Method and apparatus for determining measurement result, method and apparatus for sending

Abstract

The application discloses a kind of determination method and device of measurement result, transmitting method and device, wherein, the method comprises: receiving the service request sent by target device;Based on service request, generate measurement configuration information;Determine measurement result based on measurement configuration information.Through the present application, the technical problems that the quality of service of target device to AI service cannot be accurately measured in the related art, leading to the inability to switch base station to handle AI service in time, and then the effect that the processing of base station to AI service can be accurately measured is achieved, when the quality of service of device is not satisfied by the processing of base station to AI service, timely trigger the effect of base station switching.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the field of communication, in particular to a method and apparatus for determining measurement results, a method and apparatus for sending. BACKGROUND

[0002] With the movement of the target device and the dynamic change of network resources (including communication, calculation, memory, etc.), the base station currently serving the target device can not be able to meet the processing requirements (e.g., real-time performance, accuracy, etc.) of the target device for the target service (e.g., artificial intelligence (AI) service). In this case, the target device needs to be switched to a new base station with higher service quality to ensure that the target device can enjoy high-quality task processing services.

[0003] The base station switching mechanism in the related art is mainly based on the measurement and triggering of wireless signal strength and other indicators, without considering the demand of the target device for the service quality of the service. This leads to that when the base station cannot accurately measure the demand of the target device for the service quality of the service, the traditional mechanism is difficult to timely trigger the switching of the base station when the processing of the service does not meet the processing requirements of the target device, thereby affecting the processing experience of the target device for the task. SUMMARY

[0004] Embodiments of the present application provide a method and apparatus for determining measurement results, a method and apparatus for sending, to at least solve the technical problem that the demand of the target device for the service quality of the AI service cannot be accurately measured in the related art, leading to the inability to timely switch the base station to process the AI service.

[0005] According to an aspect of embodiments of the present application, a method for determining measurement results is provided, applied to a base station, comprising: receiving a service request sent by a target device, wherein the service request includes processing requirements of the target device for a target service, and the target service includes an artificial intelligence (AI) service; generating measurement configuration information based on the service request, wherein the measurement configuration information is used to guide the target device to perform measurement of the service quality of the target service, the measurement configuration information includes measurement object configuration and measurement result configuration, the measurement object configuration is used to represent a network entity corresponding to the target service to be measured, and the measurement result configuration is used to represent a time and a manner of sending a measurement result by the target device; sending the measurement configuration information to the target device, and receiving a first measurement result sent by the target device, wherein the first measurement result is a result obtained by the target device based on the measurement configuration information, by measuring the service quality of the target service.

[0006] According to another aspect of the embodiments of the present application, a method for sending measurement result is provided, which is applied to a target device and includes: sending a service request to a base station, wherein the service request includes a processing requirement of the target device for a target service, and the target service includes an artificial intelligence (AI) service; receiving measurement configuration information sent by the base station, wherein the measurement configuration information is generated by the base station based on the service request, and includes measurement object configuration and measurement result configuration, the measurement object configuration is used to indicate a network entity corresponding to the target service to be measured, and the measurement result configuration is used to indicate a time and a manner for sending a measurement result by the target device; measuring a service quality of the target service based on the measurement configuration information to obtain a first measurement result; and sending the first measurement result to the base station, wherein the base station is used to generate handover decision information based on the first measurement result, and the handover decision information is used to indicate whether to hand over a connection between the target device and the base station.

[0007] According to another aspect of the embodiments of the present application, a device for determining measurement result is also provided, which includes a first memory, a first processor, and a first computer program stored in the first memory and executable on the first processor, and when the first processor executes the first computer program, the following operations are implemented: receiving a service request sent by a target device, wherein the service request includes a processing requirement of the target device for a target service, and the target service includes an artificial intelligence (AI) service; generating measurement configuration information based on the service request, wherein the measurement configuration information is used to guide the target device to perform measurement on a service quality of the target service, and includes measurement object configuration and measurement result configuration, the measurement object configuration is used to indicate a network entity corresponding to the target service to be measured, and the measurement result configuration is used to indicate a time and a manner for sending a measurement result by the target device; sending the measurement configuration information to the target device, and receiving a first measurement result sent by the target device, wherein the first measurement result is obtained by the target device based on the measurement configuration information.

