Background A basic concept in the 3 ^rd Generation Partnership Project (3GPP) Evolved Packet Core (EPC) architecture is a Packet Data Network (PDN). A PDN is an Internet Protocol (IP) network. This is typically the Internet, but it can also be a closed corporate network or an operator service network like an IP Multimedia system (IMS). A PDN has one or more names, each name represented in a string called Access Point Name (APN). A PDN gateway (PDN-GW or PGW) is a functional node that provides access to one or more PDNs.

A PDN connection provides a user equipment (UE) with an access channel to a PDN. It is a logical IP tunnel between UE and PGW. Each PDN connection has a single IP address/prefix. A UE can setup multiple PDN connections, possibly to the same APN.

Figure 1 is a signalling diagram showing how a UE attaches to a Wireless Local Area Network (WLAN) in single-connection mode. This is a simplified version of the procedure described more fully in 3GPP TS 23.402 section 16.2.

The UE receives a Media Access Control (MAC) address of a Trusted Wireless Access Gateway (TWAG) as part of an EAP-AKA' authentication process (block 4). In single- connection mode the UE uses that MAC address to set a destination address field in the MAC header of uplink packets.

Two types of WLAN-WLAN handover may be defined. The first type is an Access Point (AP) to AP handover where both source and target AP are associated with one and the same TWAG. The second type is an AP to AP handover where the source and target AP are associated with two different TWAGs. We denote this as a TWAG-to-TWAG handover. Figure 2 is a signalling diagram showing a TWAG-to-TWAG handover. Note that a TWAG is in a non-3GPP access domain. Handover from non-3GPP access to another non-3GPP access is out-of-scope for 3GPP standardization, and therefore the signalling diagram of Figure 2 is not included in 3GPP TS 23.402. However, given the 3GPP TS 23.402 procedure for handover from 3GPP to WLAN, it can be anticipated that the procedure of Figure 2 will be used for TWAG-to-TWAG handovers.

As a pre-condition, the UE has a PDN connection via a source AP. In step 1 the UE finds a target AP. The UE then sends an association request (step 2) and includes its existing authentication keys according to IEEE 802.1 1 r. The target TWAG does not have a security context for the UE and therefore it starts a full EAP-AKA' authentication process (block 4 comprising steps 5-9). Because this call flow is not yet standardized, it is unclear whether the UE will indicate handover in step 5.

IEEE 802.1 1 r is an amendment to the IEEE 802.1 1 -2007 specification and defines an optimization for AP-to-AP handover. The UE is already authenticated on a source AP and re-uses the same authentication keys when a handover to a target AP is performed. So, upon an AP-to-AP handover within the same TWAG, the EAP-AKA' authentication block can be skipped if IEEE 802.1 1 r is supported. This would give the signalling diagram of Figure 3.

Note that the Association Request (step 2) and Association Response (step 3) are between the UE and the TWAG in the signalling diagrams disclosed herein. These signals are according to IEEE 802.1 1 defined between an STA (a station e.g., an end device or the UE) and the AP. In one implementation, an IEEE 802.1 1 AP may be divided into a radio head-end (mostly also called an AP) and a controller (mostly called an AC). The exact division of functionality between the AP and the AC is implementation specific. For example, one implementation supports IEEE 802.1 1 r and the Association Request is only forwarded by the AP to the AC upon an AP-to-AP handover. The AP does not forward the Association Request when the UE re- associates to the same AP. The interface between the AP and the AC is often a proprietary extension of the Control and Provisioning of Wireless Access Points (CAPWAP) protocol (see RFC 5415). Figures 1 -3 assume a so-called single-connection mode. If a UE is in this mode then the number of PDN connections over the WLAN is restricted to a maximum of one. If a UE is in multi-connection mode, then it may have zero, one or multiple PDN connections over the WLAN. See 3GPP TS 23.402 section 16.1.2 for a definition of these modes. The UE and the network negotiate the mode of operation during one or more EAP-AKA' messages.

