Self-updating method, device and equipment of interoceptive perception algorithm

Abstract

This application provides a self-updating method, apparatus, and device for an integrated sensing algorithm. The method obtains a first response message sent by a target NWDAF via a SF (Signal Transfer Function). Based on the first response message, an initial model algorithm is deployed, and the initial model algorithm is used to analyze and process the sensed information to obtain sensed data (information reported by the base station). Based on the sensed data, a first request message is generated and sent to the target NWDAF. A second response message sent by the target NWDAF is obtained, indicating the processing method of the initial model algorithm. The initial model algorithm is updated and saved based on the second response message to obtain the target model algorithm. This method, through the interaction between the SF and the NWDAF, simultaneously sends sensed data and correct data to the NWDAF, enabling the NWDAF to self-update and optimize the model algorithm based on the data, thus improving the optimization and updating efficiency of the model algorithm.

Description

Technical Field

[0001] This application relates to the field of integrated communication and sensing technology, and in particular to a self-updating method, apparatus and device for an integrated communication and sensing algorithm. Background Technology

[0002] 5G-A (5G Advanced), as an evolution of 5G technology, is leading a new round of revolution in the communications field. Among its advancements, the integrated sensing network mode has become a research hotspot. This mode uses 5G base stations to perform real-time signal scanning of objects within its service range, acquiring information such as the object's type, location, speed, and route, providing strong support for emerging fields such as the low-altitude economy and smart cities.

[0003] In the 5G-A integrated sensing network, the 5G base station on the radio network side transmits sensing information to the newly added Sensing Function (SF). The SF is responsible for receiving and processing this information, and performing calculations and outputs according to application requirements. Currently, analytical computing capabilities are mainly achieved through one-time deployment and subsequent upgrades to update and optimize algorithm models.

[0004] However, current sensing and communication technologies have shortcomings in algorithm model updating and optimization. Because frequent performance optimization is required in the early stages of technology development, existing manual model upgrade methods are neither flexible nor convenient enough. This upgrade method not only increases maintenance costs but may also affect system stability and reliability. Therefore, how to more efficiently update and optimize algorithm models has become a key issue that urgently needs to be addressed in the development of current 5G-A sensing and communication technologies. Summary of the Invention

[0005] This application provides a self-updating method, apparatus, and device for a synesthetic sensing algorithm, which enables the algorithm model to self-learn and self-optimize.

[0006] In a first aspect, embodiments of this application provide a self-updating method for a synesthetic sensing algorithm, applied in SF, comprising:

[0007] Obtain the first response message sent by the target NWDAF, wherein the first response message includes a model algorithm that meets the SF requirements;

[0008] The initial model algorithm is deployed according to the first response message, and the initial model algorithm is used to analyze and process the sensing information to obtain sensing data, wherein the sensing information is the information reported by the base station.

[0009] Based on the perceived data, a first request message is generated and sent to the target NWDAF. The first request message includes the perceived data.

[0010] Obtain the second response message sent by the target NWDAF, the second response message being used to indicate the processing method of the initial model algorithm;

[0011] The initial model algorithm is updated and saved according to the second response message to obtain the target model algorithm.

[0012] Optionally, before obtaining the first response message sent by the target NWDAF, the method further includes:

[0013] Send a second request message to the NRF, the second request message being used to instruct the NRF to query an NWDAF that meets the SF-aware requirements;

[0014] Obtain a third response message sent by the NRF, the third response message including multiple NWDAF information that conform to the SF requirements;

[0015] Based on the third response message and its own configuration, the target NWDAF is determined, and a third request message is sent to the target NWDAF, the third request message including multiple requirements of the model algorithm.

[0016] Optionally, the step of analyzing and processing the perceived information using the initial model algorithm to obtain perceived data includes:

[0017] Obtain sensing information uploaded by the base station;

[0018] The perceived information is analyzed and processed according to the initial model algorithm to obtain the perceived target and its motion data.

[0019] The application platform is a system for receiving and storing motion information of the perceived target. The system acquires the real data of the perceived target sent by the application platform.

[0020] Determine whether the actual data and the motion data are consistent;

[0021] If the real data is consistent with the motion data, then first perception data is generated, which is used to indicate that the model algorithm meets the perception requirements.

[0022] If the real data is inconsistent with the motion data, then second perception data is generated, and the first perception data is used to indicate that the model algorithm does not meet the perception requirements.

[0023] Optionally, the step of updating and saving the initial model algorithm according to the second response message to obtain the target perception algorithm includes:

[0024] The second response message is parsed and processed to obtain the indication information of the second response message;

[0025] When the indication information is a confirmation information, the current model algorithm is confirmed as the target perception algorithm and the target perception algorithm is saved;

[0026] When the indication information is a model algorithm, the initial model algorithm is updated and optimized according to the model algorithm to obtain a second model algorithm;

[0027] The perceived information is analyzed and processed using the second model algorithm, and the first request message and the second response message are resent until the indication information of the second response message is a confirmation message.

[0028] Secondly, embodiments of this application provide a self-updating method for a synesthetic sensing algorithm, applied to NWDAF, comprising:

[0029] After receiving the third request message sent by SF, determine the model algorithm that matches SF, and send the model algorithm as the first response message to SF;

[0030] After receiving the first request message sent by the SF, the corresponding sensing data is determined based on the first request message;

[0031] When the perceived data is the first perceived data, the current model algorithm is mapped to obtain confirmation information;

[0032] When the perceived data is the second perceived data, the current model algorithm is updated based on the second perceived data to obtain a processing result, which includes the updated model algorithm.

[0033] A second response message is generated based on the confirmation information or processing result, and the second response message is sent to the SF.

[0034] Optionally, the mapping process for the current model algorithm to obtain confirmation information includes:

[0035] Based on the first perceived data, the current model algorithm and the current business requirement are determined. The current model algorithm is the model algorithm that generates the first perceived data, and the current business requirement is the specific information of the second request message sent by the SF.

[0036] A mapping relationship between the current model algorithm and the current business requirement is constructed, and the mapping relationship is stored to obtain confirmation information, which is used to indicate that the model algorithm has been confirmed and stored.

[0037] Thirdly, embodiments of this application provide a self-updating device for a synergistic sensing algorithm, applied in SF, comprising:

[0038] The acquisition module is used to acquire the first response message sent by the target NWDAF, wherein the first response message includes a model algorithm that meets the requirements of the SF;

[0039] The processing module is used to deploy an initial model algorithm according to the first response message, and to analyze and process the sensing information using the initial model algorithm to obtain sensing data, wherein the sensing information is the information reported by the base station.

