Managing network functions associated with user equipment task offloading

Abstract

Techniques for managing one or more network functions associated with user equipment data exchange functions are disclosed. While not necessarily limited to this, the disclosed techniques are well suited to implement to manage charging functions associated with offloading of work tasks of user equipment participating in split artificial intelligence / machine learning (AIML) model processing in a communication network environment.

Description

Technical Field

[0001] The field relates generally to communication networks and, more particularly, but not exclusively, to network function management in such communication networks. Background Art

[0002] This section introduces various aspects that may help to better understand the present invention. Therefore, the statements in this section should be read in this light and should not be understood as admissions of the contents in or outside the prior art.

[0003] While fourth generation (4G) wireless mobile communication technology, also known as Long Term Evolution (LTE) technology, was designed to provide high capacity mobile multimedia at high data rates, particularly for human interaction, fifth generation (5G) technology is being used not only for human interaction but also for machine-type communications in so-called Internet of Things (IoT) networks.

[0004] More specifically, 5G networks are designed to enable large-scale IoT services (e.g., large numbers of capacity-limited devices) and mission-critical IoT services (e.g., requiring high reliability), while also providing improvements to traditional mobile communication services in the form of enhanced mobile broadband (eMBB) services, providing better wireless Internet access for mobile devices.

[0005] In an example communication system, a user equipment (5G UE in a 5G network, or more broadly, UE) such as a mobile terminal (subscriber) communicates over an air interface with a base station or access point of an access network, referred to as a 5G AN, in the 5G network. An access point (e.g., gNB) is illustratively part of the access network of the communication system.

[0006] For example, in a 5G network, the access network, referred to as the 5G AN, is described in the 5G Technical Specification (TS) 23.501, entitled “Technical Specification Group Services and System Aspects; System Architecture for the 5G System”, and TS 23.502, entitled “Technical Specification Group Services and System Aspects; Procedures for the 5G System (5GS)” (the entire contents of which are incorporated herein by reference). Typically, an access point (e.g., a gNB) provides a UE with access to a core network (CN or 5GC), which then provides the UE with access to other UEs and / or data networks, such as packet data networks (e.g., the Internet). TS 23.501 goes on to define a 5G service-based architecture (SBA) that models services as network functions (NFs) that communicate with each other using representative state transfer application programming interfaces (Restful APIs). In addition, TS 33.501, entitled “Technical Specification Group Services and System Aspects; Security Architecture and Procedures for the 5G System” (the entire contents of which are incorporated herein by reference) further describes security management details associated with 5G networks.

[0007] However, in some cases, UEs may communicate with each other via a secure direct connection. One such exemplary case is known as work offloading, i.e., where one UE offloads work to another UE via a secure direct connection between the UEs. In this case, management of attributes associated with such use cases by one or more network functions of the core network (e.g., CN or 5GC) is an important consideration. However, as attempts are made to improve the architecture and protocols associated with 5G networks to improve network efficiency and / or subscriber convenience, network function management issues associated with such use cases may present significant challenges. Summary of the invention

[0008] The illustrative embodiments provide techniques for managing one or more network functions associated with user device data exchange functions. Although not necessarily limited thereto, the illustrative embodiments are well suited for implementation to manage billing functions associated with workload offloading of user devices participating in split artificial intelligence / machine learning (AIML) model processing in a communication network environment.

[0009] In one illustrative embodiment, from the perspective of a network entity, a method includes: receiving a message from an application requesting a service via a communication network. The method also includes: assisting in managing one or more charging policies for the service when the service is performed by a first user device and a second user device, wherein the first user device and the second user device are in a data exchange relationship with respect to the service.

[0010] In another illustrative embodiment, from the perspective of a consumer user device, a method includes: receiving, at a first user device, a message from an application requesting a service via a network entity of a communication network. The method also includes: establishing, by the first user device, a data exchange relationship with a second user device proximate to the first user device to share performance of the service. The method also includes: sending, by the first user device, a message to the network entity, the message including: computing usage data for providing the service, and also indicating a role performed by the first user device in the data exchange relationship, so that the communication network can implement one or more charging policies for the service with respect to the second user device.

[0011] In another illustrative embodiment, from the perspective of a provider user device, a method includes: receiving, at a first user device, a message from a second user device proximate to the first user device, the message requesting a connection for establishing a data exchange relationship with the first user device to share performance of a service requested via a network entity of a communication network. The method also includes: sending, by the first user device, a message to the network entity, the message including computing usage data for providing the service and also indicating a role performed by the first user device in the data exchange relationship to enable the communication network to implement one or more charging policies for the service with respect to the first user device.

[0012] Additional illustrative embodiments are provided in the form of a non-transitory computer-readable medium having an executable program code embodied therein, which, when executed by a processor, causes the processor to perform the above and / or other steps, operations, etc. Additional illustrative embodiments include an apparatus having a processor and a memory, the memory being configured to perform the above and / or other steps, operations, etc. Some illustrative embodiments include a system configured to perform the above and / or other steps, operations, etc. Furthermore, some illustrative embodiments include an apparatus or system including components for performing the above and / or other steps, operations, etc.

[0013] Advantageously, the illustrative embodiments provide techniques for managing billing functionality in a communication network in a data exchange context (e.g., a work task offloading context between at least two UEs) with respect to performance of services (such as, but not limited to, split AIML model processing services).