[0008] According to another aspect of the embodiments of the present application, a sending device of measurement result is further provided, which comprises a second memory, a second processor, and a second computer program stored in the second memory and executable in the second processor, and when the second processor executes the second computer program, the following operations are implemented: sending a service request to a base station, wherein the service request comprises a processing requirement of a target device to a target service, and the target service comprises an artificial intelligence (AI) service; receiving measurement configuration information sent by the base station, wherein the measurement configuration information is information generated by the base station based on the service request, and the measurement configuration information comprises measurement object configuration and measurement result configuration, the measurement object configuration is used to indicate a network entity corresponding to the target service to be measured, and the measurement result configuration is used to indicate a time and a manner of sending a measurement result by the target device; measuring a service quality of the target service based on the measurement configuration information to obtain a first measurement result; and sending the first measurement result to the base station, wherein the base station is used to generate handover decision information based on the first measurement result, and the handover decision information is used to indicate whether to hand over a connection between the target device and the base station.

[0009] According to still another aspect of the embodiments of the present application, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program, and the computer program is configured to be executed by a processor to perform the steps in any one of the method embodiments.

[0010] According to still another aspect of the embodiments of the present application, a computer program product or a computer program is provided, and the computer program product or the computer program comprises computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to enable the computer device to perform the steps in any one of the method embodiments.

[0011] According to still another aspect of the embodiments of the present application, an electronic device is further provided, which comprises a memory and a processor, the memory stores a computer program, and the processor is configured to execute the computer program to perform the steps in any one of the method embodiments.

[0012] According to the present application, since the base station receives the service request sent by the target device, the base station can immediately understand the specific processing needs of the target device for the AI service, and accordingly generate the corresponding measurement object configuration and measurement result configuration, which enables the network to be dynamically adjusted to better adapt to the changes in AI service demand; the measurement configuration information is sent to the target device, and the target device measures the quality of service of the AI service based on the measurement configuration information, which provides accurate measurement instructions for the target device and can ensure the accuracy and pertinence of the measurement results; so that the measurement results can be more accurately determined, and whether to switch the connection between the target device and the base station is determined in time, ensuring the continuity and quality of service of the AI service and promoting the efficient use of network resources. Therefore, the technical problem that the quality of service of the AI service of the target device cannot be accurately measured in the related art, resulting in the inability to timely switch the base station to process the AI service, can be solved, and the effect that the processing of the AI service of the base station can be accurately measured, and the base station is timely triggered when the processing of the AI service of the base station does not meet the quality of service requirements of the device. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 is an application scenario of a measurement result determination method according to an embodiment of the present application;

[0014] Figure 2 is a flowchart of an optional measurement result determination method according to an embodiment of the present application;

[0015] Figure 3 is a flowchart of another optional measurement result sending method according to an embodiment of the present application;

[0016] Figure 4 is another optional interaction flowchart between a UE and a base station according to an embodiment of the present application;

[0017] Figure 5 is a structural block diagram of an optional measurement result determination apparatus according to an embodiment of the present application;

[0018] Figure 6 is a structural block diagram of another optional measurement result sending apparatus according to an embodiment of the present application;

[0019] Figure 7 is a structural block diagram of a computer system of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0020] In the following, the technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the drawings in the embodiments of the present application, so that those skilled in the art can better understand the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work should be within the scope of protection of the present application.

[0021] It should be noted that the terms "first", "second" and the like in the description and claims of the present application and the above drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in other than the order illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a list of steps or units need not be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to such processes, methods, products or devices.