[0040] The generation module is configured to generate a first request message based on the perceived data and send the first request message to the target NWDAF, wherein the first request message includes the perceived data.

[0041] The acquisition module is further configured to acquire a second response message sent by the target NWDAF, the second response message being used to indicate the processing method of the initial model algorithm;

[0042] The processing module is further configured to update and save the initial model algorithm according to the second response message to obtain the target model algorithm.

[0043] Optionally, the device further includes: a sending module and a confirmation module;

[0044] The sending module is used to send a second request message to the NRF, the second request message being used to instruct the NRF to query an NWDAF that meets the SF awareness requirements;

[0045] The acquisition module is also used to acquire a third response message sent by the NRF, the third response message including multiple NWDAF information that conform to the SF requirements;

[0046] The confirmation module is used to determine the target NWDAF based on the third response message and its own configuration, and send a third request message to the target NWDAF. The third request message includes multiple requirements of the model algorithm.

[0047] Optionally, the device further includes: a determination module;

[0048] The acquisition module is also used to acquire sensing information uploaded by the base station;

[0049] The processing module is further configured to perform algorithmic analysis and processing on the perceived information according to the initial model algorithm to obtain the perceived target and the motion data of the perceived target;

[0050] The acquisition module is also used to acquire the real data of the perceived target sent by the application platform, wherein the application platform is a system for receiving and storing motion information of the perceived target;

[0051] The judgment module is used to determine whether the real data and the motion data are consistent;

[0052] The generation module is further configured to generate first perception data if the real data is consistent with the motion data, and the first perception data is used to indicate that the model algorithm meets the perception requirements.

[0053] The generation module is further configured to generate second perception data if the real data is inconsistent with the motion data, wherein the first perception data is used to indicate that the model algorithm does not meet the perception requirements.

[0054] Optionally, the processing module is further configured to parse the second response message to obtain indication information of the second response message;

[0055] The confirmation module is further configured to confirm the current model algorithm as a target perception algorithm and save the target perception algorithm when the indication information is confirmation information;

[0056] The processing module is further configured to update and optimize the initial model algorithm according to the model algorithm when the indication information is a model algorithm, so as to obtain a second model algorithm;

[0057] The processing module is further configured to analyze and process the perceived information using the second model algorithm, resend the first request message and obtain the second response message, until the indication information of the second response message is confirmation information.

[0058] Fourthly, embodiments of this application provide a self-updating device for a synesthetic sensing algorithm, applied to NWDAF, comprising:

[0059] The determination module is used to determine the model algorithm that conforms to SF after receiving the third request message sent by SF, and send the model algorithm as the first response message to SF;

[0060] The determining module is further configured to determine the corresponding sensing data based on the first request message after receiving the first request message sent by the SF;

[0061] The processing module is used to perform mapping processing on the current model algorithm when the perceived data is the first perceived data, so as to obtain confirmation information;

[0062] The processing module is further configured to update the current model algorithm based on the second sensing data when the sensing data is the second sensing data, and obtain a processing result, the processing result including the updated model algorithm;

[0063] The generation module is used to generate a second response message based on the confirmation information or processing result, and send the second response message to the SF.

[0064] Optionally, the determining module is further configured to determine the current model algorithm and the current business requirement based on the first perceived data, wherein the current model algorithm is the model algorithm that generates the first perceived data, and the current business requirement is the specific information of the second request message sent by the SF;

[0065] The processing module is further configured to construct a mapping relationship between the current model algorithm and the current business requirement, and to store the mapping relationship to obtain confirmation information, which is used to indicate that the model algorithm has been confirmed and stored.

[0066] Fifthly, embodiments of this application provide a self-updating device for a synergistic sensing algorithm, comprising: a memory and a processor;

[0067] The memory stores computer-executed instructions;

[0068] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect, the second aspect, and / or various possible implementations of the first aspect and the second aspect as described above.

[0069] In a sixth aspect, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect, the second aspect, and / or various possible implementations of the first aspect and the second aspect.

[0070] In a seventh aspect, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect, the second aspect, and / or various possible implementations of the first aspect and the second aspect.

[0071] The self-updating method, apparatus, and device for the integrated sensing algorithm provided in this application are applied to an intelligent network sensing and optimization system. The NWDAF first receives a third request message from the SF and, after determining a model algorithm that matches the SF, sends it to the SF. The SF then deploys an initial model algorithm and uses it to determine sensing data before sending it to the NWDAF. The NWDAF processes the sensing data, obtains confirmation information or processing results, and feeds them back to the SF. Finally, the SF updates and saves the initial model algorithm based on the feedback from the NWDAF, obtaining the target model algorithm. This method, through the interaction between the NWDAF and the SF, achieves dynamic determination, deployment, verification, and updating of the model algorithm, effectively improving the accuracy and optimization efficiency of network sensing and promoting continuous optimization of intelligent network performance. Attached Figure Description

[0072] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0073] Figure 1 This is a schematic diagram of the structure of the intelligent network sensing and optimization system shown in this application;

[0074] Figure 2 A flowchart illustrating a self-updating method for a synesthetic sensing algorithm provided in this application. Figure 1 ;

[0075] Figure 3 Signaling interaction diagram of a self-updating method for a synesthetic sensing algorithm provided in this application. Figure 1 ;

[0076] Figure 4 Signaling interaction diagram of a self-updating method for a synesthetic sensing algorithm provided in this application. Figure 2 ;

[0077] Figure 5 A flowchart illustrating a self-updating method for a synesthetic sensing algorithm provided in this application. Figure 2 ;

[0078] Figure 6 A flowchart illustrating a self-updating method for a synesthetic sensing algorithm provided in this application. Figure 3 ;

[0079] Figure 7 A flowchart illustrating a self-updating method for a synesthetic sensing algorithm provided in this application. Figure 4 ;

[0080] Figure 8 A schematic diagram of the structure of a self-updating device for a synesthetic sensing algorithm provided in this application. Figure 1 ;

[0081] Figure 9 A schematic diagram of the structure of a self-updating device for a synesthetic sensing algorithm provided in this application. Figure 2 ;

[0082] Figure 10 This is a schematic diagram of the structure of a self-updating device for a synesthetic sensing algorithm provided in this application.

