Orchestrating paging in a communication network

Abstract

The present disclosure relates to a method (200) for orchestrating paging of a user equipment, UE, (101) connected to a communication network (100) The method (200) comprises identifying (201) a location of a mobility prediction model of the UE (101). The method (200) comprises obtaining (202) a predicted cell location of the UE (101), by executing the mobility prediction model in the identified location, using a last know cell location of the UE (101) as input to the mobility prediction model. The method (200) comprises initiating (203) paging in the predicated cell location of the UE (101). Further the disclosure relates to a network node (600) for orchestrating paging of the UE (101).

Description

[0001] ORCHESTRATING PAGING IN A COMMUNICATION NETWORK

[0002] TECHNICAL FIELD

[0003] The present disclosure relates to a method for orchestrating paging of a user equipment in a communication network. The present disclosure relates to a network node, a computer program and a computer program product to carry out the method for orchestrating paging of a user equipment in a communication network.

[0004] BACKGROUND

[0005] Wireless access networks and systems commonly referred to as fourth generation (4G) and fifth generation (5G) networks have been defined by the third-generation partnership project (3GPP). In these networks, wireless devices connect to a core network through access points called eNodeBs for 4G and gNodeBs for 5G, facilitating wireless access.

[0006] The 3GPP wireless access networks incorporate an idle mode for wireless devices that are not actively connected to the network. In the idle mode, the energy consumption of the wireless device is minimized, and network traffic associated with the device is low. When substantial data transmission is required, the device transitions from idle mode to a connected mode where it becomes more active.

[0007] During idle mode, when a wireless device such as user equipment (UE) needs to be reached by the network, the UE goes through a paging process. Paging occurs when the network needs to contact an idle UE, e.g., due to incoming data traffic or an incoming call. The network sends a signal to the device, requesting it to connect. However, the network typically does not have precise information about the device's location or which access point to use, so it starts from the device's last known position and expands the search area, including more access points, until the device receives the paging message and responds to the network.

[0008] Each access point provides network access to a particular area. A geographical area to which an access point (eNodeB or gNodeB) provides network access is referred to as a cell. The cells or the access points providing network access are organized into tracking areas (TA). Each TA is identified by a Tracking Area Code / ldentity (TAC / TAI). TAs are further grouped into Registration Areas (RAs). An RA is a group of TAs specifically grouped for a particular UE. RA is a UE-specific list of TAs reflecting the mobility pattern of the UE.

[0009] A wireless device is associated with a current TA, representing the last known location, that is, the TA where the wireless device was last seen, and also with a Tracking Area List (TAL) which comprises additional TAs close to the current TA which the wireless device may have entered without the knowledge of the network.

[0010] Paging is performed by Access and Mobility Function (AMF) in 5G networks and by Mobility Management Entity (MME) in 4G networks. Paging usually happens in stages, with each stage having its own configured algorithm. When paging happens, there is a 4-step paging process initiated (the paging process may be slightly different for 4G / 5G and may also be vendor dependent):

[0011] Step-1 : The network first checks if the UE responds to signaling in a cell to which it was last connected to. If the UE does not respond the network performs step 2.

[0012] Step-2: The network checks if the UE responds to signaling in a current TA. If the UE does not respond, the network continuous to step 3.

[0013] Step-3: The network now checks if the UE responds in any of the cells belonging to a TAL. If the UE does not respond the network does step 4.

[0014] Step-4: The network sends signals to all cells of the entire network.

[0015] When the UE is in IDLE mode it still listens to signaling. When it detects that the best cell has a different TAC than the one it is currently registered in, the UE makes a Tracking Area Update (4G) or a Registration Request (5G).

[0016] An operator can optimize its network by:

[0017] -grouping the cells into suitable small TAs. A small TA means fewer cells are signaled when paging a TA, but on the other hand there would be more signaling from UEs when they switch TA, and

[0018] -selecting the length of the TAL that is used in step 3. A shorter list means less signaling as fewer cells are a part of the TAL, but on the other hand, the likelihood of the UE not being found, and having to use step 4 increases.

[0019] Recent literature on paging discloses using mobility prediction models to predict an access point to which the UE may have likely been connected to or a list of access points to which the UE may have likely been connected to. Mobility prediction is based on trained machine learning models and uses previous location (cell) of the UE as input. A well performing mobility prediction model has proven to increase the performance of the paging process by increasing the likelihood of finding the UE within a predicted list of eNBs or gNBs. The predicted list may be shorter than the preconfigured list (TAL) of historic eNBs or gNBs in the standard algorithm.