[0022] According to an aspect of the embodiments of the present application, a determination method of a measurement result is provided. Optionally, in the present embodiment, the determination method of the measurement result can be applied to, but is not limited to, a hardware environment as shown in the figure comprising a terminal device 102 and a server 104. The server 104 can be connected with the terminal device 102 through a network, and can be used to provide services (for example, application services, etc.) for the terminal device 102 or a client installed on the terminal device 102, and a database can be set on the server 104 or independently of the server 104, and used to provide data storage services for the server 104. Figure 1

[0023] The network can include, but is not limited to, at least one of the following: wired network, wireless network. The wired network can include, but is not limited to, at least one of the following: wide area network, metropolitan area network, local area network. The wireless network can include, but is not limited to, at least one of the following: Wireless Fidelity (WIFI), Bluetooth. The terminal device 102 can be, but is not limited to, a personal computer (PC), a mobile phone, a tablet computer, etc. The server 104 can be, but is not limited to, a cloud server, a server cluster or other server types.

[0024] ​The determination method of the measurement result in the embodiment of the application can be executed by the server 104, or by the terminal device 102, or by the server 104 and the terminal device 102 jointly. The terminal device 102 executing the handover method of the base station in the embodiment of the application can also be executed by a client installed thereon.

[0025] Taking the determination method of the measurement result in the embodiment executed by the server 104 arranged in the base station as an example, Figure 2 is a flowchart of an optional determination method of a measurement result according to the embodiment of the application, as Figure 2 shown, the flow of the method can include the following steps:

[0026] Step S202, receiving a service request sent by a target device, wherein the service request includes processing requirements of the target device for a target service, and the target service includes an artificial intelligence (AI) service;

[0027] Step S204, generating measurement configuration information based on the service request, wherein the measurement configuration information is used to guide the target device to perform measurement of the quality of service of the target service, and the measurement configuration information includes measurement object configuration and measurement result configuration, the measurement object configuration is used to represent a network entity corresponding to the target service to be measured, and the measurement result configuration is used to represent a time and a manner of sending a measurement result by the target device;

[0028] Step S206, sending the measurement configuration information to the target device, and receiving a first measurement result sent by the target device, wherein the first measurement result is a result obtained by the target device based on the measurement configuration information, and measuring the quality of service of the target service.

[0029] The determination method of the measurement result in the embodiment can be applied to the field of communication between the base station and the target device, and can be applied to a scene relying on real-time and high-precision AI services. For example, in an autonomous vehicle, the vehicle (target device) can need to exchange data with the base station in real time to process AI services such as target detection, path planning, and environment perception. The handover method can ensure that when the quality of service (QoS) of the AI service is reduced due to changes in the computing power of the base station or the communication conditions, the vehicle can quickly switch to a base station that can provide better AI services, maintaining the stability and safety of the autonomous driving system. For another example, in remote medical services, AI technology is used for image recognition, disease diagnosis, and health monitoring, which has very high requirements for response delay and accuracy. The handover method can immediately switch to a better base station when the QoS of the AI service is damaged, reducing potential risks in the remote medical process and improving the accuracy and timeliness of medical diagnosis.

[0030] Optionally, the target device in this embodiment refers to a device that needs to communicate with the base station (such as gNB) and relies on the base station to provide certain services, for example, user equipment (User Equipment, referred to as UE), specifically including but not limited to: smart phones, wearable devices, autonomous vehicles, drones, smart cameras, remote medical devices.

[0031] Optionally, the service request in this embodiment refers to a request sent by the target device to the base station, which contains the target device's processing requirements for a specific service (such as AI model inference, training tasks, etc.), such as the required maximum response time, minimum accuracy, supported model size, and other QoS parameters.

[0032] Optionally, the processing requirement in this embodiment refers to the specific performance requirements of the target device for the target service (AI service), such as response time, service accuracy, model size limit, etc. These requirements are the minimum standards for the quality of AI services that the target device expects to obtain from the base station. For example, the processing requirement includes at least one of the following: processing latency requirement of the target service, processing accuracy requirement of the target service, computing resource requirement of the base station for processing the target service, computing capability requirement of a first network model for processing the target service, processing latency requirement of the first network model for processing the target service, computing resource requirement of the first network model for processing the target service, computing resource requirement of a second network model for processing the target service, processing latency requirement of the second network model for processing the target service, memory requirement of the target device, computing capability requirement of the target device, memory requirement of the first network model, memory requirement of the second network model, the first network model is set in the base station, and the second network model is set in the target device. Among them,

[0033] The processing latency requirement refers to the time delay in the target service processing process, that is, the time interval from sending the service request by the device to receiving the processing result. In real-time applications such as autonomous driving, remote surgery, online education interaction, etc., low latency is crucial because any delay can seriously affect user experience or safety.

