Link Management Functions (A/Gb mode).
Radio Resource Management Functions.
Network Management Functions.
UE reachability function

5.3.1 Network Access Control Functions p. 26

Network access is the means by which a user is connected to a telecommunication network in order to use the services and/or facilities of that network. An access protocol is a defined set of procedures that enables the user to employ the services and/or facilities of the network.

User network access may occur from either the mobile side or the fixed side of the network. The fixed network interface may support multiple access protocols to packet data networks, for example IP. The set of access protocols to be supported is determined by the PLMN operator.

Individual PLMN administrations may require specific access-control procedures in order to limit the set of users permitted to access the network, or to restrict the capabilities of individual users, for example by limiting the type of service available to an individual subscriber. Such access control procedures are beyond the scope of the specifications.

5.3.1.1 Registration Function p. 26

Registration is the means by which a user's Mobile Id is associated with the user's packet data protocol(s) and address(es) within the PLMN, and with the user's access point(s) to the packet data network. The association can be static, i.e. stored in an HLR, or dynamic, i.e. allocated on a per need basis.

5.3.1.2 Authentication and Authorisation Function p. 26

This function performs the identification and authentication of the service requester, and the validation of the service request type to ensure that the user is authorised to use the particular network services. The authentication function is performed in association with the Mobility Management functions.

5.3.1.3 Admission Control Function p. 26

The purpose of admission control is to calculate which network resources are required to provide the quality of service (QoS) requested, determine if those resources are available, and then reserve those resources. Admission control is performed in association with the Radio Resource Management functions in order to estimate the radio resource requirements within each cell.

5.3.1.4 Message Screening Function p. 26

A screening function concerned with filtering out unauthorised or unsolicited messages is required. This should be supported through packet filtering functions. All types of message screening are left to the operators' control, e.g. by use of Internet firewalls.

5.3.1.5 Packet Terminal Adaptation Function p. 26

This function adapts data packets received / transmitted from/to terminal equipment to a form suitable for transmission by GPRS across the packet domain network.

5.3.1.6 Charging Data Collection Function p. 26

This function collects data necessary to support subscription and/or traffic fees.

5.3.1.7 Operator Determined Barring Function |R4| p. 26

The purpose of this function is to limit the service provider's financial risk with respect to new subscribers or to those who have not promptly paid their bills by restricting a particular packet switched service.

The functionality of ODB is described in the TS 23.015.

5.3.2 Packet Routeing and Transfer Functions p. 27

A route is an ordered list of nodes used for the transfer of messages within and between the PLMN(s). Each route consists of the originating node, zero or more relay nodes and the destination node. Routeing is the process of determining and using, in accordance with a set of rules, the route for transmission of a message within and between the PLMN(s).

5.3.2.1 Relay Function p. 27

The relay function is the means by which a node forwards data received from one node to the next node in the route.

5.3.2.2 Routeing Function p. 27

The routeing function determines the core network node to which a message should be forwarded and the underlying service(s) used to reach that GPRS Support Node (GSN), S-GW or P-GW, using the destination address of the message. The routeing function selects the transmission path for the "next hop" in the route.

Data transmission between core network nodes may occur across packet data networks that provide their own internal routeing functions, for example ITU-T Recommendation X.25 [34], Frame Relay or ATM networks.

5.3.2.3 Address Translation and Mapping Function p. 27

Address translation is the conversion of one address to another address of a different type. Address translation may be used to convert an packet data network protocol address into an internal network address that can be used for routeing packets within and between the PLMN(s).

Address mapping is used to map a network address to another network address of the same type for the routeing and relaying of messages within and between the PLMN(s), for example to forward packets from one network node to another.

5.3.2.4 Encapsulation Function p. 27

Encapsulation is the addition of address and control information to a data unit for routeing packets within and between the PLMN(s). Decapsulation is the removal of the addressing and control information from a packet to reveal the original data unit.

Encapsulation and decapsulation are performed between the core network nodes, and between the GPRS serving support node and the MS.

