Method and apparatus for ai / ML model transfer / delivery for wireless communication systems
The AI/ML model transfer/delivery method in wireless communication systems addresses inefficiencies by enabling proactive and reactive model updates through UE, RAN, or OTT server triggers, ensuring timely and adaptive model delivery in response to scenario and configuration changes.
Patent Information
- Application Number
- PCT/CN2023/131322
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-11-13
- Publication Date
- 2025-05-22
AI Technical Summary
Existing wireless communication systems face challenges in efficiently transferring and delivering AI/ML models to user equipment (UE) in response to changes in scenarios, configurations, or sites, leading to potential latency and inefficiencies.
The method and apparatus for AI/ML model transfer/delivery in wireless communication systems involve triggering model transfer requests from UE, RAN, or OTT server, and facilitating the transfer of updated models through a model transfer/delivery tunnel, which can be initiated proactively or reactively.
This approach enables efficient and timely delivery of AI/ML models, reducing latency and improving system performance by allowing for adaptive model updates based on changing scenarios and configurations.
Smart Images

Figure CN2023131322_22052025_PF_FP_ABST
Abstract
Description
METHOD AND APPARATUS FOR AI / ML MODEL TRANSFER / DELIVERY FOR WIRELESS COMMUNICATION SYSTEMSFIELD
[0001] The present disclosure relates generally to communication systems, and more particularly, the method and apparatus for AI / ML model transfer and delivery for wireless communication systems.BACKGROUND
[0002] The integration of Artificial Intelligence (AI) into 3GPP and wireless technology has ushered in a new era of potential advancements and efficiencies in 5G and future 6G. In the development of AI / ML algorithms for wireless technology, it is important to tailor them to specific scenarios, locations, configurations, and deployments. This customization allows for better performance, as a one-size-fits-all approach may not be optimal. AIML algorithms can be updated through model changes, indicating the need for flexibility and adaptability in the application of AI / ML models. When AI models are designed for specific scenarios, configurations, or sites, they can be selected or activated for inference when applicable.
[0003] Model transfer / delivery is the process that enables the availability of an AI / ML model at the UE side. It becomes necessary when there is no existing AI / ML model at the UE that is applicable to the relevant scenario, configuration, or site. The term 's cenarios'could signify a range of conditions, including various deployment scenarios, different distributions of outdoor or indoor UE, a variety of UE mobility levels, or a range of carrier frequencies. Other aspects of scenarios are not excluded. Configurations might stand for parameters such as different UE settings, an assortment of gNB settings, a variety of bandwidths (like 10MHz, 20MHz) , diverse antenna port layouts (for instance, N1 / N2 / P) or different numbers of antenna ports (such as 32 ports, 16 ports) , among others. Various use cases might emphasize different elements of the scenarios and configurations. In such instances, the UE needs to download the AI / ML model trained for the specific scenario, configuration, or site.
[0004] In 3GPP meetings, it has been agreed that model transfer / delivery can be initiated reactively, meaning that an AI / ML model is downloaded when needed due to changes in scenarios, configurations, or sites. Alternatively, if the UE has the capability to store multiple AI / ML models, it can pre-download the models, allowing for model switching when changes occur. This supports proactive model transfer / delivery approaches, which have significantly shorter latency compared to reactive approaches.
[0005] In this invention, apparatus and mechanisms are sought to perform AI / ML model transfer / delivery for wireless communication systems.SUMMARY
[0006] The following presents a simplified summary of one or more aspects in order to provide a basic understanding of such aspects. This summary is not an extensive overview of all contemplated aspects, and is intended to neither identify key or critical elements of all aspects nor delineate the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form as a prelude to the more detailed description that is presented later.