[0034] The processing accuracy requirement is the target device's requirement for the accuracy of the service processing result. For AI services such as image recognition, speech-to-text, data analysis, the target device may need to achieve a certain accuracy threshold, otherwise the service will be considered unacceptable.

[0035] The computing resource requirement of the base station is to carry out the execution of the AI model, and the base station needs sufficient computing resources. This may include the utilization rate of hardware resources such as CPU, GPU, and special accelerators, as well as specific AI service processing capabilities such as the maximum number of concurrent tasks.

[0036] The first network model is usually deployed in the base station to support AI inference, training, etc. The computing power requirement involves the basic computing power required for model execution, including FLOPS (floating point operations per second) and other indicators.

[0037] The processing latency requirement of the first network model is the constraint on the latency of the first network model executing a specific AI service, ensuring that the calculation is completed within a limited time and does not affect the response speed of the entire AI service chain.

[0038] The computing resource requirement of the first network model is the specific requirement of the base station computing resource when the first network model is executed, including the occupied memory, CPU / GPU core number, etc.

[0039] The second network model is usually resident in the UE (user equipment), such as a lightweight AI model on the edge device. Its computing resource requirement reflects the local computing capability required by the UE to complete the AI service.

[0040] The processing latency requirement of the second network model is the maximum acceptable latency when the UE executes local AI services, ensuring smooth AI computing process in coordination with the base station.

[0041] The UE needs a certain amount of memory space to run specific AI applications or models. The memory requirement specifies the minimum memory capacity required by the UE to execute AI services.

[0042] In addition to memory, the computing capability of the UE is also a key factor in evaluating its ability to effectively execute AI services. This includes the performance of CPU, GPU, etc. processors, and whether it has a dedicated AI computing unit.

[0043] The first network model and the second network model are run on the base station and the UE respectively. The first network model and the second network model can be dynamically adjusted and optimized according to different service requirements and the capabilities of the UE and the base station.

[0044] Combining the above requirements, the target device will explicitly indicate its expectations for the quality of service of the target service in the service request, such as requiring a response latency of less than 100ms, an accuracy of at least 95%, or requesting a specific type of AI model (such as YOLOv7 for object detection). The base station generates measurement configurations according to these requirements to monitor the performance of its AI services, and when the performance drops to the point where it cannot meet the UE's requirements, it triggers a switching process to connect the target device to a base station that can provide higher QoS, ensuring the continuity and high quality of AI services.

[0045] Optionally, the measurement configuration information in this embodiment is a set of parameters and instructions issued by the base station (e.g. gNB) to ensure that the target device (UE) can accurately and effectively measure the quality of service of the target service. These configuration information guides the UE to perform the necessary measurements to evaluate the AI QoS performance of the current serving base station and the neighboring base station, and decides whether to perform base station switching according to the results. The measurement configuration information includes but is not limited to specific indicators that need to be measured. The measurement object configuration specifies which network entities the UE needs to perform performance measurement on. This can be the current target service base station (or cell), the neighboring base station (or cell), or even the base station of different systems (such as from 5G to 4G). Through the measurement object configuration, the UE can determine the range and type of base stations that need to be focused on in order to more accurately evaluate the quality of service of the target service. The measurement result configuration describes the rules of how and when the UE reports the measurement results to the base station. This includes the reporting period (such as reporting once every 120ms), event-triggered reporting (such as reporting when the response time exceeds 100ms), and the content format of the report. The report configuration ensures the effectiveness and timeliness of the communication between the UE and the base station, so that the base station can make switching decisions quickly when necessary.