5.3.2.5 Tunnelling Function p. 27

Tunnelling is the transfer of encapsulated data units within and between the PLMN(s) from the point of encapsulation to the point of decapsulation. A tunnel is a two-way point-to-point path. Only the tunnel endpoints are identified.

5.3.2.6 Compression Function p. 27

The compression function optimises use of radio path capacity by transmitting as little of the SDU (i.e. the exterior PDP PDU) as possible while at the same time preserving the information contained within it. Only IP header compression is supported in Iu mode. The P-GW/GGSN may instruct the SGSN to negotiate no data compression for specific PDP contexts.

5.3.2.7 Ciphering Function p. 27

The ciphering function preserves the confidentiality of user data and signalling across the radio channels and inherently protects the PLMN from intruders.

5.3.2.8 Domain Name Server Function p. 28

The Domain Name Server function resolves logical network node names to addresses. This function is standard Internet functionality according to RFC 1034, which allows resolution of any name for GSNs and other nodes within the GPRS packet domain PLMN backbone networks.

5.3.2.9 DHCP function |R8| p. 28

The Dynamic Host Configuration Function allows to deliver IP configuration information for UEs. This function is standard Internet functionality.

5.3.3 Mobility Management Functions p. 28

5.3.3.1 General |R8| p. 28

The mobility management functions are used to keep track of the current location of an MS within the PLMN or within another PLMN.

5.3.3.2 Idle Mode Signalling Reduction Function |R8| p. 28

The Idle mode Signalling Reduction (ISR) function provides a mechanism to limit signalling during cell reselection in idle mode between GERAN and E-UTRAN or between UTRAN and E-UTRAN and is described in TS 23.401.

5.3.4 Logical Link Management Functions (A/Gb mode) p. 28

Logical link management functions are concerned with the maintenance of a communication channel between an individual MS and the PLMN across the radio interface. These functions involve the co-ordination of link state information between the MS and the PLMN as well as the supervision of data transfer activity over the logical link.

Refer to TS 44.064 for further information.

5.3.4.1 Logical Link Establishment Function p. 28

Logical link establishment is performed when the MS attaches to the PS services.

5.3.4.2 Logical Link Maintenance Functions p. 28

Logical link maintenance functions supervise the logical link status and control link state changes.

5.3.4.3 Logical Link Release Function p. 28

The logical link release function is used to de-allocate resources associated with the logical link connection.

5.3.5 Radio Resource Management Functions p. 28

5.3.5.1 General |R11| p. 28

Radio resource management functions are concerned with the allocation and maintenance of radio communication paths, and are performed by the Radio Access Network. Refer to TS 43.064 and to TS 43.051 for further information on GERAN. Refer to TS 25.301 for further information on UTRAN.

5.3.5.2 RAT/Frequency Selection Priority |R11| p. 28

To support radio resource management in UTRAN/GERAN, the SGSN provides the parameter 'Index to RAT/Frequency Selection Priority' to RNC across Iu and to BSC across Gb. The RFSP Index is mapped by the RNC/BSC to locally defined configuration in order to apply specific RRM strategies. The RFSP Index is UE specific and applies to all the Radio Bearers. Examples of how this parameter may be used in UTRAN/GERAN:

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.

The SGSN receives the subscribed RFSP Index from the HSS (e.g., during the Attach procedure). For non-roaming subscribers the SGSN 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 SGSN chooses the RFSP Index in use based on the subscribed RFSP Index, the locally configured operator's policies and the UE related context information available at the SGSN, including the UE's usage setting and voice domain preference for E-UTRAN, if received during Attach and Routing Area Update procedures (see clause 5.3.15).

For roaming subscribers the SGSN may alternatively choose the RFSP Index in use based on the visited network policy but can take input from 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).

The SGSN forwards the RFSP Index in use to the RNC across Iu and to the BSC across Gb. The RFSP Index in use is also forwarded from source RNC to target RNC during the SRNS Relocation procedure for Intra-RAT handover.