[0007] In an aspect of the disclosure, a method, a computer-readable medium, and an apparatus are provided. The apparatus may be a UE. The overall model transfer / delivery procedure may contain the procedure of model transfer / delivery triggering and model transfer / delivery via model transfer / delivery tunnel. The model transfer / delivery triggering is to request model transfer / delivery to OTT server of UE. In different embodiments, it can initiate at UE, RAN, or OTT server of UE. In one embodiment, the UE detects the site, scenario or radio environment change, and UE triggers a model transfer request to its OTT server. In one embodiment, the RAN detects the UE which connected to itself changes the site, scenario or radio environment, or RAN changes the configuration, then RAN triggers a model transfer request to UE’s OTT server. In one embodiment, the OTT server updates the model for one scenario and initialize the model transfer procedure by itself, then indicate UE and network. In different embodiments, the model transfer triggering is initiated proactively or reactively.
[0008] The model transfer / delivery procedure is to transfer the updated model from OTT server to UE via model transfer / delivery tunnel. In different embodiments, the model transfer tunnel is from OTT server to UE via one user plane tunnel or several user plane / control plane tunnels.
[0009] To the accomplishment of the foregoing and related ends, the one or more aspects comprise the features hereinafter fully described and particularly pointed out in the claims. The following description and the annexed drawings set forth in detail certain illustrative features of the one or more aspects. These features are indicative, however, of but a few of the various ways in which the principles of various aspects may be employed, and this description is intended to include all such aspects and their equivalents.BRIEF DESCRIPTION OF THE DRAWINGS
[0010] Figure 1 illustrates a schematic system diagram illustrating an exemplary wireless network in accordance with embodiments of the current invention.
[0011] Figure 2 illustrate exemplary overall flow for the interface and tunnel of / AI / ML model transfer / delivery in accordance with embodiments of the current invention.
[0012] Figure 3 illustrate exemplary diagrams of model transfer triggering initiate by UE in accordance with embodiments of the current invention.
[0013] Figure 4 illustrate exemplary diagrams of model transfer triggering initiate by RAN in accordance with embodiments of the current invention.
[0014] Figure 5 illustrate exemplary diagrams of model transfer triggering initiate by OTT server in accordance with embodiments of the current invention.
[0015] Figure 6 illustrate exemplary diagrams of model transfer / delivery tunnel in accordance with embodiments of the current invention.
[0016] Figure 7 illustrate an exemplary overall flow to perform model transfer / delivery triggering and model transfer / delivery procedure in accordance with embodiments of the current invention.DETAILED DESCRIPTION
[0017] The detailed description set forth below in connection with the appended drawings is intended as a description of various configurations and is not intended to represent the only configurations in which the concepts described herein may be practiced. The detailed description includes specific details for the purpose of providing a thorough understanding of various concepts. However, it will be apparent to those skilled in the art that these concepts may be practiced without these specific details. In some instances, well known structures and components are shown in block diagram form in order to avoid obscuring such concepts.
[0018] Several aspects of telecommunication systems will now be presented with reference to various apparatus and methods. These apparatus and methods will be described in the following detailed description and illustrated in the accompanying drawings by various blocks, components, circuits, processes, algorithms, etc. (collectively referred to as “elements” ) . These elements may be implemented using electronic hardware, computer software, or any combination thereof. Whether such elements are implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system.
[0019] Aspects of the present disclosure provide methods, apparatus, processing systems, and computer readable mediums for NR (new radio access technology, or 5G technology) , 6G technology or other radio access technology. NR may support various wireless communication services. These services may have different quality of service (QoS) requirements e.g. latency and reliability requirements.
[0020] Figure 1 illustrates a schematic system diagram illustrating an exemplary wireless network in accordance with embodiments of the current invention. Wireless system includes one or more fixed base infrastructure units forming a network distributed over a geographical region. The base unit may also be referred to as an access point, an access terminal, a base station, a Node-B, an eNode-B, a gNB, or by other terminology used in the art. As an example, base stations serve a number of mobile stations within a serving area, for example, a cell, or within a cell sector. In some systems, one or more base stations are coupled to a controller forming an access network that is coupled to one or more core networks. gNB 1and gNB 2 are base stations in NR or future technology, the serving area of which may or may not overlap with each other. As an example, UE1 or mobile station is only in the service area of gNB 1 and connected with gNB1. UE1 is connected with gNB1 only, gNB1 is connected with gNB 2 and 3 via Xn interface. UE2 is in the overlapping service area of gNB1 and gNB2. The core network may also be referred to as an CN or other terminology used in the art. It may further refer to AMF or UPF or OAM entity or any of the combination of them.