Figure 4 shows how an attach to the WLAN in multi-connection mode works. This is a simplified version of the procedures in 3GPP TS 23.402 section 16.2.1 and 16.8. Block 7 (comprising steps 8-12) is repeated for each PDN connection. Step 10 is only executed upon handover from a 3GPP access. The UE would in that case have indicated handover in step 8. The signalling in block 7 between the UE and the TWAG is performed by a WLAN Control Protocol (WLCP) defined in 3GPP TS 23.402 section 16.1 .4A. WLCP is a User Datagram Protocol (UDP)/IP based protocol, so in order to do WLCP signalling the UE first acquires a Non-Seamless WLAN Offload (NSWO) address in step 6. When the UE sends a WLCP signal to the TWAG, it uses the TWAG address received as part of step 4.

Figure 5 describes how a TWAG-to-TWAG handover would work in multi-connection mode. Just as for single-connection mode, TWAG-to-TWAG handover in multi- connection mode is out-of-scope for 3GPP and therefore the call flow below is not included in 3GPP TS 23.402. However, given the 3GPP TS 23.402 procedure for handover from 3GPP to WLAN, it can be anticipated that the procedure below will be used for TWAG-to-TWAG handovers.

In step 2 of Figure 5 the UE will indicate its authentication keys according to IEEE 802.1 1 r. However, the target TWAG will not recognize these keys, and the complete EAP-AKA' authentication procedure will be performed. After this, block 7 will be performed for each PDN connection. Because this call flow is not standardized, it is not clear if the UE will indicate handover in step 8.

Figures 1 -5 describe the scenario where one or more PDN connection is setup. That is, the traffic is routed through the PGW. This is called "EPC-routed". An alternative is NSWO. NSWO traffic does not pass through the PGW. Instead, the TWAG routes the traffic directly to the Internet. For EPC-routed PDN connections the PGW assigns the IP address. For NSWO the TWAG assigns the IP address. In single-connection mode, the UE requests either EPC-router or NSWO by means of a flag in EAP-AKA' authentication. In multi-connection and IPv4 the UE uses its NSWO IPv4 address not only for NSWO signalling towards Internet but also for IPv4 WLCP signalling. For multi- connection and IPv6 the UE uses a NSWO address with link-local scope for WLCP signalling and another NSWO address with global scope for signalling towards Internet.

This document deals primarily with TWAG-to-TWAG handover of PDN connections. NSWO handling is only covered when needed for WLCP signalling.

One problem with existing TWAG-to-TWAG handovers is that the full EAP-AKA' authentication procedure needs to be performed. This takes time; typically in the order of seconds. That severely limits handover performance and causes large packet drops.

Another problem is that the TWAG-to-TWAG handover is not a standardized call flow. It is therefore undefined if the UE will indicate handover to the network when a TWAG- to-TWAG handover is performed. Summary

According to the invention in a first aspect, there is provided a target trusted wireless access gateway, TWAG, for establishing at least one connection with at least one user equipment, UE, at a handover from a source TWAG in a telecommunications system. The target TWAG comprises a receiving means, which may be a receiver configured to receive an association request from the at least one UE. The target TWAG comprises a location requestor means, which may be a UE location requestor configured to obtain an address of the source TWAG. The target TWAG comprises a context requesting means, which may be a UE context requester configured, based on the obtained address of the source TWAG, to control a transmitting means, which may be a transmitter configured to control the transmitter to transmit a request to modify a bearer or bearers associated with the at least one UE and to establish at least one connection with the at least one UE, based on the received UE context. Optionally, the UE location requestor is configured to control the transmitter to transmit to a first further node a request for an address of the source TWAG being further configured to receive a response comprising the address of the source TWAG. Optionally, the UE location requestor is configured to obtain the address of the source TWAG based on a Basic Service Set Identifier of the source TWAG received in the association request.

Optionally, the association request comprises at least one UE media access control, MAC, address, and wherein obtaining the address of the source TWAG is based on the at least one UE MAC address.