[0014] These and other features and advantages of the embodiments described herein will become more apparent from the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 illustrates a communication network environment that may be used to implement one or more illustrative embodiments;

[0016] Figure 2 illustrates user equipment and entities that may be used to implement one or more illustrative embodiments;

[0017] FIG. 3A to FIG. 3C illustrates the differences between scenarios involving work task offloading and scenarios without work task offloading with respect to a user device operating in a split AIML model context in which one or more illustrative embodiments may be implemented;

[0018] Figure 4 illustrates a converged charging system environment that may be used to implement one or more illustrative embodiments;

[0019] Figure 5 illustrates a process for managing billing functionality in a communication network environment having workload offloading functionality according to an illustrative embodiment; and

[0020] FIG. 6A to FIG. 6C An exemplary message structure and trigger conditions associated with managing billing functionality in a communication network environment with workload offloading functionality are illustrated in accordance with an illustrative embodiment. DETAILED DESCRIPTION

[0021] Embodiments will be described herein in conjunction with example communication systems and associated technologies for network function management in communication systems. However, it should be understood that the scope of the claims is not limited to the specific type of communication system and / or process disclosed. Embodiments may be implemented in various other types of communication systems using alternative processes and operations. For example, although described in the context of a wireless cellular system utilizing a third generation partnership project (3GPP) system element, such as a 3GPP next generation system (5G), the disclosed embodiments may be directly applicable to various other types of communication systems, such as a 6G communication system.

[0022] According to an illustrative embodiment implemented in a 5G communication system environment, one or more 3GPP Technical Specifications (TS) and Technical Reports (TR) may provide further explanation of network elements / functions and / or operations that may interact with parts of the present invention solution, such as the above-mentioned 3GPP TS23.501, TS23.502 and TS 33.501.Other 3GPP TS / TR documents can provide additional details that a person of ordinary skill in the art will implement, for example, TR 22.876: "TechnicalSpecification Group Services and System Aspects; Study on AI / ML Model TransferPhase 2", TR 22.874: "Technical Specification Group Services and SystemAspects; Study on Traffic Characteristics and Performance Requirements for AI / ML Model Transfer in 5GS", TS22.115: "Technical Specification Group Services and System Aspects; Service Aspects; Charging and Billing", TS 32.277: "Technical Specification Group Services and System Aspects; Telecommunication Management; Charging Management; Proximity-based Services (ProSe) Charging", TS32.255: "Technical Specification Group Services and System Aspects; Telecommunication Management; Charging" Management; 5G Data Connectivity Domain Charging; Stage 2" and TS 32.290: "Technical Specification Group Services and System Aspects; Telecommunication Management; Charging Management; 5G system; Services, Operations and Procedures of Charging Using Service Based Interface (SBI)", the entire contents of which are incorporated herein by reference.Note that 3GPP TS / TR documents are non-limiting examples of communication network standards (e.g., specifications, procedures, reports, requirements, recommendations, etc.) However, while well suited to 5G-related 3GPP standards, embodiments are not necessarily limited to any particular standard.

[0023] It should be understood that in some illustrative embodiments, the terms 5G network, etc. (e.g., 5G system, 5G communication system, 5G environment, 5G communication environment, etc.) may be understood to include all or part of the access network and all or part of the core network. However, the terms 5G network, etc. may occasionally be used interchangeably with the terms 5GC network, etc. herein without loss of generality, as those of ordinary skill in the art understand any distinction.

[0024] Before describing the illustrative embodiments, Figure 1 and Figure 2 A general description describing some of the main components of a 5G network.

[0025] Figure 1 1 shows a communication system 100 in which an illustrative embodiment may be implemented. It should be understood that the elements shown in the communication system 100 are intended to represent some of the main functions provided within the system, such as control plane functions, user plane functions, etc. Figure 1 The blocks shown in refer to specific elements in the 5G network that provide some of these main functions. However, other network elements may be used to implement some or all of the main functions represented. In addition, it should be understood that Figure 1 Not all functionality of a 5G network is depicted in the figures. Rather, at least some functionality is shown to help explain the illustrative embodiments. Subsequent figures may depict some additional elements / functionality (i.e., network entities).

[0026] Thus, as shown, the communication system 100 includes a user equipment (UE) 102 that communicates with an access point 104 via an air interface 103. It should be understood that the UE 102 can use one or more other types of access points (e.g., access functions, networks, etc.) to communicate with other 5GC networks other than gNBs. By way of example only, the access point 104 can be any 5G access network (gNB), an untrusted non-3GPP access network using a non-3GPP interworking function (N3IWF), a trusted non-3GPP network using a trusted non-3GPP gateway function (TNGF), or a wired access using a wired access gateway function (W-AGF), or can correspond to a traditional access point (e.g., an eNB). In addition, the access point 104 can be a wireless local area network (WLAN) access point, which will be further explained in the illustrative embodiments described herein.

[0027] UE 102 may be a mobile station, and such a mobile station may include, for example, a mobile phone, a computer, an IoT device, or any other type of communication device. Therefore, the term "user equipment" as used herein is intended to be interpreted broadly to cover a variety of different types of mobile stations, subscriber stations, or more generally communication devices, including examples such as combinations of data cards inserted into other devices such as laptops or smartphones. Such communication devices are also intended to cover devices that are generally referred to as access terminals.

[0028] In one illustrative embodiment, the UE 102 includes a Universal Integrated Circuit Card (UICC) portion and a Mobile Equipment (ME) portion. The UICC is the user-related portion of the UE and contains at least one Universal Subscriber Identity Module (USIM) and appropriate application software. The USIM securely stores permanent subscription identifiers and their associated keys, which are used to uniquely identify and authenticate subscribers accessing the network. The ME is the user-independent portion of the UE and contains Terminal Equipment (TE) functions and various Mobile Terminal (MT) functions. Alternative illustrative embodiments may not use UICC-based authentication, such as non-public (private) networks (NPNs).

[0029] Note that in one example, the permanent subscription identifier is an International Mobile Subscriber Identity (IMSI) unique to the UE. In one embodiment, the IMSI is a fixed 15-digit length and includes a 3-digit Mobile Country Code (MCC), a 3-digit Mobile Network Code (MNC), and a 9-digit Mobile Station Identification Number (MSIN). In a 5G communication system, the IMSI is referred to as a Subscription Permanent Identifier (SUPI). In the case where the IMSI is used as the SUPI, the MSIN provides the subscriber identity. Therefore, it is usually only necessary to encrypt the MSIN part of the IMSI. The MNC and MCC parts of the IMSI provide routing information for routing to the correct home network by the serving network. When the MSIN of the SUPI is encrypted, it is referred to as a Subscription Hidden Identifier (SUCI). Another example of a SUPI uses a Network Access Identifier (NAI). NAI is commonly used for IoT communications.