[0021] Figure 1 further illustrates simplified block diagrams for UE2 and gNB2, respectively. UE has an antenna, which transmits and receives radio signals. A RF transceiver, coupled with the antenna, receives RF signals from antenna, converts them to baseband signal, and sends them to processor. In one embodiment, the RF transceiver may comprise two RF modules (not shown) . A first RF module is used for transmitting and receiving on one frequency band, and the other RF module is used for different frequency bands transmitting and receiving which is different from the first transmitting and receiving. RF transceiver also converts received baseband signals from processor, converts them to RF signals, and sends out to antenna. Processor processes the received baseband signals and invokes different functional modules to perform features in UE. Memory stores program instructions and data to control the operations of mobile station. UE also includes multiple function modules that carry out different tasks in accordance with embodiments of the current invention. An application layer, which receives and transmits the signaling and data in application layer between UE and OTT server.
[0022] A RRC config controller, which controls the RRC message transit / receive between UE and gNB.
[0023] A state controller, which controls UE RRC state according to network’s command and UE conditions. RRC supports the following states, RRC_IDLE, RRC_CONNECTED and RRC_INACTIVE.
[0024] A DRB controller, which controls to establish / add, reconfigure / modify and release / remove a DRB based on different sets of conditions for DRB establishment, reconfiguration and release.
[0025] A protocol stack controller, which manage to add, modify or remove the protocol stack for the DRB. The protocol Stack includes SDAP, PDCP, RLC, MAC and PHY layers.
[0026] A model transfer / delivery controller, which controls the model transfer / delivery controller behavior, including model transfer / delivery triggering, model transfer / delivery tunnel setup and model transfer / delivery procedure.
[0027] Similarly, gNB2 has an antenna, which transmits and receives radio signals. A RF transceiver, coupled with the antenna, receives RF signals from antenna, converts them to baseband signals, and sends them to processor. RF transceiver also converts received baseband signals from processor, converts them to RF signals, and sends out to antenna. Processor processes the received baseband signals and invokes different functional modules to perform features in gNB2. Memory stores program instructions and data to control the operations of gNB2.
[0028] Figure 1 further illustrates simplified block diagrams for OTT server. In one example, the OTT server is an UE-sided OTT server. In another example, the OTT server is a network-sided OTT server or a neutral site OTT server. The OTT server may have functional blocks that are similar to those of UE and gNB, which are used to interface with UE and gNB and handle the transmission and reception of control signaling, collected data and assistance information. The OTT server may include a model transfer / delivery controller, which interface with equivalent functional blocks on the UE side or the network side. In one example, the network entity / node / function in 5GS is DCAF. In one embodiment, the network entity / node / function in 5GS is NWDAF. In one embodiment, the network entity / node / function in 5GS is OAM.. In one embodiment, the network node / entity / function (e.g., DCAF, CN, OAM, etc. ) also request model transfer / delivery to the OTT server.
[0029] Figure 2 illustrate exemplary overall flow for the interface and tunnel of / AI / ML model transfer / delivery in accordance with embodiments of the current invention. The dataflow is between OTT server and UE, and multiple CP (control plane) / UP (user plane) tunnel may be used for the model transfer / delivery and signaling. The overall procedure may contain model transfer / delivery triggering (i.e., request to OTT server for model transfer / delivery from UE, RAN or initiate by OTT server itself) , setup model transfer / delivery tunnel (from OTT server to UE) , further contains the signaling over interface on Un, Xn, NG, etc. model transfer / delivery procedure, to delivery / transfer updated model from OTT server to UE.