The SGSN stores the subscribed RFSP Index value received from the HSS and the RFSP Index value in use. During the Routing Area Update procedure the SGSN may update the RFSP Index value in use and signal the updated value to the RNC across Iu and to the BSC across Gb, if the locally configured operator's policies indicate to do so (e.g. the SGSN may need to update the RFSP Index value in use if the UE related context information has changed). During inter-SGSN mobility procedures, the source SGSN forwards both RFSP Index values to the target SGSN. The target SGSN 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 SGSN.

The Iu messages that transfer the RFSP Index to the RNC are specified in TS 25.413.

The Gb messages that transfer the RFSP Index to the BSC are specified in TS 48.018.

5.3.5.3 Service identification for improved radio utilisation for GERAN |R11| p. 29

The Service Class Indicator (SCI) (see TS 29.281) enables the GGSN/P-GW to provide the A/Gb mode GERAN access with an indication in the downlink user plane packet to assist the A/Gb mode GERAN access in providing specific RRM treatment in order to improve radio resource control and the overall performance of the GERAN.

In the current specification, the SCI is only applicable for A/Gb mode GERAN access, and only for the Gn/Gp, S4 and the GTP based S5/S8 interfaces.

Support of SCI is optional in GERAN, A/Gb mode SGSN, and PGW/GGSN.

The GGSN/PGW is informed by the SGSN/MME of the UE's current RAT.

When the UE is using an A/Gb mode GERAN the GGSN/PGW determines the value of the SCI based on configuration.The SCI is included in the downlink user plane data packet (see TS 29.281).

There is no impact on the S-GW as part of this feature. If the SCI is received at the S-GW, the S-GW forwards them transparently.

An eNodeB or RNC shall ignore the Service Class Indicator if received over the S1-U, S12 or other interface.

When the serving A/Gb mode SGSN receives SCI in a GTP-U packet, it copies it, without modifying its value, into a Gb interface information element that is sent by the SGSN in the downlink Gb interface user data packet to the GERAN access. In order to allow the GERAN to map the SCI into RRM behaviour, the downlink Gb interface user data packet also carries the HPLMN ID (in the IMSI parameter) and additional information, added by the SGSN, which indicates whether the SCI is assigned:

by a GGSN/P-GW in the Home PLMN, or
by a GGSN/P-GW in the Visited PLMN, or
by a GGSN/P-GW for which the SCIs are coordinated across the different operator group PLMNs and the serving PLMN of the SGSN (Operator Group GGSN).

Absence of additional information is an indication of a VPLMN provided SCI.

The A/Gb mode GERAN uses the information from the SGSN to determine whether to map, and how to map, the SCI to the related RRM behaviour. If the GERAN is not configured with an SCI mapping for the SGSN provided information, then the GERAN shall treat the user plane packet normally, i.e. the GERAN ignores the SCI.

In network sharing configurations (MOCN or GWCN) the SCI can be supported as specified in TS 23.251.

5.3.6 Network Management Functions p. 30

5.3.6.1 General |R10| p. 30

Network management functions provide mechanisms to support O&M functions related to GPRS.

5.3.6.1a GTP-C signalling based Load and Overload Control |R12| p. 30

The S4-SGSN may support GTP-C based Load Control feature for enhanced GW selection procedures as described in TS 23.401, clause 4.3.7.1a.1.

The S4-SGSN may support GTP-C signalling based Overload Control feature as described in TS 23.401, clause 4.3.7.1a.2.

For details on the applicability and use of Load and Overload Control feature for PDN-GW and Serving GW, see TS 23.401, clause 4.3.7.1a.

5.3.6.2 NAS level congestion control |R10| p. 31

5.3.6.2.1 General p. 31

NAS level congestion control contains the functions: "APN based congestion control" and "General NAS level Mobility Management congestion control".

The use of the APN based MM and SM congestion control is for avoiding and handling of MM and SM signalling congestion associated with MSs with a particular APN. Both MSs and network shall support the functions to provide APN based MM and SM congestion control.

The SGSN may detect the NAS signalling congestion associated with the APN and start and stop performing the APN based congestion control based on criteria such as:

Maximum number of active PDP contexts per APN;
Maximum rate of PDP context activations per APN;
One or multiple PDN-GWs or GGSNs of an APN are not reachable or indicated congestion to the SGSN;
Maximum rate of MM signalling requests associated with the devices with a particular subscribed APN; and/or
Setting in network management.