[0020] US 2 020 275 410 A1 discloses operations of a network device analyzing a group of contextual data for a user equipment device, and mobility management behaviors historically implemented for the user equipment device and based on an indication that a page request is scheduled to be sent to the user equipment device. The operations including configuring a paging message for the user equipment device based on the contextual data and the mobility management behaviors.

[0021] Recent research also suggests using a mobility prediction model per International Mobile Subscriber Identifier (IMSI) / UE to predict mobility of the UE. Also, there is a possibility to use one mobility prediction model per group of UEs, where each UE is assigned to a group depending on some criteria.

[0022] The problem with current paging technology, both in 4G and 5G, is that mobility prediction is limited by simple models / heuristics. Predicting mobility using an ML model per UE / IMSI has proven to outperform simpler strategies, such as simply using a list of last visited cells. However, the increased accuracy comes at a price of increased latency in the paging process. The increased accuracy adds a level of complexity and requires more computational resources, for example, RAM, disk, and / or CPU cycles. Advanced machine learning models, such as convolutional and deep neural networks typically outperform simpler approaches like support vector machines (SVMs), but requires more resources both for training and inference. Typically, both training and inference is done using a centralized cloud service or a central database. Figure 1 shows multiple cells, and tracking areas comprising a group of cells, for example, Cells 1 , 2 and 3 comprise a TA. Figure 1 shows a system where models for a UE / IMSI are stored in a top database (Top db) and the base stations or access points (gNBs / eNBs) of the cells are retrieving mobility predictions from the top dB. Unfortunately, retrieving mobility predictions from the top dB every time induces latency and slows down the paging process. An efficient paging process is critical for network performance.

[0023] SUMMARY

[0024] An object of the invention is to enable paging functionality with reduced latency in a communication network so as to improve the performance of the communication network.

[0025] According to a first aspect of the invention, a method is provided for orchestrating paging of a user equipment in a communication network. The method comprises identifying a location of a mobility prediction model of the UE. The method comprises obtaining a predicted cell location of the UE, by executing the mobility prediction model in the identified location, using a last know cell location of the UE as input to the mobility prediction model. The method comprises initiating paging in the predicated cell location of the UE.

[0026] According to a second aspect of the invention, there is presented a network node in the communication network. The network node is configured to orchestrate paging of the UE in the communication network, The network node comprises processing circuitry configured to cause the network node to identify a location of a mobility prediction model of the UE. The processing circuitry is configured to obtain a predicted cell location of the UE, by executing the mobility prediction model in the identified location, using a last know cell location of the UE as input to the mobility prediction model. The processing circuitry is configured to initiate paging in the predicated cell location of the UE.

[0027] According to a third aspect of the invention, there is presented a computer program comprising instructions which when executed on a processor of a network node, causes the network node to perform a method according to any of the embodiments of the first aspect.

[0028] According to a fourth aspect of the invention, there is presented a computer program product which comprises a computer readable storage medium on which a computer program according to the third aspect is stored. Other objectives, features and advantages of the enclosed embodiments will be apparent from the following detailed disclosure, from the attached dependent claims as well as from the drawings.

[0029] Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to "a / an / the element, apparatus, component, means, module, action, etc." are to be interpreted openly as referring to at least one instance of the element, apparatus, component, means, module, action, etc., unless explicitly stated otherwise. The actions of any method disclosed herein do not have to be performed in the exact order disclosed, unless explicitly stated.

[0030] BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 shows multiple cells, and tracking areas comprising a group of cells.

[0032] Figure 2A schematically illustrates part of an LTE network.

[0033] Figure 2B schematically illustrates part of a 5G network.

[0034] Figure 3 is a flow chart illustrating a method for orchestrating paging of a UE in a communication network.

[0035] Figure 4 illustrates a distribution of databases.

[0036] Figure 5 is a flowchart illustrating one or more embodiments of the method to orchestrate paging the communication network.

[0037] Figure 6 depicts the network node.