[0030] Access point 104 is illustratively part of a radio access network or RAN of communication system 100. Such a radio access network may include, for example, a 5G system having multiple base stations. More generally, components of a radio access network may be considered “radio access entities”.

[0031] Furthermore, in this illustrative embodiment, the access point 104 is operably coupled to an access and mobility management function (AMF / SEAF) 106. In a 5G network, the AMF / SEAF supports mobility management (MM) and security anchor (SEAF) functions, among other things.

[0032] In this illustrative embodiment, the AMF / SEAF 106 is operably coupled to other network functions 108 (e.g., services that use other network functions). As shown, some of these other network functions 108 include, but are not limited to, a charging function (CHF), a charging trigger function (CTF), an account and balance management function (ABMF), a charging gateway function (CGF), a rating function (RF), a direct discovery name management function (DDNMF), and an application function (AF). These listed network function examples are typically implemented in the home network of the UE subscriber, as further explained below. Some or all of these functions are used to enable a communication service provider (CSP) to perform converged or integrated billing that combines online and offline billing systems to address new and emerging 5G monetization use cases. Typically, the CHF collects network and service usage data, and enables payment (understanding how much a user should be charged for the services consumed), and allows the CSP to create new services and quickly and efficiently provide and launch these services to consumers. The AF exposes the application layer for interaction with 5G NFs and network resources, such as splitting AIML functionality, as will be further explained in this document. The DDNMF is a logical 5G NF for handling network-related operations required for dynamic ProSe direct discovery. The DDNMF in the HPLMN may interact with the DDNMOF in the VPLMN or the local PLMN in order to manage the ProSe direct discovery service. ProSe direct discovery is a process employed by a ProSe-enabled UE to discover other ProSe-enabled UEs in its vicinity (neighborhood) based on direct radio transmissions (direct device connection or sidelink) between the two UEs using NR technology. Note that, more generally, an NF may be considered a "network entity"

[0033] Note that a UE, such as UE 102, typically subscribes to a so-called home public land mobile network (HPLMN), in which some or all of the functions 106 and 108 reside. Alternatively, a UE, such as UE 102, may receive services from an NPN, in which these functions may reside. The HPLMN is also referred to as a home environment (HE). If the UE is roaming (not in an HPLMN), it typically connects to a visited public land mobile network (VPLMN), also referred to as a visited network, and the network currently serving the UE is also referred to as a serving network. In a roaming situation, some of the functions 106 and 108 may reside in a VPLMN, in which case the functions in the VPLMN communicate with the functions in the HPLMN as needed. However, in a non-roaming scenario, the access and mobility management functions 106 and other network functions 108 reside in the same communication network (i.e., the HPLMN). Unless otherwise noted, the embodiments described herein are not necessarily limited to which functions reside in which PLMN (i.e., HPLMN or VPLMN).

[0034] Access point 104 is also operatively coupled (via one or more of functions 106 and / or 108) to a session management function (SMF) 110, which is operatively coupled to a user plane function (UPF) 112. UPF 112 is operatively coupled to a packet data network, such as the Internet 114. Note that the thicker solid lines in the figure represent the user plane (UP) of the communication network, as compared to the thinner solid lines representing the control plane (CP) of the communication network. It should be understood that Figure 1 The network (e.g., the Internet) 114 in the example may additionally or alternatively represent other network infrastructures, including but not limited to cloud computing infrastructures and / or edge computing infrastructures. Other typical operations and functions of such network elements are not described here, as they are not the focus of the illustrative embodiments and may be found in appropriate 3GPP 5G documents. Note that the functions shown in 106, 108, 110, and 112 are examples of network functions (NFs).

[0035] It should be understood that this particular arrangement of system elements is merely an example, and in other embodiments, additional or alternative elements of other types and arrangements may be used to implement the communication system. For example, in other embodiments, the communication system 100 may include other elements / functions not explicitly shown herein.

[0036] therefore, Figure 1 The arrangement of is only one example configuration of a wireless cellular system, and many alternative configurations of system elements may be used. For example, although in Figure 1In the embodiments shown, only a single element / function is shown, but this is only for simplicity and clarity of description. A given alternative embodiment may of course include a greater number of such system elements, as well as additional or alternative elements of the type normally associated with conventional system implementations.

[0037] It should also be noted that although Figure 1 The system elements are illustrated as single functional blocks, but the various subnets that make up the 5G network are divided into so-called network slices. A network slice (network partition) is a logical network that provides specific network capabilities and network characteristics that can support corresponding service types, optionally through the use of network function virtualization (NFV) on a common physical infrastructure. Using NFV, network slices can be instantiated as needed for a given service, such as eMBB services, large-scale IoT services, and mission-critical IoT services. Therefore, when an instance of a network slice or function is created, the network slice or function is instantiated. In some embodiments, this involves installing or otherwise running the network slice or function on one or more host devices of the underlying physical infrastructure. UE 102 is configured to access one or more of these services via access point 104.

[0038] Figure 2 2 is a block diagram illustrating the computing architecture of various participants in the method according to the illustrative embodiment. More specifically, the system 200 is shown as including a user equipment (UE) 202 and a plurality of entities 204-1, ..., 204-N. For example, in the illustrative embodiment, and again referring to Figure 1 , UE 202 can represent UE 102 (and any UE acting as a consumer and / or provider in the context of work task offloading, as will be further described herein), and entities 204-1,...,204-N can represent functions 106 and 108 (i.e., network entities such as but not limited to CHF, CTF, ABMF, CGF, RF, DDNMF, AF), and access point 104 (i.e., radio access entity such as but not limited to RAN node or gNB). It should be understood that UE 102 and entities 204-1,...,204-N are configured to interact to provide management (e.g., billing functions associated with work task offloading in the context of split AIML model processing) and other techniques described herein.