[0046] The measurement configuration information further includes at least one of: a measurement indicator configuration for indicating the performance indicators of the target service quality to be measured; and a measurement identifier configuration for identifying the measurement configuration information. The measurement indicator configuration is used to indicate the specific performance indicators that the UE needs to measure. In an AI QoS-oriented network, these indicators may include but are not limited to the response time, accuracy, computing resource usage, maximum AI model size supported by the UE, and traditional network performance indicators such as signal strength (RSRP, RSRQ) of the AI service. For example, the base station may require the UE to measure the end-to-end AI service response time (E2E_Response_Latency) and the AI service accuracy (Accuracy). The measurement identifier configuration is used to uniquely identify a measurement configuration information, which facilitates the tracking and management of different measurement configurations issued at different times between the UE and the base station. The measurement identifier can be a simple number or string, which ensures that the base station can identify when the UE reports the results corresponding to which configuration, thereby avoiding confusion and incorrect decisions.

[0047] These components of the measurement configuration information work together to provide the UE with a detailed measurement guidance scheme, ensuring the accuracy and timeliness of the measurement data, and thus helping the base station and network management system make optimization and handover decisions based on AI QoS. For example, the UE can be configured to report measurement results every 120 ms, while also reporting an event when it finds that the AI service response time of a neighboring base station is 10 ms lower than that of the current serving base station. The base station evaluates network quality based on this information, and when it detects a decrease in AI service quality or a neighboring base station that can provide better service, it can quickly perform a handover to maintain the UE's AI service experience.

[0048] Optionally, the target device in this embodiment collects data on the AI service quality of the serving base station and neighboring base stations, as well as the traditional wireless signal quality, by performing the monitoring task of the measurement configuration indication. This data is packaged in a message and reported to the current serving base station. The measurement results can include: end-to-end response time of the AI service: check if it exceeds the preset threshold. AI service accuracy: compare with the minimum accuracy requirement of the UE. Resource availability: whether it can meet the UE's demand for AI services. Wireless signal quality indicators (such as RSRP, RSRQ, SINR, etc.): evaluate the stability of the communication link.

[0049] Optionally, the first measurement result in this embodiment is generated by the target device after measuring the AI service performance of the current serving base station according to the measurement configuration information issued by the base station, and can include key indicators such as response time, accuracy, and computing resource availability, as well as performance measurement data of neighboring base stations.

[0050] Through the embodiments provided in this application, since the base station receives the service request sent by the target device, the base station can immediately understand the specific processing needs of the target device for AI services, and accordingly generate corresponding measurement object configuration and measurement result configuration, which enables the network to dynamically adjust to better adapt to changes in AI service demand; the measurement configuration information is sent to the target device, and the target device measures the service quality of the AI service based on the measurement configuration information, which provides the target device with accurate measurement instructions and can ensure the accuracy and relevance of the measurement results; thereby the measurement results can be more accurately determined, and whether to switch the connection between the target device and the base station can be determined in a timely manner, ensuring the continuity and service quality of the AI service and promoting the efficient use of network resources. Therefore, the technical problem that the service quality demand of the target device for the AI service cannot be accurately measured in related technologies, resulting in the inability to timely switch the base station to process the AI service, can be solved, and the effect that the processing of the AI service by the base station can be accurately measured, and the base station can be timely triggered to switch when the processing of the AI service by the base station does not meet the service quality demand of the device.

[0051] In an example embodiment, after the measurement configuration information is generated based on the service request, the method further comprises: measuring the quality of service of the target service of the base station by using the measurement configuration information to obtain a second measurement result, wherein the measurement of the quality of service of the target service of the base station comprises at least one of the following: measuring the processing delay of processing the target service, measuring the processing accuracy of processing the target service, and measuring the computing resources used for processing the target service.

[0052] Optionally, after the measurement configuration information is sent to the target device and the first measurement result sent by the target device is received, the method further comprises: generating the handover decision information based on the first measurement result and / or the second measurement result.