[0030] Figure 3 illustrate exemplary diagrams of model transfer triggering initiate by UE in accordance with embodiments of the current invention. In one embodiment, the UE detects the site, scenario or radio environment change, and the model transfer / delivery triggering is initiate from UE to its OTT server. The term 'scenarios' could signify a range of conditions, including various deployment scenarios, different distributions of outdoor or indoor UE, a variety of UE mobility levels, or a range of carrier frequencies. Other aspects of scenarios are not excluded. Configurations might stand for parameters such as different UE settings, an assortment of gNB settings, a variety of bandwidths (like 10MHz, 20MHz) , diverse antenna port layouts (for instance, N1 / N2 / P) or different numbers of antenna ports (such as 32 ports, 16 ports) , among others. In one embodiment, the OTT server further send a model transfer / delivery indication to UE and / or network after receiving the model transfer / delivery request from UE. In these steps, the delivery tunnel can be CP / UP tunnel.
[0031] In one embodiment (as shown in sub-figure①) , the model transfer / delivery request is delivered from UE to RAN node, then further delivered to 5GS / OAM, and then further delivered to OTT server. In these steps, the delivery tunnel can be CP / UP tunnel. The model transfer / delivery request from UE to RAN can also be delivered via legacy measurement procedure (e.g., SON / MDT, UE measurement report) . In one embodiment, the model transfer / delivery request from UE to RAN can also be delivered through RRC / MAC / PHY signaling.
[0032] In one embodiment (as shown in sub-figure②) , the model transfer / delivery request is delivered from UE to 5GS / OAM, and further delivered to OTT server. In these steps, the delivery tunnel can be CP / UP tunnel. In one embodiment, 5GS / OAM further indicates RAN for the model transfer / delivery preparation (e.g., to setup model transfer / delivery tunnel) .
[0033] In one embodiment (as shown in sub-figure③) , the model transfer / delivery request is delivered from UE to OTT server via dataflow. In this case, the model transfer / delivery request may be performed without RAN / 5GC / OAM awareness.
[0034] Figure 4 illustrate exemplary diagrams of model transfer triggering initiate by RAN in accordance with embodiments of the current invention. In one embodiment, the RAN detects the UE which connected to itself changes the site, scenario or radio environment, or RAN changes the configuration, then the model transfer triggering is initiate from RAN to UE’s OTT server. In one embodiment, the OTT server further send a model transfer / delivery indication to UE and / or network for the confirmation.
[0035] In one embodiment (as shown in sub-figure①) , the model transfer / delivery request is delivered from RAN to 5GS / OAM, then further delivered to OTT server. In these steps, the delivery tunnel can be CP / UP tunnel. In one embodiment, RAN further indicates UE for the model transfer / delivery preparation (e.g., to setup model transfer / delivery tunnel) .
[0036] In one embodiment (as shown in sub-figure②) , RAN indicates UE for the model transfer / delivery indication, and UE further request a model transfer / delivery to its OTT server. In one embodiment, this transfer / delivery request from UE to OTT server is delivery via dataflow.
[0037] In one embodiment (as shown in sub-figure③) , the model transfer / delivery request is delivered from RAN to UEs OTT server via CP or UE tunnel. In one embodiment, RAN further indicates UE for the model transfer / delivery preparation. In one embodiment, UE further request a model transfer / delivery to its OTT server after receives the indication from RAN. In one embodiment, the OTT server further send a model transfer / delivery indication to UE after receives the request from RAN.
[0038] Figure 5 illustrate exemplary diagrams of model transfer triggering initiate by OTT server in accordance with embodiments of the current invention. In one embodiment, the OTT server updates the model for one scenario and initialize the model transfer procedure proactively by itself, then OTT server send model transfer / delivery indication to UE and network.