The SGSN may detect the NAS signalling congestion associated with the MSs belonging to a particular group. The SGSN may start and stop performing the group based congestion control based on criteria such as:

Maximum rate of MM and SM signalling requests associated with the devices of a particular group; and/or
Setting in network management.

The SGSN may detect the NAS signalling congestion associated with the MSs that belong to a particular group and are subscribed to a particular APN. The SGSN may start and stop performing the APN and group specific NAS level congestion control based on criteria such as:

Maximum number of active PDP contexts per group and APN;
Maximum rate of MM and SM signalling requests associated with the devices of a particular group and a particular subscribed APN; and/or
Setting in network management.

The SGSN should not apply NAS level congestion control for emergency services.

The SGSN may also use the reject of NAS level Mobility Management signalling requests under general congestion conditions such as detecting congestion of one or several DCNs in an SGSN serving multiple DCNs.

5.3.6.2.2 APN based Session Management congestion control p. 31

The APN based Session Management congestion control may be activated by SGSN due to e.g. congestion situation at SGSN, or by OAM at SGSN, or by a restart or recovery condition of a GGSN or PDN-GW, or by a partial failure or recovery of a GGSN or PDN-GW for a particular APN(s).

The SGSN may reject the Session Management (SM) requests from the MS (e.g. Activate PDP Context and Secondary PDP Context) with a Session Management back-off timer for congested APNs. If the MS provides no APN, then the SGSN uses the APN which is used in GGSN/PDN-GW selection procedure.

The SGSN may deactivate PDP contexts belonging to a congested APN by sending the Deactivate PDP Context Request message to the MS with a Session Management back-off timer. If Session Management back-off timer is set in the Deactivate PDP Context Request message then the cause "reactivation requested" should not be set.

The SGSN may store a Session Management back-off time per MS and APN when congestion control is active for an APN if a request without the low access priority indicator is rejected by the SGSN. The SGSN may immediately reject any subsequent request from the MS targeting to the APN before the stored Session Management back-off time is expired except Modify PDP Context Request -as it may be used by the MS to report 3GPP PS Data Off status change, see clause 5.3.25). If the SGSN stores the Session Management back-off time per MS and APN and the SGSN decides to send a Session Management Request message to a MS connected to the congested APN (e.g. due to decreased congestion situation), the SGSN shall clear the Session Management back-off time prior to sending any Session Management Request message to the MS.

The SGSN should not apply APN based congestion control for emergency services.

Upon reception of the Session Management back-off timer in the Session Management reject message or in the Deactivate PDP Context Request message, the MS shall take the following actions until the timer expires:

If APN is provided in the rejected Session Management Request message or if the Session Management back-off timer is received in the Deactivate PDP Context Request message, the MS shall not initiate any Session Management procedures for the congested APN. The MS may initiate Session Management procedures for other APNs.
If APN is not provided in the rejected Session Management Request message, the MS shall not initiate any Session Management requests without APN. The MS may initiate Session Management procedures for specific APN.
Cell/RA/PLMN/RAT change do not stop the Session Management back-off timer.
The MS is allowed to initiate the Session Management procedures when it is accessing the network with AC11 15 or for emergency services even when the Session Management back-off timer is running.
The MS is allowed to perform MS-initiated PDP Context Modification procedure to report 3GPP PS Data Off status change when the Session Management back off timer is running.
If the MS receives a network initiated Session Management Request message for the congested APN while the Session Management back-off timer is running, the MS shall stop the Session Management back-off timer associated with this APN and respond to the SGSN.
If the MS is configured with a permission for overriding low access priority and the Session Management back-off timer is running due to a reject message received in response to a request with low access priority, the upper layers in MS may request the initiation of Session Management procedures without low access priority.

The MS is allowed to initiate PDN disconnection procedure (e.g. sending Deactivate PDP Context Request) when the Session Management back off timer is running.

The MS shall support a separate Session Management back-off timer for every APN that the MS may activate.