[0039] The user device 202 includes a processor 212 coupled to a memory 216 and an interface circuit system 210. The processor 212 of the user device 202 includes a management processing module 214 that can be implemented at least in part in the form of software executed by the processor. The management processing module 214 performs management described in conjunction with subsequent figures and other figures herein. The memory 216 of the user device 202 includes a management storage module 218 that stores data generated or otherwise used during management operations.

[0040] Each of the entities (individually or collectively referred to herein as 204) includes: a processor 222 (222-1, ..., 222-N) coupled to a memory 226 (226-1, ..., 226-N) and an interface circuit system 220 (220-1, ..., 220-N). Each processor 222 of each entity 204 includes: a management processing module 224 (224-1, ..., 224-N) that can be implemented at least in part in the form of software executed by the processor 222. The management processing module 224 performs management operations described in conjunction with subsequent figures and other figures in this document. Each memory 226 of each entity 204 includes: a management storage module 228 (228-1, ..., 228-N) that stores data generated or otherwise used during management operations.

[0041] Processors 212 and 222 may include, for example, a microprocessor such as a central processing unit (CPU), an application specific integrated circuit (ASIC), a digital signal processor (DSP), or other types of processing devices, as well as portions or combinations of such elements.

[0042] The memories 216 and 226 may be used to store one or more software programs that are executed by the respective processors 212 and 222 to implement at least a portion of the functionality described herein. For example, management operations and other functions described in conjunction with subsequent figures and otherwise herein may be implemented in a straightforward manner using software code executed by the processors 212 and 222.

[0043] Thus, a given one of memories 216 and 226 may be viewed as an example of what is more generally referred to herein as a computer program product, or more generally as a computer or processor readable (non-transitory or storage) medium having executable program code embodied therein. Other examples of computer or processor readable media may include any combination of disks or other types of magnetic or optical media. Illustrative embodiments may include an article of manufacture that includes such a computer program product or other computer or processor readable medium.

[0044] In addition, memories 216 and 226 may more specifically include, for example, electronic random access memory (RAM), such as static RAM (SRAM), dynamic RAM (DRAM), or other types of volatile or non-volatile electronic memory. The latter may include, for example, non-volatile memory, such as flash memory, magnetic RAM (MRAM), phase change RAM (PC-RAM), or ferroelectric RAM (FRAM). The term "memory" as used herein is intended to be interpreted broadly and may additionally or alternatively cover, for example, read-only memory (ROM), disk-based memory or other types of storage devices, and parts or combinations of such devices.

[0045] Interface circuitry 210 and 220 illustratively include transceivers, or other communication hardware or firmware that allows related system elements to communicate with each other in the manner described herein.

[0046] from Figure 2 As is apparent from FIG. 2 , the user device 202 and the plurality of entities 204 are configured to communicate with each other as management participants via their respective interface circuit systems 210 and 220. The communication involves each participant sending data to and / or receiving data from one or more other participants. The term "data" as used herein is intended to be interpreted broadly to cover any type of information that may be sent between participants, including but not limited to identity data, key pairs, key indicators, tokens, secrets, management messages, registration request / response messages and data, request / response messages, authentication request / response messages and data, metadata, control data, audio, video, multimedia, consent data, other messages, and the like.

[0047] It should be understood that Figure 2 The specific arrangement of components shown in is only an example, and in other embodiments, many alternative configurations may be used. For example, any given network element / function and / or access point may be configured to include additional or alternative components and support other communication protocols.

[0048] Each of the other system elements such as access point 104, SMF 110 and UPF 112 can be configured to include components such as processors, memories and network interfaces. In addition, entities such as third-party applications and network operators can participate in the methods described herein via computing devices configured to include components such as processors, memories, and network interfaces. These elements and devices do not need to be implemented on separate independent processing platforms, but can represent different functional parts of a single general-purpose processing platform.

[0049] More generally, Figure 2It can be considered to represent a processing device that is configured to provide corresponding management functions and is operably coupled to each other in a communication system. As an example only, all or part of each of the UE 202 and the plurality of entities 204 (e.g., processors and memories) can be considered as an example of a component for performing one or more operations, one or more steps, one or more functions, one or more processes, etc. described herein.

[0050] As described above, 3GPP TS23.501 defines the 5GC network architecture as service-based, such as a service-based architecture (SBA). It is recognized herein that when deploying different NFs, there may be many situations in which the NF may need to interact with entities external to the SBA-based 5GC network (e.g., including (multiple) corresponding PLMNs, such as HPLMN and VPLMN). Therefore, the term "internal" used herein illustratively refers to operations and / or communications within the SBA-based 5GC network (e.g., SBA-based interfaces), and the term "external" illustratively refers to operations and / or communications (non-SBA interfaces) external to the SBA-based 5GC network.

[0051] In view of the above general description of some features of the 5GC network, the problems of the existing methods for managing one or more network functions associated with a user equipment work task offloading function in a communication network environment, and the solutions proposed according to the illustrative embodiments will be described below. Although not limited to this, the illustrative embodiments will be described in the context of managing the billing function associated with the work task offloading of a user equipment participating in split artificial intelligence / machine learning (AIML) model processing in a communication network environment. In addition, it should be understood that the term "work task offloading relationship" is an example of a data exchange relationship regarding a requested service (e.g., split AIML model processing). In addition, although some illustrative embodiments are described with respect to a work task offloading relationship between two UEs, it should be understood that according to other illustrative embodiments, for a requested service, more than two UEs may participate in the same work task offloading relationship.

[0052] Work task offloading, or more specifically proximity-based work task offloading, is based on a third-party request for one UE (i.e., relay UE) to receive data from another UE (i.e., remote UE) via a secure direct (device-to-device) connection and perform computation of the work task for the remote UE. The computation result can be further sent to a network server or NF.