Optionally, the UE context request comprises at least one UE MAC address and wherein the received UE context comprises one or more of: the security context for the at least one UE; at least one PDN connection identifier for the at least one UE; a source TWAG MAC address; and a source TWAG internet protocol, IP, address.

Optionally, the UE context comprises a plurality of PDN connection identifiers, and wherein the handover controller is configured to control the transmitter to transmit a request to establish a plurality of PDN connections to the at least one UE is configured to control the transmitter to transmit a modify bearer request to a packet data network gateway, PGW.

According to the invention in a second aspect, there is provided a method for operating a TWAG for establishing at least one connection with at least one user equipment, UE, at a handover from a source TWAG in a telecommunications system. The method comprises receiving, by a receiver configured to control a transmitter comprises: transmitting, by the UE location requestor configured to control the transmitter comprises obtaining the address of the source TWAG based on a Basic Service Set Identifier of the source TWAG received in the association request.

Optionally, the association request comprises at least one UE media access control, MAC, address, and wherein obtaining the address of the source TWAG is based on the at least one UE MAC address. Optionally, the UE context request comprises at least one UE MAC address and wherein the received UE context comprises one or more of: the security context for the at least one UE; at least one PDN connection identifier for the at least one UE; a source TWAG MAC address; and a source TWAG internet protocol, IP, address.

Optionally, the UE context comprises a plurality of PDN connection identifiers, the method further comprising controlling, by the handover controller to transmit a request to establish a plurality of PDN connections to the at least one UE, based on the PDN connection identifiers.

Optionally, the method further comprises the handover controller controlling the transmitter to transmit a modify bearer request to a packet data network gateway, PGW. According to the invention in a third aspect, there is provided a computer program comprising instructions which, when executed on at least one processor, cause the at least one processor to carry out the method above.

According to the invention in a fourth aspect, there is provided a carrier containing the computer program above, wherein the carrier is one of an electronic signal, optical signal, radio signal, or non-transitory computer readable storage medium.

According to the invention in a fifth aspect, there is provided a node comprising a locator function in a telecommunications network. The locator function comprises a receiving means, which may be a receiver a request for an address of a source TWAG serving a UE, wherein the request comprises a UE identifier. The locator function comprises a receiving means, which may be a UE location responder is a UE location request comprises a target TWAG address, and wherein the UE identifier comprises a MAC address for the UE is configured use the UE identifier to obtain the source TWAG address from a look up table.

Optionally, the node is a TWAG in a telecommunications network. The method comprises receiving from a target TWAG serving a UE, wherein the request comprises a UE identifier. The method comprises obtaining, by a UE location responder to transmit the source TWAG address to the target TWAG. Optionally, the request for the address of the source TWAG is a UE location request comprises a target TWAG address, and wherein the UE identifier comprises a MAC address for the UE using the UE identifier to obtain the source TWAG address from a look up table.

Optionally, the node is a TWAG comprising instructions which, when executed on at least one processor, cause the at least one processor to carry out the method above. According to the invention in a eighth aspect, there is provided a carrier containing the computer program above, wherein the carrier is one of an electronic signal, optical signal, radio signal, or non-transitory computer readable storage medium.

According to the invention in a ninth aspect, there is provided a source TWAG for handing over at least one connection with at least one UE to a target TWAG in a telecommunications system. The source TWAG comprises a receiving means, which may be a receiver based on the UE identifier and to control a transmitter to transmit the UE context to the target TWAG. Optionally, the UE identifier is a UE MAC address.

Optionally, the UE context comprises one or more of: a security context for the at least one UE; at least one PDN connection identifier for the at least one UE; a source TWAG MAC address; and a source TWAG internet protocol, IP, address.

Optionally, the source TWAG further comprises an address obtaining means, which may be an address obtainer is configured to control the transmitter to transmit the unique address to the at least one UE for handing over at least one connection with at least one UE to a target TWAG in a telecommunications system. The method comprises receiving, by a receiver based on the UE identifier. The method comprises controlling, by the UE context responder, a transmitter to transmit the UE context to the target TWAG.