To avoid that large amounts of MSs initiate deferred requests (almost) simultaneously, the SGSN should select the Session Management back-off timer value so that deferred requests are not synchronized.

The APN based Session Management congestion control is applicable to the NAS SM signalling initiated from the MS in the control plane. The Session Management congestion control does not prevent the MS to send and receive data or initiate Service Request procedures for activating user plane bearers towards the APN(s) that are under SM congestion control.

5.3.6.2.3 APN based Mobility Management congestion control p. 32

The SGSN may perform APN based congestion control for MSs with a particular subscribed APN by rejecting Attach procedures with a Mobility Management back-off timer. If the subscription contains a wildcard APN, the SGSN should not reject the request.

When congestion control is active for MSs with a particular subscribed APN, a Mobility Management back-off timer may be sent by the SGSN to MS.

If SGSN maintains the MS context, the SGSN may store the back-off time per MS if a request without the low access priority indicator is rejected by the SGSN. The SGSN may immediately reject any subsequent request from the MS before the stored back-off time is expired.

After rejecting Attach Requests, the SGSN should keep the subscriber data for some time. This allows for the rejection of subsequent requests without HSS signalling when the congestion situation resulting from MSs with a particular subscribed APN persists. Similarly the SGSN may reject Attach Requests based on subscriber data that the SGSN may store after the Detach procedure.

While the Mobility Management back-off timer is running, the UE shall not initiate any NAS request for Mobility Management procedures. However, the UE is allowed to initiate Mobility Management procedures when it is accessing the network with AC11-15 or for emergency services even when the Mobility Management back-off timer is running.

While the Mobility Management back-off timer is running, the MS configured with a permission for overriding low access priority is allowed to initiate Mobility Management procedures without low access priority if the Mobility Management back-off timer was started due to a reject message received in response to a request with low access priority and the upper layers in MS request to activate a PDP context without low access priority or the MS has an activated PDP context that is without low access priority.

To avoid that large amounts of MSs initiate deferred requests (almost) simultaneously, the SGSN should select the Mobility Management back-off timer value so that deferred requests are not synchronized.

5.3.6.2.4 General NAS level Mobility Management congestion control p. 33

Under general overload conditions the SGSN may reject Mobility Management signalling requests from MSs. When a NAS request is rejected, a Mobility Management back-off timer may be sent by the SGSN. If SGSN maintains the MS context, the SGSN may store the back-off time per MS if a request without the low access priority indicator is rejected by the SGSN. The SGSN may immediately reject any subsequent request from the MS before the stored back-off time is expired. While the Mobility Management back-off timer is running, the MS shall not initiate any NAS request for Mobility Management procedures. While the Mobility Management back-off timer is running, the MS is allowed to access the network with AC11-15, perform Detach Procedure, perform mobile terminated services and initiate emergency services. While the Mobility Management back-off timer is running, the MS is allowed to perform Routing Area Update (or combined RA/LA update) if it is already in READY or PMM-CONNECTED state. After any such Detach procedure, the back-off timer continues to run. If the MS receives a paging request from the SGSN while the Mobility Management back-off timer is running, the MS shall stop the Mobility Management back-off timer and initiate the Service Request procedure or the Routeing Area Update procedure as described in clause 6.9.2.1.

The Mobility Management back-off timer shall not impact the UE's function to perform Cell/RAT and PLMN change. Cell/RAT and RA change do not stop the Mobility Management back-off timer. The Mobility Management back-off timer shall not be a trigger for PLMN reselection. The back-off timer is stopped as defined in TS 24.008 when a new PLMN that is not an equivalent PLMN is accessed.

When the MS receives a handover command, the MS shall proceed with the handover procedure regardless of whether the Mobility Management back-off timer is running.

The SGSN should not reject Routing Area Update procedures that are performed when the MS is already in READY or PMM-CONNECTED state.

If the SGSN rejects a Routing Area Update Request or a Service Request with a Mobility Management back-off timer which is larger than the sum of the MS's periodic RA Update timer plus the implicit detach timer, the SGSN should adjust the mobile reachable timer and/or implicit detach timer such that the SGSN does not implicitly detach the MS while the Mobility Management back-off timer is running.