[0083] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0084] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0085] With the rapid development of 5G technology, 5G-A, as an evolution of 5G, is gradually becoming a research hotspot in the field of communications. 5G-A not only inherits the high speed, large capacity, and low latency characteristics of 5G, but also further explores the integration of communication and sensing, proposing a unified sensing network model. This model aims to leverage the technological advantages of 5G base stations, such as massive MIMO (massive multiple-input multiple-output), to achieve radar-like sensing functions, thereby expanding the application scenarios of 5G networks and supporting the development of emerging fields such as low-altitude economy, traffic management, park management, intrusion detection, and smart cities. The unified sensing technology senses objects in a connectionless manner, without requiring the sensed object to interact with the network, providing a new solution for the integration of communication and sensing.

[0086] In the 5G-A integrated sensing network model, the 5G base station on the wireless network side plays a core role. The base station utilizes its communication technology to perform real-time signal scanning of objects within its service range and receives signals reflected from these objects to obtain information such as object type, location, speed, and route. This information is then passed to the newly added sensing function network element (SF). The SF is responsible for receiving and processing the sensing information fed back from the base station and calculating and outputting information about the sensed objects according to application requirements. The SF can be deployed flexibly and diversely; it can be an independent network element or co-located with other network elements, and it supports centralized or distributed deployment. In current 5G-A integrated sensing devices, the analytical computing capabilities are mainly deployed in the base station's BBU (Baseband Processing Unit) sensing board and the SF. These devices are deployed once and then upgraded later to update and optimize the algorithm model.

[0087] While current 5G-A sensing technology has made significant progress in achieving communication and sensing fusion, it still faces several technical challenges. Especially in the early stages of development, the existing manual model upgrade method is inflexible and inconvenient due to the need for frequent performance optimization. This upgrade method not only increases maintenance costs but may also affect system stability and reliability. Therefore, how to more efficiently update and optimize algorithm models has become a critical issue that urgently needs to be addressed in the development of current 5G-A sensing technology.

[0088] To address the aforementioned issues, this application proposes a self-updating method for an integrated sensing algorithm. The SF (Sensing Module) interacts with the base station to independently determine whether its model algorithm meets current service requirements. If it does not, it sends a model algorithm update request to the NWDAF (Near-Wide Grounding AF), enabling the NWDAF to update the model algorithm accordingly. This method automatically upgrades and optimizes sensing capabilities through continuous self-learning and self-optimization of the algorithm model during actual sensing operations, thus automatically and conveniently improving sensing capabilities and accuracy.

[0089] Figure 1 This is a schematic diagram of the intelligent network sensing and optimization system shown in this application. The following is in conjunction with… Figure 1 The structure of the intelligent network sensing and optimization system is described in detail.

[0090] like Figure 1 As shown, the system integrates the Sensing Function (SF), the Network Data Analytics Function (NWDAF), and the Network Repository Function (NRF).

[0091] The SF (Sensing Component) is responsible for executing actual sensing services, such as data acquisition and preprocessing. Through feedback interaction with the NWDAF (Network Window Assistance Component), it provides real-time performance data for the sensing model. The NWDAF is responsible for processing and analyzing data collected from the SF and other network functions. It uses algorithmic models to perform in-depth data analysis, generate optimization suggestions, and continuously optimizes the algorithm model based on real-time performance data through feedback interaction with the SF. The NRF (Network Response Component) stores and manages various functional entities in the network, such as the SF and NWDAF, supporting the deployment of the SF and network element registration, ensuring correct connection and interoperability of network functions. The system can automatically deploy the SF and complete network element registration, ensuring that newly added SFs can quickly integrate into the network. The system supports interoperability between different network functions, ensuring smooth data flow and sharing. By optimizing network configuration and protocols, it achieves efficient network operation and rapid fault recovery.

[0092] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0093] Figure 2 A flowchart illustrating a self-updating method for a synesthetic sensing algorithm provided in this application embodiment. Figure 1 Applied to SF, such as Figure 2 As shown, the method includes:

[0094] S101: Obtain the first response message sent by the target NWDAF.

[0095] The first response message includes a model algorithm that meets SF requirements.

[0096] Understandably, before acquiring the model, SF proactively requests NWDAF to obtain a model algorithm that meets SF's specific needs. As a core component of network data analysis, NWDAF can provide corresponding algorithm support based on SF's request. The first response message is NWDAF's reply to SF's request, containing carefully selected and configured model algorithms matched to SF's specific requirements (such as data processing capabilities, accuracy requirements, and real-time performance). In this way, SF obtains an initial model algorithm that meets its business needs and can operate efficiently.

[0097] S102: Deploy the initial model algorithm based on the first response message, and use the initial model algorithm to analyze and process the perceived information to obtain perceived data.

[0098] Among them, the sensing information is the information reported by the base station.

[0099] Understandably, upon receiving the first response message from NWDAF containing the model algorithm, SF immediately began deploying this initial model algorithm. After deployment, SF began using this algorithm to analyze and process the sensing information from the base station. Through the processing of the initial model algorithm, the raw sensing information from the base station was transformed into valuable sensing data, providing a foundation for subsequent network optimization, decision support, and other processes.

[0100] S103: Based on the sensing data, generate a first request message and send the first request message to the target NWDAF. The first request message includes the sensing data.

[0101] Understandably, after acquiring the sensing data, SF will generate a new request message, namely the first request message, based on this data. The main purpose of this request message is to feed the sensing data back to NWDAF so that NWDAF can further analyze the data and evaluate and adjust the initial model algorithm based on the analysis results. By sending the first request message containing the sensing data to NWDAF, SF achieves data sharing and collaborative work with NWDAF, jointly promoting the improvement of network intelligence.

[0102] S104: Obtain the second response message sent by the target NWDAF.

[0103] The second response message is used to indicate how the initial model algorithm should be processed.

[0104] Understandably, after receiving the first request message containing sensing data from the SF, NWDAF will conduct in-depth analysis of this data and propose specific suggestions on the processing method of the initial model algorithm based on the analysis results. These suggestions are encapsulated in a second response message and sent back to the SF. The second response message not only contains the evaluation results of the initial model algorithm's performance but may also propose optimization suggestions or new processing strategies. By receiving and interpreting the second response message, the SF can understand the performance of the initial model algorithm in practical applications and make corresponding adjustments accordingly.

[0105] S105: Update and save the initial model algorithm according to the second response message to obtain the target model algorithm.