Optionally, the UE identifier is a UE MAC address. Optionally, the UE context comprises one or more of: a security context for the at least one UE; at least one PDN connection identifier for the at least one UE; a source TWAG MAC address; and a source TWAG internet protocol, IP, address. Optionally, the method further comprises obtaining by an address obtainer to transmit the unique address to the at least one UE comprising instructions which, when executed on at least one processor, cause the at least one processor to carry out the method above. According to the invention in a twelfth aspect, there is provided a carrier containing the computer program above, wherein the carrier is one of an electronic signal, optical signal, radio signal, or non-transitory computer readable storage medium.

Brief description of drawings

Figure 1 is a signalling diagram showing how a UE attaches to a Wireless Local Area Network (WLAN) in single-connection mode;

Figure 2 is a signalling diagram showing a TWAG-to-TWAG handover in single- connection mode;

Figure 3 is a signalling diagram showing an AP-to-AP handover within the same TWAG;

Figure 4 is a signalling diagram showing an attach to a WLAN in multi-connection mode;

Figure 5 is a signalling diagram showing a TWAG-to-TWAG handover in multi- connection mode;

Figure 6a shows an architecture diagram;

Figure 6b shows a schematic diagram of a TWAN;

Figure 7 is a signalling diagram showing a method for TWAG-TWAG handover for the single-connection mode;

Figure 8 is a signalling diagram showing an initial WLAN attach to a TWAG;

Figure 9 is a signalling diagram for TWAG-to-TWAG handover with a point-to-point link between the UE and the source TWAG;

Figure 10 is a signalling diagram showing an initial WLAN attach in multi-connection mode where a point-to-point link between UE and TWAG can be assumed; Figure 1 1 is signalling diagram showing initial WLAN attach in multi-connection mode when no point-to-point link between TWAG and UE can be assumed;

Figure 12 shows a schematic diagram of a target TWAG;

Figure 13 shows a schematic diagram of a source TWAG; and

Figure 14 shows a schematic diagram of a locator function.

Detailed description

Generally, disclosed herein are methods and apparatus for hiding a TWAG-to-TWAG handover for the UE. The UE will not see a difference between an AP-to-AP handover within the same TWAG, and a TWAG-to-TWAG handover. This may be achieved by copying a state from the source TWAG to the target TWAG. The state would include a security context, PDN connection information, and a TWAG address(es). It is assumed herein that the AC function is part of the TWAG. In practical implementations, the AC and the TWAG function may be in different nodes. In such cases, an interface between the AC and the TWAG is needed for the ideas presented here to work. Figure 6a shows an architecture diagram for connection of a UE 600 to a PGW 602 via a source (or first) TWAN 604a. The source TWAN 604a is the TWAN to which the UE 600 is currently connected. The UE 600 is in electrical communication with the source TWAN 604a via a first SWw interface 606a. The source TWAN 604a is in electrical communication with the PGW 602 by a first S2a interface 608a. The PGW 602 then provides access to other nodes in the Home Public Land Mobile Network (HPLMN).

A target (or second) TWAN 604b is also in electrical communication with the PGW 602 via a second S2a interface 608b. The target TWAN 604b is a TWAN to which the UE wishes to connect as part of a TWAG to TWAG handover from the source TWAN 604a. The UE 600 may establish electrical communication with the target TWAN 604b over a second SWw interface 606b, which is shown as a dashed line to indicate that the electrical communication will be established as part of the TWAG to TWAG handover.

It will be understood that after a handover, the target TWAG will become the source TWAG. Figure 6b shows a schematic diagram of a TWAN 604. The TWAN 604 may be a source TWAN 604a or a target TWAN 604b and comprises a WLAN Access Network 610 (one or more APs), a Trusted WLAN AAA Proxy 612 and a TWAG 614. It is noted one or more of the features of a source TWAN 604a may be the same as those of a target TWAN 604b. Therefore, features of a source (or first) TWAN 604a are given a suffix 'a' and the features of a target (or second) TWAN 604b are given a suffix 'b'. The SWw interface 606 provides electrical communication to the WLAN Access Network 610. In addition, the S2a interface 608 provides electrical communication with the TWAG 614.