[0053] In the context of a specific use case, proximity-based work offloading can be used for AIML inference model processing. Model splitting is an important feature of AI inference model processing, where different devices or functions (e.g., one or more UEs and one or more network functions or servers) compute different layers of the AIML inference model. In some scenarios, it is recognized herein that the number of devices or functions computing layers, and the amount of data transfer, corresponds to different model split points. For example, Figure 3A A layer-wise computation / communication resource evaluation 300 of an example AIML model (e.g., an AlexNet model) is shown. A general trend is achieved where the more layers a UE computes, the less intermediate data needs to be sent to a network server (e.g., an application). In addition, when the computational power of the UE is low (e.g., due to low battery), the application can change the split point so that the UE can compute fewer layers while increasing the data rate in the Universal Mobile Telecommunications System (UMTS) air interface (referred to as the Uu interface) to send a higher load of intermediate data to the network server.

[0054] However, sometimes the data rate cannot be increased due to limitations of radio resources (e.g., gNB). In this case, the UE with lower computing power needs to offload the computing tasks to the neighboring UE (e.g., relay UE), while still maintaining the computing service and letting the neighboring UE send the computing data to the network server. Therefore, by offloading the work tasks using secure direct device connection, the computing load of the original UE will be released, and the data rate in the Uu interface will not necessarily increase, which results in better network performance.

[0055] More specifically, Figure 3B The service flow 310 is shown without any workload offloading. Figure 3C A service flow 320 is shown where work task offloading occurs.

[0056] like Figure 3B As shown in service flow 310 in FIG. 1 , UE-A is using Figure 3A A convolutional neural network (CNN) of the AlexNet model represented in performs image recognition. Assume that UE-A selects split point 3 for AI reasoning. The achieved end-to-end (E2E) service latency (including image recognition latency and intermediate data transmission latency) is 1 second. However, when UE-A's battery becomes low, it cannot undertake the heavy workload of the AlexNet model (i.e., computing layers 1-15 of the AlexNet model on the local (UE) side).

[0057] Therefore, if Figure 3CAs shown in the service flow 320 in FIG. 1 , it is also assumed that, when managed by the 5G network (CN or 5GC), UE-A finds UE-B (e.g., another mobile subscriber terminal or customer premises equipment (CPE)) that has installed the same model and is willing to take on the offload task from UE-A. Note that the 5G network does not store the location data of UE-A and UE-B.

[0058] like Figure 3C As further shown in , UE-A establishes a side link (i.e., a secure direct device connection) with UE-B. During the side link establishment, it is assumed that UE-B also obtains information about the total service delay (including image recognition delay and intermediate data transmission delay) and the processing time consumed by UE-A to calculate layers 1-4.

[0059] Since UE-B has obtained the E2E service delay and the processing time consumed by UE-A, and also knows its own processing time for calculating layers 5-15, UE-B can determine the quality of service (QoS) parameters applied to both the Uu interface and the sidelink while maintaining the same (i.e., 1 second) E2E service delay.

[0060] Note that it is assumed that UE-A and UE-B have the same computing power, i.e. the time used to compute a specific AlexNet model layer is the same for UE-A and UE B. Otherwise, the data rate on the Uu interface and the sidelink may be changed accordingly.

[0061] like Figure 3C As further shown in the figure, UE-A sends intermediate data (i.e., data after calculating layers 1-4) to UE-B via a side link, and UE-B performs further processing and sends the intermediate data (i.e., data after calculating layers 5-15) to the network (application) server via the Uu interface. Figure 3C The specific model layers computed by UE-A and UE-B are shown in . UE-A continues to perform image recognition by utilizing the sidelink and the computational power of UE-B, while the source and destination Internet Protocol (IP) addresses of the image recognition service and the E2E service latency remain unchanged.

[0062] Therefore, if Figure 3C As shown in , a direct device connection (e.g., sidelink) can be advantageously used to implement proximity-based work offloading. In this case, the data rate of UE-A on the Uu interface does not need to be increased, and the computational load of UE-A is offloaded to UE-B.

[0063] However, it is recognized herein that, as mentioned above in FIG. 3A to FIG. 3CAs described in the context of , etc., such proximity-based work offloading regarding split AIML model processing may pose significant challenges to the converged billing system used by CN or 5GC to collect billing information.

[0064] For example, the above-mentioned 3GPP TS22.115 considers the requirement that 5GC can collect and provide billing information in terms of duration and amount of data sent / received, QoS, etc. Such requirements apply to traditional call processing-related billing. However, it is recognized herein that for use cases such as split AIML model processing using computing resources of neighboring UEs, users of neighboring UEs that provide such resources should benefit based on the amount of shared computing resources. For such requirements, it is recognized herein that 5GC needs to collect AIML computing-specific information and use it to determine charging strategies.

[0065] This article also recognizes that 5GC needs to allow operators or authorized third parties to configure charging policies for AIML's proximity-based workload offloading. These charging policies should be defined in a way that incentivizes users of UEs to agree to share their computing resources. These policies will be different from traditional billing, in which operators charge UEs for providing services.

[0066] It is further recognized herein that in some scenarios, a third party may be able to provide high computing power to a UE specifically for providing proximity-based work offloading services to AIML. By way of example only, an airport authority or local government (more generally, an entity) may decide to install equipment at an airport that not only provides additional AIML computing power to the airport's UEs, but also shares airport safety-related information within the device and to a network (application) server. In such a scenario, since the equipment installed by the airport authority provides additional computing power, the billing policy should also be defined by the airport authority. It should be understood that such equipment installed at the locations explained herein may be referred to as customer premises equipment (CPE). Such CPE (whether fixed and / or mobile) may be more generally referred to as one or more UEs.

[0067] The illustrative embodiments enable the above-described and other features and advantages, and thereby overcome the above-described and other technical disadvantages of existing approaches by providing improved techniques for managing billing functionality in the context of work task offloading with respect to split AIML model processing, etc.

[0068] As used herein with respect to one or more illustrative embodiments, the term “AIML service provider” is used for a UE that provides AIML work offloading service to another UE in proximity, while the term “AIML service consumer” is used for a UE that requests AIML work offloading service from an AIML service provider.