[0106] Finally, SF updates and optimizes the initial model algorithm based on the suggestions or strategies proposed in the second response message sent by NWDAF. This step may involve adjusting algorithm parameters, optimizing the algorithm structure, or introducing new algorithm components. After the update process, SF will obtain a more complete and higher-performing target model algorithm. This target model algorithm will be saved and will play an important role in subsequent network analysis and optimization work.

[0107] The self-updating method of the integrated sensing algorithm provided in this application is applied to the SF (Signal Provider). The SF obtains a first response message sent by the target NWDAF (Near-Wideband Data Center), deploys an initial model algorithm based on the first response message, and analyzes and processes the sensing information using the initial model algorithm to obtain sensing data. The sensing information is the information reported by the base station. Based on the sensing data, a first request message is generated and sent to the target NWDAF. A second response message sent by the target NWDAF is then obtained, indicating the processing method of the initial model algorithm. The initial model algorithm is updated and saved based on the second response message to obtain the target model algorithm. This method, through interaction with the NWDAF, simultaneously sends sensing data and correct data to the NWDAF, enabling the NWDAF to self-update and optimize the model algorithm based on the data, thus improving the optimization and updating efficiency of the model algorithm.

[0108] Figure 3 Signaling interaction diagram of a self-updating method for a synesthetic sensing algorithm provided in this application embodiment. Figure 1 ,like Figure 3 As shown, in this embodiment... Figure 2 Based on the embodiments, the interaction process between SF and NWDAF is described in detail. The method includes:

[0109] S201: After NWDAF receives the third request message sent by SF, it determines the model algorithm that conforms to SF.

[0110] S202: NWDAF sends the model algorithm as the first response message to SF.

[0111] Understandably, NWDAF's main task at this step is to receive and analyze the third request message sent by SF to determine a model algorithm that meets SF's specific needs and business scenario. NWDAF utilizes its powerful data processing and analysis capabilities to parse and evaluate the information in the request message, then selects the most suitable model algorithm and sends it back to SF as the first response message.

[0112] S203: SF deploys the initial model algorithm based on the first response information and uses the initial model algorithm to determine the sensing data.

[0113] S204: SF sends the perceived data as the first request message to NWDAF.

[0114] Steps S203-S204 are similar to steps S102-S103, and will not be described again here.

[0115] S205: NWDAF determines the corresponding sensing data based on the first request message, processes the sensing data, and obtains confirmation information or processing results.

[0116] S206: NWDAF sends the confirmation message or processing result to SF as a second response message.

[0117] Understandably, in this step, the main task of NWDAF is to receive the first request message sent by SF and determine the sensing data based on this message. Subsequently, NWDAF performs detailed processing and analysis on this sensing data. If the sensing data contains indications that the algorithm is correct, NWDAF will generate a confirmation message; if the sensing data is incorrect, the current model algorithm will be updated, and the updated model algorithm will be used as the processing result. Finally, NWDAF sends this confirmation message or processing result back to SF as a second response message, thereby ensuring the accuracy and integrity of the algorithm, or providing corrected and optimized information.

[0118] S207: SF updates and saves the initial model algorithm based on the second response message to obtain the target model algorithm.

[0119] Step S207 is similar to step S105, and will not be described again here.

[0120] This application provides a self-updating method for a syn-sensing integrated sensing algorithm, which is applied to an intelligent network sensing and optimization system. Through the interaction between NWDAF and SF, the dynamic determination, deployment, verification and updating of the model algorithm are realized, which effectively improves the accuracy and optimization efficiency of network sensing and promotes the continuous optimization of intelligent network performance.

[0121] Figure 4 Signaling interaction diagram of a self-updating method for a synesthetic sensing algorithm provided in this application embodiment. Figure 2 ,like Figure 4 As shown, in this embodiment... Figure 2 Based on the embodiments, the process of determining the target NWDAF using SF and NRF is described in detail. The method includes:

[0122] S301: SF sends a second request message to NRF.

[0123] The second request message is used to instruct the NRF to query the NWDAF that meets the SF-aware requirements.

[0124] Understandably, SF needs to find one or more Network Data Analysis Function (NWDAF) instances that meet its business needs or perception requirements. To achieve this, SF constructs a second request message that explicitly includes its specific perception requirements for NWDAF, such as data processing capabilities, response speed, and geographic location preferences. SF then sends this request message to NRF, requesting NRF to help query and return NWDAF information that meets these conditions.

[0125] S302: The NRF searches for multiple NWDAF information that meet the SF requirements based on the second request message.

[0126] Understandably, upon receiving the second request message from SF, NRF will begin searching its maintained Network Function Registration Information (NFRAI) database for matching NWDAF instances based on the SF awareness requirements provided in the message. This process may involve matching and analyzing multiple dimensions of the NWDAF instance, such as its capability description, current load, and geographical location. Ultimately, NRF will filter out information on multiple NWDAF instances that meet SF's requirements. This information may include key data such as the NWDAF's identifier, address, and supported interface protocols.

[0127] S303: NRF sends multiple NWDAF messages as a third response message to SF.

[0128] Understandably, after successfully finding NWDAF instance information that matches SF's requirements, NRF will encapsulate this information into a third-party response message and send it to SF over the network. This response message contains detailed information about all the NWDAF instances that meet the criteria found by NRF, enabling SF to fully understand the capabilities and status of these NWDAF instances. This step is NRF's direct response to SF's request and is also the basis for SF's subsequent selection of target NWDAFs.

[0129] S304: SF determines the target NWDAF based on the third response message and its own configuration.

[0130] Understandably, upon receiving the third response message from the NRF, SF will conduct further analysis and evaluation based on the NWDAF instance information provided in the message, combined with its own business logic, performance requirements, cost considerations, and other factors. This process may involve comprehensive consideration of multiple aspects such as the NWDAF instance's response time, data processing capabilities, reliability, and geographical location. Ultimately, SF will select one or more optimal NWDAF instances as target NWDAFs for subsequent data analysis and processing tasks.

[0131] S305: SF sends a third request message to the target NWDAF.

[0132] The third request message includes multiple requirements for the model algorithm.

[0133] Understandably, once the target NWDAF is identified, SF will construct a third request message based on specific business needs. This message details the specific requirements for the model algorithm that the target NWDAF needs to execute. These requirements may include key information such as the algorithm type, input data format, expected output results, and processing time requirements. Subsequently, SF sends this request message to the target NWDAF, requesting it to perform data analysis and processing according to these requirements. This step marks the formal commencement of business interaction and data flow between SF and the target NWDAF.