Single-connection mode

Figure 7 is a signalling diagram describing a method for TWAG-TWAG handover for the single-connection mode. The target TWAG 614b does not send a Create Session Response to the PGW 602, only a Modify Bearer Response. From the UE 600 point of view, the procedure is the same as an AP-to-AP handover.

The locator function is a database that keeps track of which source TWAG 614a currently serves which UE 600. Using the locator function, the target TWAG 614b finds out what the source TWAG 614a is (steps 3 and 4). That is, in step 3, the target TWAG 614b transmits a UE Location Request comprising the target TWAG address and the UE MAC to the locator function. The locator function processes the request and responds to the target TWAG 614b with a UE Location Response comprising the source TWAG address.

The locator function can be implemented in several ways.

The locator function may logically be a central register. The UE's MAC address, received by the target TWAG 614b in step 2, is sent in step 3 to the locator function and used by the locator function as a lookup key. The locator function replies to the target TWAG 614b with the address of the source TWAG 614a that is currently serving this UE 600 (step 4). Another possible approach is to implement the locator function in a distributed way. Each TWAG 314 could be pre-configured with a list of Basic Service Set Identifiers (BSSIDs) mapped to a TWAG address. A BSSID is the MAC address of an AP. The UE 600 sends the BSSID of the source AP in step 2 according to IEEE 802.1 1 r. Based on this, the target AP can find the address of the source TWAG 614a. In this variant, the locator function would be internal to the TWAG 614.

It will be understood that other ways to implement the locator function are also possible.

After the exchange between the locator function, the target TWAG 614b fetches the UE context from the source TWAG 614a (step 5-6) based on the source TWAG address obtained from the locator function. The target TWAG 614b transmits a UE context request to the source TWAG 614a using the source TWAG address. The UE context request comprises the UE MAC address. The source TWAG 614a sends a UE context response to the target TWAG 614b.

The UE context response may include one or more of: · Security information like authentication keys. This way, the target TWAG

614b can apply the IEEE 802.1 1 r procedures and skip the EAP-AKA' procedure.

• PDN connection context. This way, the target TWAG 614b can request the PGW 602 to perform a handover in a modify bearer request (step 7-9).

• Other information related to the sending and receiving of packets between target TWAG 614b and UE 600, such as netmask, Maximum Transfer Unit (MTU), etc.

In single-connection mode, there is by definition a point-to-point link between the UE 600 and the source TWAG 614a (see 3GPP TS 23.402 section 16). This means that all packets sent by the UE will end up on the source TWAG 614b regardless of the destination IP address the UE 600 would use in uplink packets. In typical systems, the UE 600 receives the TWAG MAC address as part of the EAP- AKA' authentication signaling as explained above. That address is used by the UE 600 as a destination MAC address in uplink packets. After handover to the target TWAG 614b, several implementations are possible when it comes to addressing the target TWAG 614b.