[0038] DETAILED DESCRIPTION

[0039] To minimize the latency for getting a prediction, it would be beneficial to have the inference as close to the edge (base station or the user equipment (UE)) of the network as possible. This way, a request to perform mobility prediction, from the network, for a UE need not traverse to a central cloud or a top database (as in Figure 1 ), and similarly, the result of the mobility prediction need not traverse the central cloud to reach the base stations that actually perform the paging according to the result of the mobility prediction. The desire to do model inference for a UE at the edge, for example, in a base station (that may actually perform the paging) close to the UE, means that some entity must have knowledge about a suitable edge location for model inference. But on the other hand, predicted locations of the UE are not known before the model has been executed, therefore, suitable edge location, that is, a base station close to the UE is not known before the model inference.

[0040] The present disclosure proposes a distributed setup in the communication network to store mobility prediction models. The setup is such that the models are stored closer to the UE, that is, in base stations closer to the UE and inference is performed closer to the UE. The present disclosure proposes a method for orchestrating paging of a UE in the communication network. The proposed method for orchestrating paging enables executing mobility prediction models closer to the UE and thus, reducing latency in paging the UE. The proposed disclosure enables life cycle management of mobility prediction models for a UE.

[0041] Figures 2A and 2B illustrates a communication network 100. Communication network

[0042] 100 can be a long term evolution (LTE) network or 4thgeneration (4G) network. Communication network 100 can be a 5thgeneration (5G) network. Figure 2A schematically illustrates part of a LTE network, i.e., a 4G network. An access point in a 4G network is commonly referred to as an eNodeB 102. A mobility management entity (MME) 103 is a node in the LTE network which is responsible for idle mode wireless device paging and tagging procedures including retransmissions. It is involved in the bearer activation / deactivation process and is also responsible for choosing the serving gateway (SGW) 104 for a wireless device or a user equipment

[0043] 101 at the initial attach and at time of intra-LTE handover involving Core Network (CN) node relocation. The Non-Access Stratum (NAS) signaling terminates at the MME 103 and it is also responsible for generation and allocation of temporary identities to wireless devices.

[0044] Figure 2B shows some details of a 5G network. An access point in a 5G network is commonly referred to as a gNodeB 106. The gNodeB 106 is connected to a user plane function (UPF) which handle data flow to and from the user equipment 101 . A 5G Core Access and Mobility Management Function (AMF) 105 receives connection and session related information from the user equipment 101 and is responsible for handling connection and mobility management tasks. A 5G network may comprise many AMF instances, a Globally Unique AMF Identifier (GUAMI) is therefore employed. The user equipment 101 accessing the network specifies this in the first NAS message it sends to the network, which is routed to the required AMF 105 by the gNodeB. Applicable to both 3GPP access and non-3GPP access, the GlIAMI also ensures that messages from a wireless device, registered through both access networks, get forwarded to the same AMF. The Non-3GPP Interworking Function (N3IWF) is responsible for routing messages outside the 5G RAN.

[0045] Figure 3 is a flow chart illustrating a method, 200, for orchestrating paging of the UE 101 in the communication network 100. The method 200 comprises identifying, 201 , a location of a mobility prediction model of the UE 101. The mobility prediction model of the UE 101 may be designed in multiple ways. In its most simple form, it may take a previous location (cell) of the UE 101 as input and may produce a single location (cell) of the UE 101 as output. Alternatively, the output could be a list of cells in which the UE 101 may be present. Advanced mobility prediction models may take a history of previous locations (cells) as input. Other parameters could also be fed into the model as inputs, such as external events surrounding the UE 101- Toad closed’, ‘football game’, ‘holiday’, etc. The mobility prediction model may be a machine learning (ML) model.

[0046] Figure 4 illustrates a distribution of databases. In Figure 4, each cell may have a base station (gNodeB or eNodeB) providing network access to each of them. For example, Cell A may have a Base Station A providing network access to it. Base Station A may host a database which can be referred to as local database. The local database may store mobility prediction model of a UE. Base Station A may perform inference using the mobility prediction model of the UE.

[0047] Similar to Cell A, Cells B and C may have Base Stations B and C. Base Stations B and C may each host a local database. Local databases of Cells A, B and C may be associated with a regional database. Local database of Cell A may comprise mobility prediction models of all the UEs that are in communication with it. The regional database may comprise mobility prediction models of all the UEs that are in communication with Cell A. The regional database may comprise mobility prediction models of all the UEs that are in communication with Cell B. The regional database may comprise mobility prediction models of all the UEs that are in communication with Cell C. Each local database in the communication network may be associated with a regional database and vice versa. Top DB or top database may also be referred to as central database. The central database may comprise mobility prediction models of all the UEs in the communication network 100, as illustrated in Figures 2A and 2B. Therefore, reference numerals may be used in the following text to refer to elements as illustrated in Figures 2A and 2B and discussed in the descriptions relating thereto.