[0134] This application provides a self-updating method for an integrated sensing algorithm, applied to an intelligent network sensing and optimization system. The method involves the SF (Self-Focusing Provider) requesting and identifying NWDAFs (Network Controller Area Functions) that meet its sensing requirements from the NRF (Network Controller Area Function). Subsequently, the SF sends a request containing model algorithm requirements to the selected NWDAF, achieving efficient matching and distribution of model algorithm requirements within the intelligent network sensing and optimization system. This process optimizes resource allocation, improves the flexibility and accuracy of network sensing, and accelerates the formulation and execution of network optimization decisions.

[0135] Figure 5 A flowchart illustrating a self-updating method for a synesthetic sensing algorithm provided in this application embodiment. Figure 2 ,like Figure 5 As shown, this embodiment is applied to SF, in Figure 2 Based on the embodiments, the process of SF determining sensing data is described in detail, and the method includes:

[0136] S401: Obtain sensing information uploaded by the base station.

[0137] Understandably, the sensing information uploaded by base stations refers to the data collected and reported by base stations (usually devices deployed in specific locations for wireless communication and / or sensing) to the system center or data processing unit. Sensing information typically includes signal strength, signal characteristics, and other information. This information is uploaded to the data processing center via the base station, providing raw material for subsequent analysis and processing.

[0138] S402: Based on the initial model algorithm, perform algorithmic analysis and processing on the perceived information to obtain the perceived target and its motion data.

[0139] Understandably, analyzing and processing the sensed information using the initial model algorithm is a crucial step in extracting valuable information. In this step, the pre-defined model algorithm performs in-depth mining and intelligent analysis of the sensed information uploaded by the base station, thereby identifying the sensed target (such as a specific user or object) and its motion data (such as trajectory, speed changes, etc.). This process relies on advanced algorithm technology and computing power to achieve efficient and accurate information extraction.

[0140] S403: Obtain the real data of the perceived target sent by the application platform.

[0141] The application platform is a system for receiving and storing motion information of perceived targets.

[0142] Understandably, obtaining the actual motion data of the perceived target sent by the application platform is to verify the accuracy of the algorithm's analysis results. The application platform, as a system specifically designed to receive and store motion information of the perceived target, can provide the actual motion data of the perceived target. This data serves as the "gold standard" and is used to compare with the motion data obtained from the algorithm's analysis to evaluate the algorithm's accuracy and reliability.

[0143] S404: Determine whether the actual data and motion data are consistent. If yes, proceed to step S405; otherwise, proceed to step S406.

[0144] Understandably, determining whether the real data matches the motion data is a crucial step in ensuring algorithm performance. By comparing the real data with the motion data obtained from algorithm analysis, the algorithm's performance can be intuitively evaluated. If they match, it means the algorithm accurately reflects the actual motion of the perceived target; if they don't match, it indicates that the algorithm has errors or shortcomings and needs further improvement and optimization.

[0145] S405: Generate the first perception data, which is used to indicate that the model algorithm meets the perception requirements.

[0146] Understandably, when the motion data obtained from algorithm analysis matches the real data, it indicates that the current model algorithm can effectively process perceived information and accurately extract valuable data. This initial perceived data not only proves the effectiveness of the algorithm but also provides reliable data support for subsequent business applications.

[0147] S406: Generate second perception data. The first perception data is used to indicate that the model algorithm does not meet the perception requirements.

[0148] Understandably, when the motion data obtained from algorithm analysis is inconsistent with the real data, this inconsistent data is categorized as secondary sensing data. This data reveals problems and shortcomings in the algorithm, providing valuable feedback for subsequent algorithm optimization. This data is then sent to systems such as Network Data Analysis Function (NWDAF) to adjust and improve the algorithm, thereby enhancing its accuracy and reliability.

[0149] This application provides a self-updating method for an integrated sensing algorithm, applied to SF (Signal Server). It acquires sensing information uploaded by a base station, analyzes and processes the sensing information according to an initial model algorithm, and obtains the sensing target and its motion data. It also acquires real data of the sensing target sent by an application platform, which is a system for receiving and storing motion information of the sensing target. The method determines whether the real data and motion data are consistent. If they are, first sensing data is generated, indicating that the model algorithm meets the sensing requirements; otherwise, second sensing data is generated, indicating that the model algorithm does not meet the sensing requirements. This method efficiently acquires and analyzes sensing information uploaded by the base station, verifies it using real data from the application platform, accurately determines whether the model algorithm meets the sensing requirements, and thus quickly generates accurate sensing data, effectively improving the intelligence and accuracy of network sensing.

[0150] Figure 6 A flowchart illustrating a self-updating method for a synesthetic sensing algorithm provided in this application embodiment. Figure 3 ,like Figure 6 As shown, in this embodiment... Figure 2Based on the embodiments, the process of SF target perception algorithm is described in detail, which includes:

[0151] S501: Parse and process the second response message to obtain the indication information of the second response message.

[0152] Understandably, upon receiving the second response message from NWDAF, SF will first perform detailed parsing and processing. This step aims to extract key indicative information from the response message. The indicative information in the second response message mainly falls into two categories: one is confirmation information, indicating that the currently submitted model algorithm has been confirmed and saved by NWDAF, meaning that the algorithm can be directly invoked when encountering the same business requirements in the future without resubmitting; the other is the processing result, which usually refers to the result after updating or optimizing the model algorithm, such as an updated model algorithm. This design aims to improve the efficiency and flexibility of algorithm usage.

[0153] S502: Determine whether the instruction information is a confirmation message. If yes, proceed to step S503; otherwise, proceed to step S504.

[0154] Understandably, after successfully parsing the indication information in the second response message, SF will proceed to the next step. It first checks whether the indication information is a confirmation message. If it indicates that the algorithm has been confirmed and saved, this means that the currently submitted model algorithm meets the NWDAF requirements and can be considered a valid target perception algorithm. SF will then execute step S503. If the indication information is not a confirmation message but indicates that the model algorithm needs further optimization or updating, SF will proceed to step S504.

[0155] S503: Confirm the current model algorithm as a target perception algorithm and save the target perception algorithm.

[0156] Understandably, once the confirmation message indicates that the current model algorithm has been accepted and saved by NWDAF, SF will officially recognize this algorithm as a target-aware algorithm. This step is a crucial part of the algorithm's lifecycle, signifying that the algorithm has passed the necessary verification and review and can be officially used in business processing. Simultaneously, SF will also be responsible for storing this target-aware algorithm to ensure it can be quickly invoked when needed to meet business requirements. Storage methods may include storing the algorithm in a local database or on a remote server for rapid access.