In one implementation, the target TWAG 614b may simply ignore the destination MAC address in uplink packets. The target TWAG 614b only examines the packet source MAC address; i.e. the UE's address. In such an implementation it is not necessary to include the source TWAG MAC address in the context from the source TWAG 614a (step 6 above). In this implementation the destination MAC address used by the UE 600 may, for example, be the same for all UEs. In another implementation, the target TWAG 614b routes the incoming traffic based on the packet destination MAC address only. The target TWAG 614b may inspect the packet source MAC address to e.g. verify the UE 600 identity, but user plane handling in the target TWAG 614b is not dependent on it. In this case the implementation would also require that the packet destination MAC address is unique within a network of TWAGs in order to identify the UE 600 and the traffic uniquely. Such a requirement may arise from the way the point-to-point link is implemented. In such a scenario, the source TWAG MAC address sent to the UE 600 in the initial WLAN attach would need to be unique for each individual UE 600. This could be implemented by the source TWAG 614a taking such a MAC address from a MAC address pool common to all TWAGs. In such a scenario the source TWAG MAC address would need to be sent in the context (step 6 above). By copying the source TWAG MAC address to the target TWAG 614b, the UE 600 can continue to use the same MAC address as destination address in uplink packets. In yet another implementation, the target TWAG 614b may handle the user plane based on both the packet source MAC address and the packet destination MAC address. In such a scenario the source TWAG MAC address would need to be sent in the context (step 6 above). A MAC address pool could be used also in this case, but there is however no strict requirement that the destination address is unique within the network. In the initial WLAN attach, the UE 600 will not have been registered yet in the locator function, so step 3 of Figure 7 will result in a registration. Figure 8 is a signalling diagram showing an initial WLAN attach to a TWAG, which will become the source TWAG 614a during a TWAG to TWAG handover. Steps 5 and 6 are only performed when there is a requirement on the uniqueness of the source/target TWAG MAC address during handover and may be performed directly after step 4 as indicated below, but may also be performed later. Steps 5 and 6 are performed before the MAC address is sent to the UE 600 in the EAP-AKA' signalling.

The ideas presented for single-connection mode can also be applied in the multi- connection mode. Contrary to single-connection mode, in the current specification there is not necessarily a point-to-point link between UE and TWAG in multi-connection mode. The following sub-section describes the case when there is a point-to-point link between UE and TWAG in multi-connection mode. The case when no point-to-point link can be assumed is described below.

Multi-connection mode

The following describes the case when there is a point-to-point link between the UE 600 and the source TWAG 614a in multi-connection mode. A later section describes the case when no point-to-point link can be assumed. Figure 9 is a signalling diagram for TWAG-to-TWAG handover with a point-to-point link between the UE 600 and the source TWAG 614a. Contrary to the single-connection mode, the UE Context Response (step 6) may include a security context for multiple PDN connections. Each PDN connection is then handed over (block 9) by way of a modify bearer request and response between the target TWAG 614b and the PGW 602. Note that only a Modify Bearer Request towards the PGW 602 (step 10) is needed, no Create Session Request. In multi-connection mode, the source TWAG 614a has an IP address that is used by the UE 600 to address the WLCP function on the source TWAG 614b. This IP address is included in the UE Context Response (step 6). By copying the source TWAG IP address to the target TWAG 614b, the UE 600 does not need to perform any WLCP signalling as part of the TWAG-to-TWAG handover. Note that, as we assume a point- to-point link between the UE 600 and the source TWAG 614a, there is no requirement on global uniqueness of the source TWAG IP address for WLCP signalling.

For the assignment of and usage of the source TWAG MAC address for WLCP signalling, the same options apply as discussed above in respect of single-connection mode. So, depending on the implementation of the point-to-point link, the source TWAG MAC may or may not need to be unique for each UE 600 and within the network of TWAGs. If uniqueness is required, the MAC pool can be used as discussed above.

In multi-connection mode, the specification (3GPP TS 23.402 section 16) requires that each PDN connection is assigned a MAC address that is used to identify traffic belonging to that PDN Connection. The MAC address needs to be unique at least for each UE 600 and source TWAG 614a. For the assignment of the TWAG MAC address similar variants apply as discussed above in respect of single-connection mode, i.e. depending on the way the point-to-point link is implemented the MAC for each PDN connection may or may not need to be unique within the network of TWAGs. In any case, the assigned MACs need to be included in the UE context (step 6). Figure 10 is a signaling diagram describing an initial WLAN attach in multi-connection mode. A new source TWAG MAC address is taken for the TWAG WLCP signaling and for every PDN connection. Note that a UE 600 at any time can add an additional PDN connection by executing block 12 (comprising steps 13-19). As in single-connection mode, steps 5 and 6 are only performed when there is a requirement on the uniqueness of the source/target TWAG MAC address and may be performed directly after step 4 as indicated below, but may also be performed later. Steps 5 and 6 may be performed before the UE 600 starts user plane signaling. The MAC address pool for steps 16 and 17 would become a TW AG-internal address pool if there is no requirement on the uniqueness of the source/target TWAG MAC addresses within the network of TWAGs. Regarding the locator function the same applies as for the single-connection mode; i.e. in the initial WLAN attach the UE's source TWAG 614a gets registered in the locator function.