[0069] As will be further described and explained, illustrative embodiments provide techniques that allow network operators (e.g., CSPs) to define billing policies for proximity-based work task offloading of AIML. In some illustrative embodiments, the defined policies enable the AIML service provider UE to obtain credit for offloaded computing work if the owner of the UE has provided user consent. In some illustrative embodiments, the policies are defined based on one or more of the number of computing layers provided, the number of times such computing services are used by neighboring AIML service consumer UEs, the number of times expected performance key performance indicators (KPIs) are met, and one or more of the priorities of the requests served.

[0070] If the AIML service provider UE is owned by a third party, for example, an airport authority or local government decides to install such an AIML assisted UE (e.g., CPE) at an airport or such a public place, an offline charging agreement may be reached between the network operator and the third party. In such a scenario, the charging policy may be defined according to such an agreement. For such a scenario, the charging policy may include bulk credits based on the number of such UEs installed and the number of times the expected performance KPIs are met.

[0071] In some illustrative embodiments, policies may be implemented to monitor performance KPIs before providing credits to AIML service provider UEs. In one non-limiting example scenario, if a UE with sufficient AIML computing power provides user consent, but fails to meet expected performance KPIs, such UEs do not obtain credits for providing computing services and their user consent is revoked.

[0072] In some demonstrative embodiments, billing information is collected from an AIML service provider UE to provide credit for computing services provided by the UE. In addition to the information collected from an AIML service consumer UE, one or more billing functions may also collect this information, which will be explained in further detail below.

[0073] Figure 4An exemplary converged charging system environment 400 is illustrated that can be used to implement one or more illustrative embodiments. Converged or converged billing combines online and offline billing systems together to address both billing scenarios that network operators conduct with their users. As generally shown in the converged billing system environment 400, a communication network 402 includes a converged billing system 404 that is operably coupled to a set of domains 406. The converged billing system 404 includes a charging function (CHF) that is operably coupled to an account and balance management function (ABMF), a charging gateway function (CGF), and a rating function (RF), while the set of domains 406 includes a service domain, a subsystem domain, and a core network (CN) domain, each domain having a charging trigger function (CTF). The CGF is operably coupled to a charging domain 408. Further details of the operation of the various functions in the converged billing system environment 400 can be found in the above-mentioned 3GPP TS32.277. However, as described above, when implemented with existing billing management functionality, the converged billing system environment 400 cannot address workload offloading scenarios, particularly in the context of split AIML model processing.

[0074] Reference now Figure 5 , according to an illustrative embodiment, a process 500 for managing billing functionality in a communication network environment with work task offloading functionality is depicted. More specifically, the process 500 manages converged billing functionality for a work task offloading scenario in the context of split AIML model processing. Thus, the process 500 may be based on Figure 4 The converged billing system environment 400 or any other suitable billing system in the communication network environment is implemented.

[0075] As shown, process 500 involves UE 502 (e.g., CTF of UE1 acting as an AIML service consumer), UE 504 (e.g., CTF of UE2 acting as an AIML service provider), DDNMF 506 (e.g., CTF of DDNMF), split AIML AF 508, and CHF 510. In this example, DDNMF 506, split AIML AF 508, and CHF 510 are part of the HPLMN of UE 502 and 504. In the case that UE 502 and / or UE 504 are roaming, the DDNMF in the corresponding VPLMN (not explicitly shown) can communicate with the DDNMF 506 of the HPLMN as needed.

[0076] Also shown in process 500 is a network operator 512 that is operably coupled to CHF 510 and (multiple) user(s) 514. As described above, recall that the network operator 512 (e.g., CSP) can agree on billing policies (i.e., including credit policies) with (multiple) user(s) 514 of a UE (e.g., UE 504) that provides work task offloading functionality to other UEs (e.g., UE 502). These policies can include, but are not limited to, AIML's proximity-based billing policies for work task offloading, where an AIML service provider UE can obtain credits for offloaded computing work if the owner of the UE has provided user consent. In some illustrative embodiments, the policy is defined based on one or more of the number of computing layers provided, the number of times such computing services are used by neighboring AIML service consumer UEs, the number of times expected performance key performance indicators (KPIs) are met, and the priority of the requests served. If the AIML service provider UE is owned by a third party, for example, an airport authority or local government decides to install such an AIML assisted UE (e.g., CPE) at an airport or such a public place, an offline billing agreement can be reached between the network operator and the third party. In such a scenario, the billing policy can be defined according to such an agreement. For such a scenario, the billing policy may include: batch credits based on the number of such UEs installed and the number of times the expected performance KPIs are met. In some illustrative embodiments, policies may be implemented to monitor performance KPIs before providing credits to the AIML service provider UE. In a non-limiting example scenario, if a UE with sufficient AIML computing power provides user consent, but fails to meet the expected performance KPIs, such UEs do not obtain credits for providing computing services and their user consent is revoked.

[0077] It should be understood that such a policy may be set by the network operator 512 in the converged charging system including the CHF 510 before steps 0 to 12 are performed, or updated in other ways during steps 0 to 12 as needed / required.

[0078] Step 0: As part of the non-access stratum (NAS) registration, assume that UE 502 (UE1) registers as an AIML service consumer and UE 504 (UE2) registers as an AIML service provider. The AIML service registered by UE 502 and 504 is a split AIML model processing service managed by split AIMLAF 508.

[0079] Step 1, 1a: The service request of the AIML workload split is shared from the split AIML AF 508 to the DDNMF 506, and then to the UE 502 registered as an AIML service consumer.

[0080] Step 2: As a result, after the discovery process triggered from UE 502, UE 502 establishes direct communication with UE 504 registered as an AIML service provider.

[0081] Step 3: The authentication and security procedures between UE 502 and UE 504 are completed.

[0082] Step 4: For the requested split task, a one-to-one direct service is established between UE 502 and UE 504 (via a secure direct device connection, such as a sidelink), and UE 504 performs the assigned required tasks (e.g., recall the above split AIML model processing example, where a consumer UE with a low battery condition computes layers 1-4 of the AIML model on behalf of the consumer UE, and the provider UE computes layers 5-15 of the AIML model on behalf of the consumer UE).