[0157] S504: Update and optimize the initial model algorithm based on the model algorithm to obtain the second model algorithm.

[0158] Understandably, if the indication in the second response message suggests that the model algorithm needs to be updated and optimized, SF will initiate an update and optimization process. This step may involve various techniques and methods, such as machine learning algorithm tuning, parameter adjustment, and feature selection, aiming to improve the algorithm's performance or adaptability. After this series of optimization processes, SF will obtain a new, improved second model algorithm. This algorithm should theoretically be more efficient or accurate than the previous algorithm, better meeting business needs.

[0159] S505: Analyze and process the perceived information using the second model algorithm, resend the first request message and obtain the second response message until the indication information of the second response message is confirmation information.

[0160] Understandably, after obtaining the updated second model algorithm, SF will use this new algorithm to analyze and process the sensed information. This step is crucial for verifying the performance of the new algorithm and ensuring its effectiveness in real-world business applications. After processing the sensed information, SF may resend the first request message to NWDAF based on the analysis results to obtain a new response message and indication information. This process may be repeated multiple times until the indication information returned by NWDAF indicates that the current algorithm has been confirmed and saved. This cyclical process ensures that the algorithm can be continuously optimized and improved to adapt to current business needs and environmental conditions.

[0161] This application provides a self-updating method for a sensory integration algorithm, applied to SF (Sensitive Sensory Architecture). The method involves parsing a second response message to obtain its indication information. It then determines whether the indication information is a confirmation message. If so, the current model algorithm is confirmed as the target sensory algorithm and saved. If not, the initial model algorithm is updated and optimized based on the previous model algorithm to obtain a second model algorithm. This second model algorithm is then used to analyze and process the sensory information, and the first request message and second response message are resent until the indication information in the second response message is a confirmation message. This method analyzes the messages transmitted by NWDAF (Non-Wideband Assisted Sensory Architecture) to determine whether the current model algorithm needs updating, ensuring the real-time accuracy of the algorithm model.

[0162] Figure 7 A flowchart illustrating a self-updating method for a synesthetic sensing algorithm provided in this application embodiment. Figure 4 ,like Figure 7 As shown, in this embodiment... Figure 3 Based on the embodiments, the process of NWDAF according to the perceptual data processing model algorithm is described in detail. The method includes:

[0163] S601: After receiving the first request message sent by SF, determine the corresponding sensing data based on the first request message.

[0164] Understandably, in the NWDAF workflow, once the first request message from SF is successfully captured, this step marks the start of the data processing flow. NWDAF first parses the request message to extract key information, which is used to locate and acquire the corresponding sensing data. Sensing data is crucial information in the network operation process and is essential for subsequent analysis and decision-making.

[0165] S602: Determine whether the perceived data is the first perceived data. If yes, proceed to step S603; otherwise, proceed to step S605.

[0166] Understandably, after acquiring the sensing data, NWDAF will further determine the specific type of the sensing data and proceed with the next step based on the specific type of the sensing data.

[0167] S603: Based on the first perception data, determine the current model algorithm and the current business requirements.

[0168] The current model algorithm is the model algorithm that generates the first perception data, and the current business requirement is the specific information of the second request message sent by SF.

[0169] Understandably, for the first perception data, NWDAF will determine the current model algorithm and the matching current business requirements based on the data's characteristics and content. Here, the current model algorithm refers to the algorithm that generated this perception data, while the current business requirements are obtained by parsing the second request message sent by SF, which contains specific business requirements and expected analysis results.

[0170] S604: Construct a mapping relationship between the current model algorithm and the current business requirements, store the mapping relationship, and obtain confirmation information.

[0171] The confirmation information is used to indicate that the model algorithm has been confirmed and stored.

[0172] Understandably, after determining the current model algorithm and current business requirements, NWDAF will construct a mapping relationship between the two and store this mapping relationship. The purpose of this step is to ensure that the corresponding model algorithm can be quickly found based on business needs, thereby improving the efficiency of data processing and analysis. For example, when the next SF request with the same business requirements determines the model algorithm, this algorithm model can be sent first. After storage processing is complete, NWDAF will generate a confirmation message indicating that the model algorithm has been confirmed and successfully stored, providing a foundation for subsequent data processing and analysis.

[0173] S605: Update the current model algorithm based on the second perception data to obtain the processing result.

[0174] The processing results include the updated model algorithm.

[0175] Understandably, if the perceived data is not the first perception data but the second perception data, which includes motion data and the difference between the motion data and the real data, NWDAF will update the current model algorithm based on this perceived data and the difference data. The update process may include adjusting model parameters, optimizing the algorithm structure, or introducing new algorithms to achieve better data processing and analysis results. After processing, NWDAF will generate a result containing the updated model algorithm.

[0176] S606: Generate a second response message based on the confirmation information or processing result, and send the second response message to SF.

[0177] Understandably, NWDAF will eventually construct a second response message based on the confirmation information or processing results generated in the previous steps and send it back to SF. This response message contains the final result or status information of NWDAF's data processing, such as confirmation information of the model algorithm, the updated model algorithm, etc. In this way, SF can understand the progress and results of NWDAF's data processing based on this response message, and thus make corresponding decisions or adjustments.

[0178] This application provides a self-updating method for a sensory integration algorithm, applied to NWDAF. It determines the corresponding sensing data based on a first request message. If the sensing data is the first sensing data, and if so, it determines the current model algorithm and the current business requirement based on the first sensing data. The current model algorithm is the one that generated the first sensing data, and the current business requirement is the specific information in the second request message sent by the SF (Sensitive Message). A mapping relationship between the current model algorithm and the current business requirement is constructed and stored to obtain confirmation information, which indicates that the model algorithm has been confirmed and stored. If not, the current model algorithm is updated based on the second sensing data to obtain a processing result, which includes the updated model algorithm. A second response message is generated based on the confirmation information or the processing result and sent to the SF. This method analyzes the sensing data to determine the accuracy of the model algorithm. When the model algorithm is inaccurate, it is optimized based on the sensing data; when the model algorithm is accurate, it is saved. This achieves self-learning and self-optimization of sensing capabilities, saving labor costs.