[0048] Identifying the location of the mobility prediction model of the UE 101 may comprise checking whether the mobility prediction model of the UE 101 is present in a first local database associated with the UE 101. The first local database may refer to a local database present in a last known cell location of the UE 101 . The first local database may be hosted by the eNodeB node 102 providing network access to the last know cell location of the UE 101. The first local database may be hosted by the gNodeB node 106 providing network access to the last know cell location of the UE 101 .

[0049] Identifying the location of the mobility prediction model of the UE 101 may comprise checking whether the mobility prediction model of the UE 101 is present in a first regional database. The first regional database may be associated with the first local database. Identifying the location of the mobility prediction model of the UE 101 may comprise checking whether the mobility prediction model of the UE 101 is present in the first regional database when the mobility prediction model is not present in the first local database.

[0050] Identifying the location of the mobility prediction model of the UE 101 comprises checking whether the mobility prediction model of the UE 101 is present in a central database when the mobility prediction model is not present in the first local and regional databases.

[0051] Once the location of the mobility prediction model of the UE 101 is identified, the mobility prediction model may be executed to obtain prediction(s) regarding the location of the UE 101. The method 200, as illustrated in Figure 3 and discussed in the descriptions relating thereto, comprises obtaining, 202, a predicted cell location of the UE 101 , by executing the mobility prediction model in the identified location, using the last know cell location of the UE 101 as input to the mobility prediction model. The mobility prediction model, when executed, may use different inputs depending on the location where the model is identified. For example, the mobility prediction model may only use a longer history of previous cell locations when identified and executed at the regional database compared to the local database. The mobility prediction model may only use a longer history of previous cell locations when identified and executed at the central database compared to the local and regional databases.

[0052] The method 200 comprises initiating, 203, paging in the predicted cell location of the UE 101. Initiating paging may comprise signaling an eNodeB node and / or a gNodeB node, providing network access to the predicted cell location, to page the predicted cell location to check whether the UE is actually present in the predicted cell location.

[0053] The method 200 may be performed by a network node comprising Mobility Management Entity, MME, 103 in a 3GPP defined wireless communication network 100.

[0054] The method 200 may be performed in a network node implementing an Access and Mobility Function, AMF, 105 in a 3GPP defined wireless communication network 100.

[0055] The method 200 may comprise storing, 204, the predicted cell location in the network node as a current cell location of the UE 101 upon receiving a confirmation, from the UE 101 , indicating that the UE 101 is present in the predicted cell location when paged in the predicted cell location.

[0056] The UE 101 refers to a device capable, configured, arranged and / or operable to communicate wirelessly with network nodes and / or other UEs in a communication network 100. Examples of a UE include, but are not limited to, a smart phone, mobile phone, cell phone, voice over IP (VoIP) phone, wireless local loop phone, personal digital assistant (PDA). Other examples include any UE identified by the 3rd Generation Partnership Project (3GPP), including a narrow band internet of things (NB-loT) UE, a machine type communication (MTC) UE, and / or an enhanced MTC (eMTC) UE. A UE in the form of an Internet of Things (loT) device may be a device for use in one or more application domains, these domains comprising, but not limited to, home, city, wearable technology, extended reality, industrial application, and healthcare.

[0057] By way of example, the loT device for a home, an office, a building or an infrastructure may be a baking scale, a coffee machine, a grill, a fridge, a refrigerator, a freezer, a microwave oven, an oven, a toaster, a water tap, a water heater, a water geyser, a sauna, a vacuum cleaner, a washer, a dryer, a dishwasher, a door, a window, a curtain, a blind, a furniture, a light bulb, a fan, an air-conditioner, a cooler, an air purifier, a humidifier, a speaker, a television, a laptop, a personal computer, a gaming console, a remote control, a vent, an iron, a steamer, a pressure cooker, a stove, an electric stove, a hair dryer, a hair styler, a mirror, a printer, a scanner, a photocopier, a projector, a hologram projector, a 3D printer, a drill, a hand-dryer, an alarm clock, a clock, a security camera, a smoke alarm, a fire alarm, a connected doorbell, an electronic door lock, a lawnmower, a thermostat, a plug, an irrigation control device, a flood sensor, a moisture sensor, a motion detector, a weather station, an electricity meter, a water meter, and a gas meter.