[0053] Optionally, the generation of the handover decision information based on the measurement result comprises: generating the handover decision information based on the first measurement result and / or the second measurement result.

[0054] Optionally, the generation of the handover decision information based on the first measurement result and / or the second measurement result comprises: generating the handover decision information in a case where it is determined from the first measurement result and / or the second measurement result that the processing of the target service by the base station does not meet the processing requirement.

[0055] Optionally, the handover of the base station in the embodiment refers to the process of switching from a current serving base station to a new serving base station, aiming to meet the AI service quality requirement of the UE and improve the user experience. The handover decision is based on the first measurement result reported by the UE. If the AI service quality of the current base station does not meet the requirement of the UE, the handover mechanism will identify a neighboring base station with better service quality and perform the corresponding connection switching.

[0056] For example, in a specific embodiment, a vehicle (UE) equipped with an automatic driving system is using a real-time target detection AI service provided by a base station (gNB). The vehicle requires that the end-to-end response delay of the AI service be ≤100ms, the accuracy be ≥90%, and the AI model size received by the vehicle side be ≤200M. When the vehicle moves, the computing load of the current serving base station increases and the wireless signal quality between the base station and the vehicle decreases, resulting in an end-to-end response delay of the AI service greater than 100ms, which cannot meet the AI service quality requirement of the vehicle. The specific steps include:

[0057] S1, the vehicle sends a service request to the current serving base station, which includes the processing requirement of the real-time target detection AI service, such as response delay, accuracy, and model size limit.

[0058] S2, the base station generates measurement configuration information based on the vehicle's service request, including the measurement of response delay, accuracy, computing resource availability, and other indicators of AI services, as well as the wireless signal quality measurement configuration of neighboring base stations.

[0059] S3, the base station sends the generated measurement configuration information to the vehicle, and the vehicle starts measuring the current service base station according to the configuration information. When the vehicle detects that the response delay of the current base station exceeds 100ms, it will report the first measurement result to the base station.

[0060] S4, after receiving the first measurement result from the vehicle, the base station finds that the current base station cannot meet the vehicle's AI service quality requirements, searches for neighboring base station information, finds a base station that can provide better AI services, and then initiates a handover process to switch the vehicle from the current base station to a new base station, ensuring that the vehicle can continue to enjoy high-quality AI services.

[0061] Optionally, the second measurement result in this embodiment is direct measurement data of the base station based on the measurement configuration information to measure the performance of processing specific services. These data cover key performance indicators such as processing delay, processing accuracy, and computing resources used, which are used to evaluate whether the base station can meet the requirements of the target service.

[0062] Optionally, the processing delay measurement in this embodiment is to measure the total time required to process target services (such as AI model inference, data transmission, etc.), which is an important indicator to judge the response speed and efficiency of the base station. For example, processing delay may include AI model calculation time, data transmission time, etc.

[0063] Optionally, the processing accuracy measurement in this embodiment is to measure the accuracy of the target service processing result according to the service type, usually expressed in percentage form. For image recognition, natural language processing, and other AI services, processing accuracy is directly related to the reliability and user experience of the service.

[0064] Optionally, the computing resource measurement in this embodiment is to monitor the utilization rate of computing resources (such as CPU, GPU, TPU) used for processing target services in the base station, to understand the remaining space of current computing capacity, which is crucial for resource scheduling and optimization.

[0065] For example, in a specific embodiment, assume that in a smart city environment, a smart drone (target device) is performing a high-precision real-time video analysis task that requires low-latency and high-precision AI services. The specific steps include:

[0066] S1, the intelligent UAV sends a service request to the network, including the latency and accuracy requirements for AI services. The network generates measurement configuration information based on the request and sends it to the UAV through second signaling. The configuration information includes measurement requirements for processing latency, processing accuracy, and computing resource usage.

[0067] S2, the network (such as the current serving base station) starts to monitor its processing capacity for the target UAV service according to the measurement configuration information. For example, it measures the AI model inference latency as 120 ms, the model processing accuracy as 88%, and the current CPU and GPU utilization rates as 45% and 65%, respectively.