[0179] Figure 8A schematic diagram of the structure of a self-updating device for a synesthetic sensing algorithm provided in this application. Figure 1 Applied to SF, such as Figure 8 As shown, the self-updating device 70 for the integrated sensing algorithm provided in this embodiment includes:

[0180] The acquisition module 701 is used to acquire the first response message sent by the target NWDAF, wherein the first response message includes a model algorithm that meets the SF requirements;

[0181] The processing module 702 is used to deploy an initial model algorithm according to the first response message, and to analyze and process the sensing information using the initial model algorithm to obtain sensing data, wherein the sensing information is information reported by the base station.

[0182] The generation module 703 is configured to generate a first request message based on the perceived data and send the first request message to the target NWDAF, wherein the first request message includes the perceived data.

[0183] The acquisition module 701 is further configured to acquire a second response message sent by the target NWDAF, the second response message being used to indicate the processing method of the initial model algorithm;

[0184] The processing module 702 is further configured to update and save the initial model algorithm according to the second response message to obtain the target model algorithm.

[0185] Optionally, the device further includes: a sending module 704 and a confirmation module 705;

[0186] The sending module 704 is used to send a second request message to the NRF, the second request message being used to instruct the NRF to query an NWDAF that meets the SF awareness requirements;

[0187] The acquisition module 701 is further configured to acquire a third response message sent by the NRF, the third response message including multiple NWDAF information that conform to the SF requirements;

[0188] The confirmation module 705 is used to determine the target NWDAF based on the third response message and its own configuration, and send a third request message to the target NWDAF, the third request message including multiple requirements of the model algorithm.

[0189] Optionally, the device further includes: a determination module 706;

[0190] The acquisition module 701 is also used to acquire sensing information uploaded by the base station;

[0191] The processing module 702 is further configured to perform algorithmic analysis and processing on the perceived information according to the initial model algorithm to obtain the perceived target and the motion data of the perceived target;

[0192] The acquisition module 701 is also used to acquire the real data of the perceived target sent by the application platform, wherein the application platform is a system for receiving and storing motion information of the perceived target.

[0193] The judgment module 706 is used to determine whether the real data and the motion data are consistent;

[0194] The generation module 703 is further configured to generate first perception data if the real data is consistent with the motion data, wherein the first perception data is used to indicate that the model algorithm meets the perception requirements;

[0195] The generation module 703 is further configured to generate second perception data if the real data is inconsistent with the motion data, wherein the first perception data is used to indicate that the model algorithm does not meet the perception requirements.

[0196] Optionally, the processing module 702 is further configured to parse the second response message to obtain indication information of the second response message;

[0197] The confirmation module 705 is further configured to confirm the current model algorithm as a target perception algorithm and save the target perception algorithm when the indication information is confirmation information;

[0198] The processing module 702 is further configured to update and optimize the initial model algorithm according to the model algorithm when the indication information is a model algorithm, so as to obtain a second model algorithm;

[0199] The processing module 702 is further configured to analyze and process the perceived information using the second model algorithm, resend the first request message and obtain the second response message until the indication information of the second response message is confirmation information.

[0200] The self-updating device for the integrated sensing algorithm provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0201] Figure 9 A schematic diagram of the structure of a self-updating device for a synesthetic sensing algorithm provided in this application. Figure 2 Applied to NWDAF, such as Figure 9 As shown, the self-updating device 80 for the integrated sensing algorithm provided in this embodiment includes:

[0202] The determining module 801 is used to determine the model algorithm that matches SF after receiving the third request message sent by SF, and send the model algorithm as the first response message to SF;

[0203] The determining module 801 is further configured to determine the corresponding sensing data based on the first request message after obtaining the first request message sent by the SF;

[0204] Processing module 802 is used to perform mapping processing on the current model algorithm to obtain confirmation information when the perceived data is the first perceived data;

[0205] The processing module 802 is further configured to update the current model algorithm based on the second sensing data when the sensing data is the second sensing data, and obtain a processing result, wherein the processing result includes the updated model algorithm;

[0206] The generation module 803 is used to generate a second response message based on the confirmation information or processing result, and send the second response message to the SF.

[0207] Optionally, the determining module 801 is further configured to determine the current model algorithm and the current business requirement based on the first perception data, wherein the current model algorithm is the model algorithm that generates the first perception data, and the current business requirement is the specific information of the second request message sent by the SF;

[0208] The processing module 802 is further configured to construct a mapping relationship between the current model algorithm and the current business requirement, and to store the mapping relationship to obtain confirmation information, which is used to indicate that the model algorithm has been confirmed and stored.

[0209] The self-updating device for the integrated sensing algorithm provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0210] Figure 10 This is a schematic diagram of the structure of a self-updating device for a synesthetic sensing algorithm provided in this application. Figure 10 As shown, the electronic device 90 provided in this embodiment includes at least one processor 901 and a memory 902. Optionally, the device 90 further includes a communication component 903. The processor 901, memory 902, and communication component 903 are connected via a bus 904.

[0211] In a specific implementation, at least one processor 901 executes computer execution instructions stored in memory 902, causing at least one processor 901 to perform the above-described method.

[0212] The specific implementation process of processor 901 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0213] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0214] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0215] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0216] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0217] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0218] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0219] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0220] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0221] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0222] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0223] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0224] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0225] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A self-updating method of a common-sense integrated sensing algorithm, characterized in that, Applied to SF, including: Obtain the first response message sent by the target NWDAF, wherein the first response message includes a model algorithm that meets the SF requirements; The initial model algorithm is deployed according to the first response message, and the sensing information uploaded by the base station is obtained; the sensing information is analyzed and processed according to the initial model algorithm to obtain the sensing target and the motion data of the sensing target. The application platform is a system for receiving and storing motion information of the perceived target. The system acquires the real data of the perceived target sent by the application platform. Determine whether the actual data and the motion data are consistent; If the real data is consistent with the motion data, then first perception data is generated, which is used to indicate that the model algorithm meets the perception requirements. If the real data is inconsistent with the motion data, second sensing data is generated. The first sensing data is used to indicate that the model algorithm does not meet the sensing requirements. The sensing information is the information reported by the base station. Based on the perceived data, a first request message is generated and sent to the target NWDAF. The first request message includes the perceived data, so that after the target NWDAF receives the first request message sent by the SF, it determines the corresponding perceived data based on the first request message. When the perceived data is the first perceived data, it performs mapping processing on the current model algorithm to obtain confirmation information. When the perceived data is the second perceived data, it performs update processing on the current model algorithm based on the second perceived data to obtain a processing result. The processing result includes the updated model algorithm. Based on the confirmation information or the processing result, a second response message is generated and sent to the SF. Obtain the second response message sent by the target NWDAF, the second response message being used to indicate the processing method of the initial model algorithm; The initial model algorithm is updated and saved according to the second response message to obtain the target model algorithm.