For IPv4, the UE 600 uses the same TWAG IPv4 address for both NSWO and WLCP signalling. If this would be a globally unique address assigned in the source network, then the address topologically belongs to the source network. If such address is copied to the target network, then downlink NSWO traffic from Internet will still be routed to the source network. This may become a problem, in particular when source and target are disjointed Ethernet networks. There are multiple solutions to this. One solution is to co-locate a Network Address Translator (NAT) in the target/source TWAG. It would then be possible to copy the IP address from the source TWAG 614a to the target TWAG 614b so the UE 600 can continue to use that same address for WLCP signalling. The UE 600 can even continue to use the same for NSWO signalling, but on-going sockets will get released as the outer NAT address is changed.

If a point-to-point link between the UE 600 and the source TWAG 614a cannot be assumed then the requirements on uniqueness of the addresses become higher.

TWAG IP address for WLCP signalling: The UE 600 received the TWAG IP address (IPv4 or IPv6 or both) as part of authentication signalling. Copying that address to the target TWAG 614b is only possible if the address is not already in use in the target TWAN 604b. In other words, the address needs to be unique within the network of TWAGs. Furthermore, the address needs to be unique per UE 600. Otherwise, it would not be possible to move a first UE 600 to a target TWAG 614b while keeping a second UE 600 on the source TWAG 614a. Setting up a TWAG IPv6 address that is unique per UE 600 and within the network is possible by generalizing the MAC pool to an "address pool", where address is either a MAC address or an IP address. TWAG IPv4 address for NSWO signalling: As there is no point-to-point link, the UE 600 may have received the NSWO address from a Dynamic Host Conversion Protocol (DHCP) v4 server that is not co-located with the source/target TWAG. The same routing problem arises as explained in the previous section. The same solution can apply as well, with an addition requirement that the DHCPv4 server is placed in the source/target TWAG.

TWAG MAC addresses for WLCP and for PDN connections: Copying these addresses from source TWAG 614a to target TWAG 614b is only possible if the addresses are unique per UE 600 and unique within the network of TWAGs. The idea of the address pool can be used.

The resulting signalling diagram for initial WLAN attach, as shown in Figure 1 1 , remains the same as when a point-to-point link between the UE 600 and the source TWAG 614a. The signalling for the initial WLAN attach is the same, with the addition that also the IP address needs to be taken from an address pool.

Figure 12 shows a schematic diagram of a target TWAG 1200. The target TWAG 1200 may be used in the architecture shown in Figure 6a.

The target TWAG 1200 comprises a transmitter 1202 and a receiver 1204. The transmitter 1202 and receiver 1204 are in electrical communication with other communication units, nodes, UEs, servers and/or functions in a telecommunications network and are configured to transmit and receive data accordingly.

It is noted that the term "electrical communication" encompasses both wired and wireless electrical communication. Therefore, electrical communication may be, for example, a network communication over a wired connection or a network communication of over a radio frequency connection. The target TWAG 1200 further comprises at least one memory 1206 and at least one processor 1208. The memory 1206 may comprise a non-volatile memory and/or a volatile memory. The memory 1206 may have a computer program 1210 stored therein. The computer program 1210 may be configured to undertake the methods disclosed herein. The computer program 1210 may be loaded in the memory 1206 from a non-transitory computer readable medium 1212, on which the computer program is stored. The processor 1208 is configured to undertake at least the functions of UE location requester 1214, a UE context requestor 1216, a handover controller 1218 and a NAT 1220, as set out herein.