[0039] In one embodiment (as shown in sub-figure①) , the model transfer / delivery indication is delivered from OTT server to RAN, then further delivered to UE. In one embodiment (as shown in sub-figure②) , the model transfer / delivery indication is delivered from OTT server to 5GS / OAM, then further delivered to RAN, then further delivered to UE. In one embodiment (as shown in sub-figure③) , the model transfer / delivery indication is delivered from OTT server to 5GS / OAM, then further delivered to UE. In these steps in sub-figure①, ② and③, the delivery tunnel can be CP / UP tunnel. In one embodiment (as shown in sub-figure④) , the model transfer / delivery indication is delivered from OTT server to UE via dataflow. In this case, the model transfer / delivery indication may be delivered without RAN / 5GC / OAM awareness. In one embodiment, the application layer of UE informs the request to RRC layer, then the UE sends model transfer / delivery request to RAN / 5GC / OAM.
[0040] Figure 6 illustrate exemplary diagrams of model transfer / delivery tunnel in accordance with embodiments of the current invention. The model transfer / delivery tunnel is setup after model transfer / delivery triggering initiated by UE / RAN / OTT server. In one embodiment (as shown in sub-figure ①) , the model transfer / delivery tunnel is from OTT server to UE via dataflow. In this case, the model transfer / delivery may be performed without RAN / 5GC / OAM awareness. In one embodiment (as shown in sub-figure②) , the model transfer / delivery tunnel is from OTT server to RAN, then from RAN to UE. In one embodiment (as shown in sub-figure③) , the model transfer / delivery tunnel is from OTT server to 5GC / OAM, then from 5GC / OAM to UE. In one embodiment (as shown in sub-figure④) , the model transfer / delivery tunnel is from OTT server to 5GC / OAM, then from 5GC / OAM to RAN, then from RAN to UE. In these steps in sub-figure ② , ③ and ④, the model transfer / delivery tunnel can be CP / UP tunnel.
[0041] Figure 7 illustrate an exemplary overall flow to perform model transfer / delivery triggering and model transfer / delivery procedure in accordance with embodiments of the current invention. In the beginning, the UE detects the site, scenario or radio environment change, and the model transfer / delivery triggering is initiate from UE to its OTT server. The model transfer / delivery request is delivered from UE to RAN node, then further delivered to 5GS / OAM, and then further delivered to OTT server. Just as shown and described in Figure 3 sub-figure①. In other embodiments, other model transfer / delivery triggering as described in Figure 3, 4, 5 is used.
[0042] Model transfer / delivery procedure is performed from OTT server to UE via model transfer / delivery tunnel. In one embodiment, the model transfer / delivery tunnel is from OTT server to UE via dataflow. In other embodiments, other forms of transfer / delivery tunnel as described in Figure 6 is used. UE receives the updated model from OTT server via model transfer / delivery tunnel. In one embodiment, UE further request model identification to network (RAN, 5GC or OAM) after receiving the updated model. In one embodiment, UE further reports the updated UE capability for the updated model to the network (RAN, 5GC or OAM) .
[0043] It is understood that the specific order or hierarchy of blocks in the processes / flowcharts disclosed is an illustration of exemplary approaches. Based upon design preferences, it is understood that the specific order or hierarchy of blocks in the processes / flowcharts may be rearranged. Further, some blocks may be combined or omitted. The accompanying method claims present elements of the various blocks in a sample order, and are not meant to be limited to the specific order or hierarchy presented.