[0058] By further ways of example, the loT device for use in a city, urban, or rural areas may be connected street lighting, a connected traffic light, a traffic camera, a connected road sign, an air control / monitor, a noise level detector, a transport congestion monitoring device, a transport controlling device, an automated toll payment device, a parking payment device, a sensor for monitoring parking usage, a traffic management device, a digital kiosk, a bin, an air quality monitoring sensor, a bridge condition monitoring sensor, a fire hydrant, a manhole sensor, a tarmac sensor, a water fountain sensor, a connected closed circuit television, a scooter, a hoverboard, a ticketing machine, a ticket barrier, a metro rail, a metro station device, a passenger information panel, an onboard camera, and other connected device on a public transport vehicle.

[0059] As further way of example, the communication loT device may be a wearable device, or a device related to extended reality, wherein the device related to extended reality may be a device related to augmented reality, virtual reality, merged reality, or mixed reality. Examples of such loT devices may be a smart-band, a tracker, a haptic glove, a haptic suit, a smartwatch, clothes, eyeglasses, a head mounted display, an ear pod, an activity monitor, a fitness monitor, a heart rate monitor, a ring, a key tracker, a blood glucose meter, and a pressure meter.

[0060] As further ways of example, the loT device may be an industrial application device wherein an industrial application device may be an industrial unmanned aerial vehicle, an intelligent industrial robot, a vehicle assembly robot, and an automated guided vehicle.

[0061] As further ways of example, the loT device may be a transportation vehicle, wherein a transportation vehicle may be a bicycle, a motor bike, a scooter, a moped, an auto rickshaw, a rail transport, a train, a tram, a bus, a car, a truck, an airplane, a boat, a ship, a ski board, a snowboard, a snow mobile, a hoverboard, a skateboard, rollerskates, a vehicle for freight transportation, a drone, a robot, a stratospheric aircraft, an aircraft, a helicopter and a hovercraft.

[0062] Figure 5 is a flowchart illustrating one or more embodiments of the method 200 to orchestrate paging the communication network 100.

[0063] In step 301 , the network node initiates paging in the last known cell location of the UE 101. If the UE 101 is found, the cell location where the UE 101 is found is updated as the current cell location of the UE 101 in step 302. If the UE 101 is not found, in step 303, the network node checks whether the mobility prediction model of the UE 101 is present in the first local database associated with the UE 101 . The first local database may refer to the local database present in the last known cell location of the UE 101 . If the mobility prediction model is found in the first local database, the model is executed, in step 304, in the first local database to obtain a predicted cell location of the UE 101. In step 305, the network node may initiate paging in the predicted cell location of the UE 101. If the UE 101 is found, the predicted cell location where the UE 101 is found is updated as the current cell location of the UE 101 in step 306. In step 306, the mobility prediction model may be copied from the identified location and stored in a second local database associated with the current cell location of the UE 101 . The identified location may be the first local database. The local database in the current cell location of the UE 101 may be referred to as the second local database. The mobility prediction model may be copied from the identified location and stored in a second regional database associated with the second local database.

[0064] If the UE 101 is not found in step 305 or if the mobility prediction model is not found in step 303, then in step 307, the network node checks whether the mobility prediction model of the UE 101 is present in the first regional database. If the mobility prediction model is found in the first regional database, the model is executed, in step 308, by the first regional database to obtain a predicted cell location of the UE 101. In step 309, the network node may initiate paging in the predicted cell location of the UE 101. If the UE 101 is found, the predicted cell location where the UE 101 is found is updated as the current cell location of the UE 101 in step 310. In step 310, the mobility prediction model may be copied from the identified location and stored in the second local database associated with the current cell location of the UE 101. The mobility prediction model may be copied from the identified location and stored in the second regional database associated with the second local database. The identified location may be the first regional database.