Each of the transmitter 1202 and receiver 1204, memory 1206, processor 1208, UE location requester 1214, UE context requestor 1216, handover controller 1218 and NAT 1220 is in electrical communication with the other features of the target TWAG 1200. The target TWAG 1200 can be implemented as a combination of computer hardware and software. In particular, the UE location requester 1214, a UE context requestor 1216 and a handover controller 1218 may be implemented as software configured to run on the processor 1208. The at least one memory 1206 stores the various programs/executable files that are implemented by a processor 1208, and also provides a storage unit for any required data. The programs/executable files stored in the memory 1206, and implemented by the processor 1208, can include the UE location requester 1214, a UE context requestor 1216, a handover controller 1218 and a NAT 1220, but are not limited to such.

Figure 13 shows a schematic diagram of a source TWAG 1300. The source TWAG 1300 may be used in the architecture shown in Figure 6a. It is noted that in addition to the features shown in Figure 13, the source TWAG 1300 may comprise the same or similar features as those in the target TWAG 1300 and vice-versa. Different features are shown in the TWAGs of Figures 12 and 13 for illustrative purposes. The source TWAG 1300 comprises a transmitter 1302 and a receiver 1304. The transmitter 1302 and receiver 1304 are in electrical communication with other communication units, nodes, UEs, servers and/or functions in a telecommunications network and are configured to transmit and receive data accordingly. The source TWAG 1300 further comprises at least one memory 1306 and at least one processor 1308. The memory 1306 may comprise a non-volatile memory and/or a volatile memory. The memory 1306 may have a computer program 1310 stored therein. The computer program 1310 may be configured to undertake the methods disclosed herein. The computer program 1310 may be loaded in the memory 1306 from a non-transitory computer readable medium 1312, on which the computer program is stored. The processor 1308 is configured to undertake at least the functions of UE context responder 1314 and an address obtainer 1316, as set out herein. Each of the transmitter 1302 and receiver 1304, memory 1306, processor 1308, UE context responder 1314 and address obtainer 1316 is in electrical communication with the other features of the source TWAG 1300. The source TWAG 1300 can be implemented as a combination of computer hardware and software. In particular, the UE context responder 1314 and address obtainer 1316 may be implemented as software configured to run on the processor 1308. The at least one memory 1306 stores the various programs/executable files that are implemented by a processor 1308, and also provides a storage unit for any required data. The programs/executable files stored in the memory 1306, and implemented by the processor 1308, can include the UE context responder 1314 and address obtainer 1316, but are not limited to such.

Figure 14 shows a schematic diagram of a locator function 1400. The locator function 1400 may be used in the architecture shown in Figure 6a.

The locator function 1400 comprises a transmitter 1402 and a receiver 1404. The transmitter 1402 and receiver 1404 are in electrical communication with other communication units, nodes, UEs, servers and/or functions in a telecommunications network and are configured to transmit and receive data accordingly.

The locator function 1400 further comprises at least one memory 1406 and at least one processor 1408. The memory 1406 may comprise a non-volatile memory and/or a volatile memory. The memory 1406 may have a computer program 1410 stored therein. The computer program 1410 may be configured to undertake the methods disclosed herein. The computer program 1410 may be loaded in the memory 1406 from a non-transitory computer readable medium 1412, on which the computer program is stored. The processor 1408 is configured to undertake at least the functions of a UE location responder 1414, as set out herein. Each of the transmitter 1402 and receiver 1404, memory 1406, processor 1408 and UE location responder 1414 is in electrical communication with the other features of the locator function 1400. The locator function 1400 can be implemented as a combination of computer hardware and software. In particular, the UE context responder 1414 and address obtainer 1416 may be implemented as software configured to run on the processor 1408. The at least one memory 1406 stores the various programs/executable files that are implemented by a processor 1408, and also provides a storage unit for any required data. The programs/executable files stored in the memory 1406, and implemented by the processor 1408, can include the UE location responder 1414, but are not limited to such.