[0044] The previous description is provided to enable any person skilled in the art to practice the various aspects described herein. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects. Thus, the claims are not intended to be limited to the aspects shown herein, but is to be accorded the full scope consistent with the language claims, wherein reference to an element in the singular is not intended to mean “one and only one” unless specifically so stated, but rather “one or more. ” The word “exemplary” is used herein to mean “serving as an example, instance, or illustration. ” Any aspect described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other aspects. Unless specifically stated otherwise, the term “some” refers to one or more. Combinations such as “at least one of A, B, or C, ” “one or more of A, B, or C, ” “at least one of A, B, and C, ” “one or more of A, B, and C, ” and “A, B, C, or any combination thereof” include any combination of A, B, and / or C, and may include multiples of A, multiples of B, or multiples of C. Specifically, combinations such as “at least one of A, B, or C, ” “one or more of A, B, or C, ” “at least one of A, B, and C, ” “one or more of A, B, and C, ” and “A, B, C, or any combination thereof” may be A only, B only, C only, A and B, A and C, B and C, or A and B and C, where any such combinations may contain one or more member or members of A, B, or C. All structural and functional equivalents to the elements of the various aspects described throughout this disclosure that are known or later come to be known to those of ordinary skill in the art are expressly incorporated herein by reference and are intended to be encompassed by the claims. Moreover, nothing disclosed herein is intended to be dedicated to the public regardless of whether such disclosure is explicitly recited in the claims. The words “module, ” “mechanism, ” “element, ” “device, ” and the like may not be a substitute for the word “means. ” As such, no claim element is to be construed as a means plus function unless the element is expressly recited using the phrase “means for. ”
[0045] While aspects of the present disclosure have been described in conjunction with the specific embodiments thereof that are proposed as examples, alternatives, modifications, and variations to the examples may be made. Accordingly, embodiments as set forth herein are intended to be illustrative and not limiting. There are changes that may be made without departing from the scope of the claims set forth below.
Claims
1.A method for UE to perform AI / ML model transfer / delivery for wireless communication system, comprising the steps of:Setup model transfer / delivery tunnel with OTT server;Receive the update AI / ML model from OTT server.2.The method of claim 1, further comprising UE request the model transfer / delivery to its OTT server.3.The method of claim 2, wherein the model transfer / delivery triggering is initiated from UE.4.The method of claim 3, further comprising UE detects the site, scenario or radio environment change, and triggers a model transfer / delivery request to its OTT server.5.The method of claim 2, wherein the model transfer / delivery triggering is initiated from RAN node.6.The method of claim 5, further comprising UE receives the model transfer / delivery indication from RAN and request a model transfer / delivery to its OTT server.7.The method of claim 2, wherein the model transfer / delivery request is delivered via control plane signaling or user plane.8.The method of claim 7, wherein the model transfer / delivery request is delivered to RAN node, and further delivered to OTT server; or wherein the model transfer / delivery request is delivered to RAN node, then further delivered to 5GS / OAM, and then further delivered to OTT server.9.The method of claim 7, wherein the model transfer / delivery request is delivered to 5GS / OAM, and further delivered to OTT server.10.The method of claim 8, wherein the model transfer / delivery request is delivered from UE to RAN by legacy measurement procedure (e.g., SON / MDT, UE measurement report) .11.The method of claim 1, further comprising UE receives the model transfer / delivery indication from the RAN node.12.The method of claim 1, wherein the model transfer / delivery tunnel is from OTT server to UE via a dataflow tunnel.13.The method of claim 1, wherein the model transfer / delivery tunnel is consist of several separate user plane or control plane tunnels.14.The method of claim 13, wherein the model transfer / delivery tunnel is from OTT server to RAN, then from RAN to UE, or wherein the model transfer / delivery tunnel is from OTT server to 5GS / OAM, then from 5GS / OAM to UE, or wherein the model transfer / delivery tunnel is from OTT server to 5GS / OAM, then from 5GS / OAM to RAN, then from RAN to UE.15.The method of claim 1, further comprising UE applies the received AI / ML model.16.The method of claim 15, wherein the AI / ML model is updated proactively and UE stores the model in the buffer for future use, or wherein the AIML model is updated reactively and UE activate the model for the current scene.17.The method of claim 1, further comprising UE request model identification to network (RAN, 5GC or OAM) after receiving the updated model.18.The method of claim 1, further comprising UE reports the updated UE capability for the updated model to the network (RAN, 5GC or OAM) .
Citation Information
Patent Citations
Model discovery and selection for cooperative machine learning in cellular networks
US20220182263A1