[0065] If the UE 101 is not found in step 309 or if the mobility prediction model is not found in step 307, then, in step 311 , the network node checks whether the mobility prediction model of the UE 101 is present in the central database. If the mobility prediction model is found in the central database, the model may be executed, in step 312, in the central database to obtain a predicted cell location of the UE 101. In step 313, the network node may initiate paging in the predicted cell location of the UE 101. If the UE 101 is found, the predicted cell location where the UE 101 is found is updated as the current cell location of the UE 101 in step 314. In step 314, the mobility prediction model may be copied from the identified location and stored in the second local database associated with the current cell location of the UE 101. The mobility prediction model may be copied from the identified location and stored in the second regional database associated with the second local database. The identified location may be the central database. If the UE 101 is not found in the central database, the network node may perform global paging to obtain the current cell location of the UE 101. Global paging may refer to signaling all the eNodeBs and / or gNodeBs in the communication network 100 to page the UE 101. If the UE 101 is found in step 315, the mobility prediction model may be updated in step 316. Updating the mobility prediction model may comprise re-training the mobility prediction model of the UE 101. In step 319, the updated mobility prediction model may be saved to the central database, the updated mobility prediction model may be copied and stored to the second local and regional databases associated with the current cell location of the UE 101.

[0066] If the mobility predication model is not present in the central database in step 311 , the network node may perform global paging to obtain the current cell location of the UE 101. If the UE 101 is found in step 317, a mobility prediction model for the UE 101 may be created in step 318. In step 319, the updated mobility prediction model may be saved to the central database, the updated mobility prediction model may be copied and stored inthe second local and regional databases associated with the current cell location of the UE 101 . In an embodiment, the mobility prediction model of the UE 101 may be removed from the identified location after the mobility prediction model of the UE 101 is copied and stored in the second local and regional databases associated with current location of the UE 101.

[0067] In an embodiment, the method 200 may comprise identifying the location of the mobility prediction model of the UE 101 using an umbrella model. The umbrella model may refer to a mobility prediction model that is common the communication network 100. The umbrella model may not be specific to the UE 101. For any UE 101 , the umbrella model may use a previous cell location of the UE 101 as an input to predict a coarser location of the UE 101. The UE 101 may be found in the predicted coarser location when paged or the network node may check whether the mobility prediction for the UE 101 is present in a local database in the predicted coarser location. The umbrella model may be executed in the central database.

[0068] By copying and storing the mobility prediction model of the UE 101 to the second local and regional databases associated with the current cell location of the UE 101 , the method 200 updates the location of the mobility prediction model of the UE 101. By creating and updating the mobility prediction model of the UE 101 , the method 200 enables an availability of mobility prediction model of the UE 101 for the next instance of paging. By updating the location of the mobility prediction model of the UE 101 and by making sure that the mobility prediction model of the UE 101 is available in the second local and regional databases associated with the current cell location of the UE 101 , the method 200 reduces the latency associated with searching for the mobility prediction model of the UE 101 in the communication network 100. By making sure that the mobility prediction model of the UE 101 is copied and stored to the local database, the method 200 enables the execution of the mobility prediction model closer to the UE 101 and thus, reducing latency in the paging process.

[0069] Figure 6 depicts the network node 600. The network node 600 comprises a processing circuitry 601 . The processing circuitry may comprise one or more processors 602, such as Central Processing Units (CPUs), microprocessors, application processors, application-specific processors, Graphics Processing Units (GPUs), and Digital Signal Processors (DSPs), or a combination thereof, and a memory 603 comprising a computer program 604 comprising instructions. The computer program 604 may be downloaded to the memory 603 as a data carrier signal carrying the computer program 604. When executed by the processor(s) 602, the instructions cause the network node 600 to become operative in accordance with embodiments of the invention described above. The memory 603 may, e.g., be a Random-Access Memory (RAM), a Read- Only Memory (ROM), a Flash memory, or the like. The processing circuitry 601 may alternatively or additionally comprise one or more Application-Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or the like, which are operative to cause the MR device 100 to become operative in accordance with embodiments of the invention described above. A computer program product 605 comprises a computer readable storage medium, such as the memory 603, on which a computer program 601 is stored.

Claims

CLAIMS1. A method (200) for orchestrating paging of a user equipment, UE, (101 ) connected to a communication network (100), the method (200) comprising:- identifying (201 ) a location of a mobility prediction model of the UE (101 );- obtaining (202) a predicted cell location of the UE (101 ), by executing the mobility prediction model in the identified location, using a last know cell location of the UE (101 ) as input to the mobility prediction model; and- initiating (203) paging in the predicated cell location of the UE (101 ).