2. The method of claim 1, wherein, Before obtaining the first response message sent by the target NWDAF, the method further includes: Send a second request message to the NRF, the second request message being used to instruct the NRF to query an NWDAF that meets the SF-aware requirements; Obtain a third response message sent by the NRF, the third response message including multiple NWDAF information that conform to the SF requirements; Based on the third response message and its own configuration, the target NWDAF is determined, and a third request message is sent to the target NWDAF, the third request message including multiple requirements of the model algorithm.

3. The method of claim 1, wherein, The step of updating and saving the initial model algorithm according to the second response message to obtain the target perception algorithm includes: The second response message is parsed and processed to obtain the indication information of the second response message; When the indication information is a confirmation information, the current model algorithm is confirmed as the target perception algorithm and the target perception algorithm is saved; When the indication information is a model algorithm, the initial model algorithm is updated and optimized according to the model algorithm to obtain a second model algorithm; The perceived information is analyzed and processed using the second model algorithm, and the first request message and the second response message are resent until the indication information of the second response message is a confirmation message.

4. A self-updating method of a common-sense integrated sensing algorithm, characterized in that, Applied to target NWDAF, including: After receiving the third request message sent by the SF, a model algorithm that matches the SF is determined, and the model algorithm is sent to the SF as a first response message. This allows the SF to deploy an initial model algorithm based on the first response message, obtain sensing information uploaded by the base station, perform algorithmic analysis and processing on the sensing information according to the initial model algorithm, obtain the sensing target and the motion data of the sensing target, and obtain the real data of the sensing target sent by the application platform. The application platform is a system for receiving and storing the motion information of the sensing target. It determines whether the real data and the motion data are consistent. If the real data and the motion data are consistent, first sensing data is generated. The first sensing data is used to indicate that the model algorithm meets the sensing requirements. If the real data and the motion data are inconsistent, second sensing data is generated. The first sensing data is used to indicate that the model algorithm does not meet the sensing requirements. The sensing information is the information reported by the base station. Based on the sensing data, a first request message is generated and sent to the target NWDAF. The first request message includes the sensing data. After receiving the first request message sent by the SF, the corresponding sensing data is determined based on the first request message; When the perceived data is the first perceived data, the current model algorithm is mapped to obtain confirmation information; When the perceived data is the second perceived data, the current model algorithm is updated based on the second perceived data to obtain a processing result, which includes the updated model algorithm. A second response message is generated based on the confirmation information or processing result, and the second response message is sent to the SF.

5. The method of claim 4, wherein, The mapping process for the current model algorithm to obtain confirmation information includes: Based on the first perceived data, the current model algorithm and the current business requirement are determined. The current model algorithm is the model algorithm that generates the first perceived data, and the current business requirement is the specific information of the second request message sent by the SF. A mapping relationship between the current model algorithm and the current business requirement is constructed, and the mapping relationship is stored to obtain confirmation information, which is used to indicate that the model algorithm has been confirmed and stored.

6. A self-updating device for a synesthetic sensing algorithm, characterized in that, Applied to SF, including: The acquisition module is used to acquire the first response message sent by the target NWDAF, wherein the first response message includes a model algorithm that meets the requirements of the SF; The processing module is configured to deploy an initial model algorithm based on the first response message and acquire sensing information uploaded by the base station; perform algorithmic analysis and processing on the sensing information according to the initial model algorithm to obtain the sensing target and the motion data of the sensing target; acquire the real data of the sensing target sent by the application platform, wherein the application platform is a system for receiving and storing the motion information of the sensing target; determine whether the real data and the motion data are consistent; if the real data and the motion data are consistent, generate first sensing data, which is used to indicate that the model algorithm meets the sensing requirements; if the real data and the motion data are inconsistent, generate second sensing data, wherein the first sensing data is used to indicate that the model algorithm does not meet the sensing requirements, wherein the sensing information is information reported by the base station; The generation module is configured to generate a first request message based on the perceived data and send the first request message to the target NWDAF. The first request message includes the perceived data, so that after the target NWDAF receives the first request message sent by the SF, it determines the corresponding perceived data based on the first request message. When the perceived data is the first perceived data, it performs mapping processing on the current model algorithm to obtain confirmation information. When the perceived data is the second perceived data, it performs update processing on the current model algorithm based on the second perceived data to obtain a processing result. The processing result includes the updated model algorithm. The module generates a second response message based on the confirmation information or the processing result and sends the second response message to the SF. The acquisition module is further configured to acquire a second response message sent by the target NWDAF, the second response message being used to indicate the processing method of the initial model algorithm; The processing module is further configured to update and save the initial model algorithm according to the second response message to obtain the target model algorithm.

7. A self-updating device of a common-sense integrated sensing algorithm, characterized in that, Applied to target NWDAF, including: The determination module is used to determine the model algorithm that conforms to the SF after receiving the third request message sent by the SF, and send the model algorithm as a first response message to the SF, so that the SF can deploy an initial model algorithm according to the first response message, obtain the sensing information uploaded by the base station, perform algorithm analysis and processing on the sensing information according to the initial model algorithm to obtain the sensing target and the motion data of the sensing target, obtain the real data of the sensing target sent by the application platform, the application platform is a system for receiving and storing the motion information of the sensing target, and determine whether the real data and the motion data are consistent. If the real data and the motion data are consistent, first sensing data is generated. The first sensing data is used to indicate that the model algorithm meets the sensing requirements. If the real data and the motion data are inconsistent, second sensing data is generated. The first sensing data is used to indicate that the model algorithm does not meet the sensing requirements. The sensing information is the information reported by the base station. According to the sensing data, a first request message is generated and sent to the target NWDAF. The first request message includes the sensing data. The determining module is further configured to determine the corresponding sensing data based on the first request message after receiving the first request message sent by the SF; The processing module is used to perform mapping processing on the current model algorithm when the perceived data is the first perceived data, so as to obtain confirmation information; The processing module is further configured to update the current model algorithm based on the second sensing data when the sensing data is the second sensing data, and obtain a processing result, the processing result including the updated model algorithm; The generation module is used to generate a second response message based on the confirmation information or processing result, and send the second response message to the SF.

8. A self-updating device of a common-sense integrated sensing algorithm, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1 to 3, 4 to 5.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 3, 4 to 5.

10. A computer program product, characterised in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1 to 3, 4 to 5.

Images