[0083] Step 5, 5a: After the splitting task is completed, the UE 504 shares the result in the AIML Work Split Complete message to split the AIML AF 508 through the DDNMF 506.

[0084] Steps 6 and 7: The direct device connection between UE 502 and UE 504 is disconnected.

[0085] Step 8: Split AIML AF 508 checks whether the assigned tasks meet the reporting criteria (service-based criteria) based on the input received in step 5a.

[0086] Steps 9, 9a, 9b: Share the status about the reporting criteria to UE 502 and UE 504 through DDNMF 506.

[0087] Steps 9c, 9d: Share the usage report information with the DDNMF 506 from the UE 502 and the UE 504, with additional parameters indicating the roles as a split AIML service consumer and a split AIML service provider, respectively.

[0088] Step 10a: DDNMF 506 calculates the credit score of UE 504, which is the AIML service provider. In this step, the credit score (calculated credit score) is a value representing the specific quantifiable computational effort expended by UE 504 in performing the split AIML model processing (e.g., calculating layers 5-15 of the AIML model). This is different from the term credit about how much monetary credit the user (owner) of UE 504 will receive from the converged billing system for performing the computational effort.

[0089] Step 10b: A charging data request including a credit score is sent from the DDNMF 506CHF 510.

[0090] Step 11: The CHF 510 generates a ProSe Function Direct Communication Charging Data Record (PF-DC-CDR) using the received input.

[0091] Step 12: CHF 510 responds to DDNMF 506 with a charging data response including UE 504's credit score.

[0092] Note that in some demonstrative embodiments, when establishing a work task offloading relationship with UE 504, UE 502 sends to UE 504 E2E latency information (e.g., one second), the specific AIML model it supports for the split operation (e.g., the AlexNet model), and the layer support limit for the operation (e.g., layers 1-4).

[0093] In an illustrative embodiment of implementing the techniques described in the above-mentioned 3GPP TS 32.277, in particular for ProSe direct discovery, the ProSe function (PF) collects the following charging information: the identity of the mobile subscriber using the ProSe function, such as IMSI; the identity of the PLMN using the ProSe function; the specific ProSe function used, such as notification, monitoring, or matching report, split AIML; and the role of the UE in ProSe, such as notification UE, monitoring UE, discoverer UE, discoverer UE, split AIML consumer / service provider.

[0094] also, Fig. 6A and Figure 6B Variations of the above-mentioned 3GPP TS 32.255 for the charging data message structure (ie including the credit score as described above in steps 10a to 12) are depicted in FIG. 6 for message content structure 600 and message content structure 610, respectively. Figure 6C A new trigger condition in Table 620 applicable to the above-mentioned 3GPP TS 32.277 is shown for split AIML model processing in the context of a work task offloading scenario.

[0095] Thus, at least one illustrative embodiment may include an apparatus (e.g., corresponding to a network entity such as DDNMF 506), the apparatus comprising at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least: receive a message from an application requesting a service via a communication network; and assist in managing one or more billing policies for a service when the service is at least partially performed by a first user device (e.g., corresponding to a consumer UE such as UE 502) and a second user device (e.g., corresponding to a provider UE such as UE 504), wherein the first user device and the second user device are in a data exchange relationship with respect to the service.

[0096] In some additional illustrative embodiments, assisting in managing one or more billing policies may also include one or more of the following: sending a message to a first user device requesting a service; receiving a message from a second user device indicating completion of the service; sending a message to an application indicating completion of the service; receiving a message from an application indicating that service-based criteria have been met; sending a message to the first user device and the second user device indicating that service-based criteria have been met; receiving corresponding messages from the first user device and the second user device, the corresponding messages including corresponding computing usage data for providing the service, wherein the corresponding messages also indicate: the roles performed by the first user device and the second user device in the data exchange relationship (wherein the corresponding messages from the first user device and the second user device may include: corresponding computing usage data for providing the service, the corresponding computing usage data indicating: the first user device performs the service consumer role, and the second user device performs the service provider role); calculating a computing credit score for the second user device; and sending the calculated computing credit score for the second user device to a billing system to enable billing results based on one or more billing policies.

[0097] In some additional illustrative embodiments, the data exchange relationship includes: a work task offloading relationship, and the service includes: a split artificial intelligence machine learning model processing service.

[0098] In some further illustrative embodiments, one or more charging policies are settable by a network operator and include a policy for providing a charging credit to an entity associated with the second user device for computational effort expended by the second user device in performing the quantifiable portion of the service.

[0099] In some further demonstrative embodiments, one or more charging policies are settable by a network operator in conjunction with an entity associated with the second user equipment.

[0100] Further, at least one illustrative embodiment may include an apparatus (e.g., corresponding to a consumer UE such as UE 502), comprising at least one processor and at least one memory storing instructions, which, when executed by the at least one processor, causes the apparatus to at least: receive a message from an application requesting a service via a network entity of a communication network; establish a data exchange relationship with a user device adjacent to the apparatus to share performance of the service; and send a message to the network entity, the message including computing usage data for providing the service and also indicating a role performed by the apparatus in the data exchange relationship to enable the communication network to implement one or more billing policies for the service with respect to the adjacent user devices.

[0101] In addition, at least one illustrative embodiment may include an apparatus (e.g., corresponding to a provider UE such as UE 504), the apparatus including at least one processor and at least one memory storing instructions, which instructions, when executed by the at least one processor, cause the apparatus to at least: receive a message from a user device adjacent to the apparatus, the message requesting a connection for establishing a data exchange relationship with the adjacent user device to share performance of a requested service via a network entity of a communication network; and send a message to the network entity, the message including computing usage data for providing the service and also indicating a role performed by the apparatus in the data exchange relationship so that the communication network can implement one or more billing policies for the service with respect to the apparatus.