2. The method (200) of claim 1 , wherein initiating (203) paging comprises signaling an eNodeB node and / or a gNodeB node, providing network access to the predicted cell location, to page the predicted cell location to check whether the UE (101 ) is actually present in the predicted cell location.

3. The method (200) of claims 1 -2, comprising: storing (204, 306, 310, 314) the predicted cell location as a current cell location of the UE (101 ) upon receiving a confirmation, from the UE (101 ), indicating that the UE (101 ) is present in the predicted cell location when paged in the predicted cell location.

4. The method (200) of claims 1 -3, wherein identifying (201 ) the location of the mobility prediction model of the UE (101 ) comprises checking (303) whether the mobility prediction model of the UE (101 ) is present in a first local database associated with the UE (101 ).

5. The method (200) of claim 4, wherein the first local database is present in a last known cell location of the UE (101 ).

6. The method (200) of claims 4-5, wherein the first local database is hosted by an eNodeB node and / or gNodeB node, providing network access to the last know cell location of the UE (101 ).

7. The method (200) of claims 1 -6, wherein identifying the location of the mobility prediction model of the UE (101 ) comprises checking (307) whether the mobility prediction model of the UE (101 ) is present in a first regional database associated with the first local database when the mobility prediction model is not present in the first local database.

8. The method (200) of claims 1 -7, wherein identifying the location of the mobility prediction model of the UE (101 ) comprises checking (311 ) whether the mobility prediction model of the UE (101 ) is present in a central database when the mobility prediction model is not present in the first local and regional databases.

9. The method (200) according to any one of claims 1 -8, comprising: copying and storing (306, 310, 314) the mobility prediction model from the identified location to a second local database associated with the current cell location of the UE.

10. The method (200) according to any one of claims 1 -9, comprising: copying and storing (306, 310, 314) the mobility prediction model from the identified location to a second regional database associated with the second local database.11 . The method (200) of any one of claims 1 -10, comprising:- performing (315) global paging when the UE (101 ) is not present in the predicted cell location obtained by executing the mobility prediction model in the central database; and- updating (316) the mobility prediction model in the central database.

12. The method (200) of any one of claims 1 -12, comprising:- performing (317) global paging to obtain the current location cell of the UE (101 ) when the mobility prediction model is not present in the central database;- creating (318) a new mobility prediction model for the UE (101 ); and- storing (319) the new mobility prediction model to the central database and to the second regional and local databases associated with the current location of the UE (101 ).

13. The method (200) of any one of claims 1-12, wherein the mobility prediction model is a machine learning (ML) model.

14. The method (200) of any one of claims 1 -13, comprising:- identifying the location of the mobility prediction model of the UE using an umbrella model.

15. The method (200) of claim 14, wherein the umbrella model is executed in the central database.

16. The method (200) according to any one of claims 1 -15, performed in a Mobility Management Entity, MME, of a third generation partnership, 3GPP, defined wireless access network.

17. The method (200) according to any one of claims 1 -15, performed in a network node (600) implementing an Access and Mobility Function, AMF, of a third generation partnership, 3GPP, defined wireless access network.

18. A network node (600) for orchestrating paging of a user equipment, UE, (101 ) in a communication network (100), the network node (600) comprising processing circuitry (602) configured to cause the network node (600) to:- identify a location of a mobility prediction model of the UE (101 );- obtain a predicted cell location of the UE (101 ), by executing the mobility prediction model in the identified location, using a last know cell location of the UE (101 ) as input to the mobility prediction model; and- initiate paging in the predicated cell location of the UE (101 ).

19. The network node (600) as claimed in claim 18, wherein the processing circuitry (602) is further configured to cause the UE (101 ) to perform a method according to any one of claims 2 to 17.

20. The network node (600) according to any one of claims 18-19, wherein the network node (600) comprises a mobility management entity, MME, (103).

21. The network node (600) according to any one of claims 18-19, wherein the network node (600) is configured to implement an Access and Mobility Function (105), AMF.

22. A computer program (604), comprising instructions which when executed on a processor (602) of a network node (600) causes the network node (600) to perform a method according to any one of claims 2 to 17.

23. A computer program product (603) which comprises a computer readable storage medium on which a computer program (604) according to claim 22 is stored.

Images