5.3.6.2.5 Group specific NAS level congestion control |R13| p. 34

The group specific NAS level congestion control applies to a specific group of MSs. Each group has a group identifier assigned.

A MS belongs to a group, if the corresponding group identifier is stored in the MS's subscription data in the HLR/HSS. A MS may belong to multiple groups and the SGSN may perform the Group specific NAS level congestion control to an MS as described below independent of whether Group specific NAS level congestion control is activated for one, multiple, or all groups the MS belongs to. The group identifier shall be stored per MS in the HLR/HSS and obtained by the SGSN as part of normal HLR/HSS signalling. A MS is not aware of a group subscription.

The group specific NAS level congestion control may be activated for Session Management signalling, or for Mobility Management signalling, or both. The group specific NAS level congestion control is activated based on operator policies.

When the group specific NAS level congestion control for Session Management signalling is active for a particular group, the SGSN's behaviour is similar to that in clause 5.3.6.2.2, with the following modifications:

SGSN may apply Session Management congestion control to all subscribed APNs for MSs that belong to this particular group.

NOTE:
How the SGSN applies Session Management congestion control to all subscribed APNs is left to Stage 3.
The SGSN may reject the Session Management (SM) requests from the MS belonging to this particular group (e.g. Activate PDP Context, Secondary PDP Context and Modify PDP Context Requests) with a Session Management back-off timer.

When group specific NAS level congestion control for Mobility Management signalling is active for a particular group and a particular APN, the SGSN's behaviour is similar to that in clause 5.3.6.2.3, but applied to MSs with subscribed to this particular group rather that subscribed to a particular APN.

Group specific NAS level congestion control is performed at the SGSN based on the MS's subscription information provided by the HLR/HSS. There is no impact on the MS, and hence, MS's behaviour as described in clauses 5.3.6.2.2 and 5.3.6.2.3 does not change.

5.3.6.2.6 APN and group specific NAS level congestion control |R13| p. 34

The APN and group specific NAS level congestion control is the interclause of APN specific NAS level congestion control and Group specific NAS level congestion control, i.e. it applies to a specific group of MSs with a particular subscribed APN. Each group of UEs has a group identifier assigned and stored in the HSS.

A MS may belong to multiple groups and the SGSN may perform the APN and group specific NAS level congestion control to a MS as described below independent of whether the APN and group specific NAS level congestion control is activated for one, multiple or all groups the MS belongs to. The group identifier(s) shall be stored per MS in the HLR/HSS and obtained by the SGSN as part of normal HLR/HSS signalling. A MS is not aware of the group identifier(s) that the UE belongs to.

The APN and group specific NAS level congestion control may be activated for Session Management signalling, or for Mobility Management signalling, or both. The APN and group specific NAS level congestion control is activated based on operator policies.

When the APN and group specific NAS level congestion control for Session Management signalling is activated for a MS belonging to a particular group and initiating signalling to a particular subscribed APN, the SGSN's behaviour is similar to that in clause 5.3.6.2.2, with the following modifications:

The Session Management (SM) congestion control is applied to this particular APN, and for MSs belonging to this particular group.
The SGSN may reject the SM requests from the MS belonging to this particular group and attaching to this particular APN (e.g. Activate PDP Context and Secondary PDP Context) with a Session Management back-off timer. If the MS provides no APN, then the SGSN uses the APN which is used in GGSN/PDN-GW selection procedure.
The SGSN may deactivate PDP contexts from the MSs, belonging to this particular group and attaching to this particular APN, by sending the Deactivate PDP Context Request message to the MS with a Session Management back-off timer.

When APN and group specific NAS level congestion control for Mobility Management signalling is activated for a MS belonging to a particular group and with a particular subscribed APN, the SGSN's behaviour is similar to that in clause 5.3.6.2.3, but applied to MSs with this particular subscribed APN and belonging to this particular group.

APN and group specific NAS level congestion control is performed at the SGSN based on the MS's subscription information provided by the HLR/HSS. There is no impact on the MS, and hence, MS's behaviour described in clauses 5.3.6.2.2 and 5.3.6.2.3 does not change.