[0102] As used herein, it should be understood that in some embodiments, the term "communication network" may include two or more separate communication networks. In addition, the specific processing operations and other system functions described in conjunction with the figures described herein are presented only as illustrative examples and should not be interpreted as limiting the scope of the present disclosure in any way. Alternative embodiments may use other types of processing operations and messaging protocols. For example, in other embodiments, the order of the steps may vary, or some steps may be at least partially executed concurrently with each other, rather than serially. In addition, one or more steps may be repeated periodically, or multiple instances of these methods may be executed in parallel with each other.

[0103] It should be emphasized again that the various embodiments described herein are presented only as illustrative examples and should not be interpreted as limiting the scope of the claims. For example, alternative embodiments may utilize different communication system configurations, user equipment configurations, base station configurations, provisioning and use processes, messaging protocols, and message formats than the above illustrative embodiments. These and many other alternative embodiments within the scope of the appended claims will be clear to those skilled in the art.

Claims

1. A method for communication, comprising: receiving a message from an application requesting a service via a communication network; as well as assisting in managing one or more charging policies for the service when the service is performed by a first user device and a second user device, wherein the first user device and the second user device are in a data exchange relationship with respect to the service; The steps are performed by at least one processor and at least one memory, wherein the at least one memory stores instructions executable by the at least one processor.

2. The method according to claim 1, wherein the data exchange relationship comprises: Work task offloading relationship, and the service includes: splitting artificial intelligence machine learning model processing service.

3. The method of claim 1, wherein the one or more charging policies are configurable by a network operator and comprise: A policy for providing a billing credit to an entity associated with the second user equipment for the computational effort expended by the second user equipment in performing the quantifiable portion of the service.

4. The method of claim 3, wherein the one or more charging policies are settable by the network operator in conjunction with the entity associated with the second user equipment.

5. The method of claim 1, wherein assisting in managing the one or more charging policies further comprises: Sending the message requesting the service to the first user equipment; receiving a message from the second user equipment indicating completion of the service; sending the message indicating completion of the service to the application; receiving a message from the application indicating that service-based criteria have been met; sending the message to the first user equipment and the second user equipment indicating that the service-based criteria have been met; receiving corresponding messages from the first user device and the second user device, the corresponding messages including corresponding computing usage data for providing the service, wherein the corresponding messages further indicate: roles performed by the first user device and the second user device in the data exchange relationship; Calculating a credit score of the second user device; as well as The calculated credit score of the second user equipment is sent to a charging system to enable a charging result based on the one or more charging policies.

6. A device for use in a communication network, comprising: means for receiving a message from an application requesting a service via a network entity of a communication network; means for establishing a data exchange relationship with a user equipment proximate to the apparatus to share the performance of the service; as well as Means for sending a message to the network entity, the message comprising computational usage data for providing the service and further indicating a role performed by the apparatus in the data exchange relationship to enable the communications network to implement one or more charging policies for the service with respect to the proximate user equipment. 7 . The apparatus of claim 6 , wherein the message indicates that the apparatus performs a service consumer role and the neighboring user equipment performs a service provider role.

8. The apparatus according to claim 6, wherein the data exchange relationship comprises: Work task offloading relationship, and the service includes: splitting artificial intelligence machine learning model processing service.

9. The apparatus of claim 6, wherein the one or more charging policies are configurable by a network operator and comprise: A policy for providing a billing credit to an entity associated with the proximate user equipment for the computational effort expended by the proximate user equipment in performing the quantifiable portion of the service.

10. The apparatus of claim 9, wherein the one or more charging policies are settable by the network operator in conjunction with the entity associated with the proximate user equipment.

11. The apparatus according to claim 6, wherein the network entity comprises: Directly discover name management capabilities.

12. The apparatus of claim 6, wherein the apparatus comprises user equipment.

13. A method for communication, comprising: At a first user equipment, receiving a message from an application requesting a service via a network entity of a communication network; The first user equipment establishes a data exchange relationship with a second user equipment adjacent to the first user equipment to share the performance of the service; as well as sending, by the first user equipment, a message to the network entity, the message comprising computing usage data for providing the service and further indicating a role performed by the first user equipment in the data exchange relationship, so as to enable the communication network to implement one or more charging policies for the service with respect to the second user equipment; The steps are performed by at least one processor and at least one memory, wherein the at least one memory stores instructions executable by the at least one processor.

14. An apparatus for use in a communication network, comprising: means for receiving a message from a user equipment proximate to the apparatus, the message requesting a connection for establishing a data exchange relationship with the proximate user equipment to share performance of a service requested via a network entity of a communication network; as well as Means for sending a message to the network entity, the message comprising computing usage data for providing the service and also indicating a role performed by the device in the data exchange relationship to enable the communication network to implement one or more charging policies for the service with respect to the device.

15. The apparatus of claim 14, wherein the message indicates that the apparatus performs a service provider role and the neighboring user equipment performs a service consumer role.

16. The apparatus according to claim 14, wherein the data exchange relationship comprises: Work task offloading relationship, and the service includes: splitting artificial intelligence machine learning model processing service.

17. The apparatus of claim 14, wherein the one or more charging policies are configurable by a network operator and comprise: A policy for providing a billing credit to an entity associated with the device for the computational effort expended by the device in performing the quantifiable portion of the service.

18. The apparatus of claim 17, wherein the one or more charging policies are settable by the network operator in conjunction with the entity associated with the apparatus.

19. The apparatus of claim 14, wherein the network entity comprises: Directly discover name management capabilities.

20. The apparatus of claim 14, wherein the apparatus comprises user equipment.

21. A method for communication, comprising: At a first user equipment, receiving a message from a second user equipment adjacent to the first user equipment, the message requesting a connection for establishing a data exchange relationship with the first user equipment to share performance of a service requested via a network entity of a communication network; as well as sending, by the first user equipment, a message to the network entity, the message comprising computing usage data for providing the service and further indicating a role performed by the first user equipment in the data exchange relationship, so as to enable the communication network to implement one or more charging policies for the service with respect to the first user equipment; The steps are performed by at least one processor and at least one memory, wherein the at least one memory stores instructions executable by the at least one processor.

Images