5.3.6.3 GGSN control of overload |R10| p. 35

The GGSN may provide mechanisms for avoiding and handling overload situations. These include the rejection of PDP context requests from UEs.

The GGSN may detect congestion per APN and start and stop performing overload control based on criteria such as:

Maximum number of active PDP contexts per APN and/or
Maximum rate of PDP context activations per APN.

When performing overload control the GGSN rejects PDP context requests. When receiving the rejection from the GGSN, the SGSN rejects the UE's PDP context request as specified in clause 5.3.6.2. In addition the GGSN may indicate a "GGSN back-off time" for a specific APN to the SGSN. The SGSN should reject PDP context requests from UEs for the specific APN related to that GGSN during the "GGSN back-off time", by the means specified in clause 5.3.6.2. If a GGSN indicates APN congestion by the "GGSN back-off time" the SGSN may select another GGSN of that APN instead of rejecting PDP context requests. unless there is already an existing PDP context to the same APN for the MS, in which case, the SGSN shall reject PDP context request

5.3.6.4 SGSN control of overload |R10| p. 35

The SGSN contains mechanisms for avoiding and handling overload situations. In an overload situation the SGSN can request the RNC to reduce any kind of signalling traffic as specified in TS 25.413.

In addition, the SGSN can request the BSC/RNC to reject the RR(C) connection establishments from MSs that access the network with low access priority that its connected BSCs/RNCs are generating on it.

A BSC/RNC supports rejecting of RR(C) connection establishments from MSs that access the network with low access priority. When rejecting an RR(C) connection request for overload reasons the BSC/RNC indicates to the MSs an appropriate timer value that limits further RR(C) connection requests.

If the network is operating in Network Mode of Operation II, then (because at a common LA/RA boundary Location Area Updates are initiated before Routeing Area updates) MSs will often initiate signalling connections towards the SGSN while in RRC connected state. If the SGSN has indicated an overload situation to the RNC, then the RNC can use the Signalling Connection Release message to avoid establishing the signalling connection with the SGSN.

Additionally, a BSC/RNC provides support for the barring of MSs configured for Extended Access Barring, as described in TS 22.011. These mechanisms are further specified in TS 48.016 and TS 44.018 for GERAN, TS 25.331 for UTRAN.

A BSC/RNC may initiate Extended Access Barring when:

all the SGSNs (and all the MSCs) connected to a BSC/RNC request to restrict the load for MSs that access the network with low access priority; or
requested by O&M.

If a SGSN requests a BSC/RNC to restrict the load for MSs that access the network with low access priority, the SGSN should select all the BSCs/RNCs with which the SGSN has Gb/Iu interface connections (so that Extended Access Barring can be triggered if all SGSNs within a pool area are experiencing the same overload situation). Alternatively, the selected BSCs/RNCs may be limited to a subset of the BSCs/RNCs with which the SGSN has Gb/Iu interface connections (e.g. particular location area or where MSs of the targeted type are registered).

For GERAN, subsequent initial access attempts by a previously barred MS through Extended Access Barring is described in TS 44.018.

In addition, to protect the network from overload the SGSN has the option of rejecting NAS request messages which include the low access priority indicator before rejecting NAS request messages without the low access priority indicator (see clause 5.3.6.2 for more information).

5.3.6.5 S4-SGSN control of overload |R10| p. 36

Under unusual circumstances (e.g. when the S4-SGSN load exceeds an operator configured threshold), the S4-SGSN may restrict the signalling load that its SGWs are generating on it, if configured to do so.

The S4-SGSN can reject Downlink Data Notification requests for low priority traffic for UEs in idle mode or to further offload the S4-SGSN, the S4-SGSN can request the SGWs to selectively reduce the number of Downlink Data Notification requests it sends for downlink low priority traffic received for UEs in idle mode according to a throttling factor and for a throttling delay specified in the Downlink Data Notification Ack message.

SGW and S4-SGSN determine whether a bearer is for low priority traffic or not on the basis of the bearer's ARP priority level and operator policy (i.e. operator's configuration in the SGW and S4-SGSN of the ARP priority levels to be considered as priority or non- priority traffic). The S4-SGSN determines whether a Downlink Data Notification request is for low priority traffic or not on the basis of the ARP priority level that was received from the SGW and operator policy.

If ISR is not active for the UE, during the throttling delay, the SGW drops downlink packets received on all its low priority bearers for UEs known as not user plane connected (i.e. the SGW context data indicates no downlink user plane TEID) served by that S4-SGSN in proportion to the throttling factor, and sends a Downlink Data Notification message to the S4-SGSN only for the non throttled bearers.

If ISR is active for the UE, during the throttling delay, the SGW does not send DDN to the S4-SGSN and only sends the DDN to the MME. If both MME and SGSN are requesting load reduction, the SGW drops downlink packets received on all its low priority bearers for UEs known as not user plane connected (i.e. the SGW context data indicates no downlink user plane TEID) in proportion to the throttling factors.

The SGW resumes normal operations at the expiry of the throttling delay. The last received value of the throttling factor and throttling delay supersedes any previous values received from that S4-SGSN. The reception of a throttling delay restarts the SGW timer associated with that S4-SGSN.

5.3.6.6 Throttling of NIDD Submit Requests |R14| p. 36

Under unusual circumstances (e.g. when the SGSN load exceeds an operator configured threshold), the SGSN may restrict NIDD Submit Request messages that its SCEFs are generating on it, if configured to do so.

5.3.7 Selection functions |R8| p. 36

5.3.7.1 SGW/PGW/GGSN selection function (3GPP accesses) p. 36

The SGSN supporting both S4 and Gn/Gp shall support selection of SGW/PGW and GGSN.

The Gn/Gp SGSN shall support selection of GGSN and may optionally support selection of PGW.

For a given UE, the SGSN shall select the same GGSN/PGW for all the PDP contexts belonging to the same APN.

At PDP Context activation, in addition to network local policy, it shall be possible for SGSN to use:

the UE capability (as indicated in the MS Network Capability);
subscription restriction to use NR as secondary RAT;
the configuration about the roaming agreement for E-UTRAN with the HPLMN of the UE; and
the UE Usage Type if DCNs are deployed in the network,

as input to select GGSN, or a SGW and PGW.

It shall be possible to configure the selection function on the SGSN to give priority towards SGW/PGW for E-UTRAN capable UEs, and GGSN for non E-UTRAN capable UE.

It shall also be possible to configure the selection function on the SGSN to select GGSN/PGW using Gp when there is no roaming agreement for E-UTRAN with the HPLMN of the UE.

The S4-SGSN supporting GTP-C Load Control feature performs enhanced PDN-GW selection as described in the clause 4.3.8.1 of TS 23.401.

5.3.7.2 Serving GW selection function p. 37

The Serving GW selection function is described in the clause "Serving GW selection function" of TS 23.401.

5.3.7.3 SGSN selection function p. 37

The SGSN selection function selects an available SGSN to serve a UE. The selection is based on network topology, i.e. the selected SGSN serves the UE's location and in case of overlapping SGSN service areas, the selection may prefer SGSNs with service areas that reduce the probability of changing the SGSN. Other criteria for SGSN selection may be load balancing between SGSNs. In networks that deploy dedicated MMEs/SGSNs, e.g. for UEs configured for low access priority, the possible target SGSN selected by source MME/SGSN is typically restricted to SGSNs with the same dedication.

When a MME/SGSN supporting DCNs selects a target SGSN, the selected target SGSN is restricted to SGSNs that belong to the same DCN. The DNS procedure may be used by the source CN node to select the target SGSN from a given DCN. If both low access priority and UE Usage Type parameter are used for SGSN selection, selection based on UE Usage type parameter overrides selection based on the low access priority indication.

When selecting an SGSN for an MS that is accessing for the purpose of utilizing EC-GSM-IoT (see TS 43.064), the SGSN selection function in the BSC shall select an SGSN taking into account the SGSN's support (or non-support) for CIoT GSM Optimization.