[0068] S3, the above measurement data constitutes the second measurement result. The network analyzes these data and finds that the processing latency and accuracy do not meet the UAV's service request standards (assuming that the requirement is that the latency does not exceed 100 ms and the accuracy is not less than 90%).

[0069] S4, based on the analysis of the second measurement result, the network decides to switch the UAV from the current base station to another base station with more optimal performance indicators (such as lower processing latency and higher processing accuracy) to meet its AI service requirements.

[0070] S5, the network notifies the UAV to switch base stations through a signaling mechanism and updates its measurement configuration to ensure that the new base station can provide AI service quality that meets the requirements.

[0071] This embodiment enables the base station to dynamically adjust according to the target service requirements by measuring the base station's processing of the target service, ensuring that the service quality always meets the service requirements.

[0072] Optionally, this embodiment can switch the base station based on the first measurement result, or based on the second measurement result, or in combination with the first measurement result and the second measurement result.

[0073] For example, in a specific embodiment, assume that in a smart agriculture scenario, a UAV (target device) for crop health monitoring needs to analyze images in real time and return the analysis results, which involves complex AI image recognition algorithms (target service). The specific steps include:

[0074] S1, the UAV sends a service request to the network, including the processing requirements of the AI service, such as image recognition end-to-end response latency ≤ 100 ms and minimum model inference accuracy ≥ 95%.

[0075] S2, the serving base station starts to monitor its performance in processing AI services internally and finds that when the UAV is located at a certain position, the AI image recognition execution latency reaches 120 ms and the model inference accuracy is only 90%, which is significantly lower than the processing requirements.

[0076] S3, at the same time, the UAV is also performing measurement, it detects that the response of the serving base station to the image recognition task has exceeded the allowed range (such as the end-to-end response time delay is 115 ms), and the response time delay of a base station (base station B) in the vicinity is 80 ms, and the model inference accuracy reaches 98%, which obviously meets and even exceeds the processing requirements.

[0077] S4, the network performs comprehensive evaluation based on the second measurement result (the self-monitoring result of the serving base station) and the first measurement result (the measurement result reported by the UAV), and finds that the processing capability of the serving base station for the AI service indeed fails to meet the processing requirements.

[0078] S5, based on the above evaluation, the network decides to switch the UAV from the serving base station to base station B, because base station B exhibits better AI service quality, and both the response time delay and the model inference accuracy meet the service requirements of the UAV.

[0079] The base station switching through multiple measurement results in the embodiment can timely switch the resource-intensive task to the base station with more abundant resources, realize effective allocation of resources, and avoid overloading of resources of a single base station to affect the overall service processing quality.

[0080] In an example embodiment, in step S202, the service request sent by the target device is received, including: receiving first signaling sent by the target device; and parsing the service request from a first extension field in the first signaling.

[0081] Optionally, the first signaling in the embodiment refers to the signaling in which the target device first interacts with the base station to request AI service or update its service requirements. The first signaling includes but is not limited to RRC Setup Request signaling, RRC Setup Complete signaling, dedicated RRC signaling, NAS (Non-Access Stratum) signaling, and SIB (System Information Block) type signaling. For example, the UE sends the QoS request of its AI service to the base station. The UE can send the QoS request of its AI service to the network, the computing / communication capability of the UE, etc. through UE assistance information (UE Assistance Information), dedicated RRC signaling (Reconfiguration Complete), or in the access process.

[0082] Optionally, the first extension field in the embodiment refers to an extension field added on the basis of the original signaling format to adapt to emerging technologies and specific requirements. These fields can carry additional information such as detailed requirements of AI service and device capability, so that the network can better understand and respond to the requirements of the device.

[0083] For example, in one specific embodiment, assume that an autonomous vehicle (target device) is driving on a highway, and it relies on the real-time target detection AI service provided by nearby base stations to ensure safe driving. The following are the specific implementation steps:

[0084] S1, the autonomous vehicle sends a service request to the current serving base station, which is done by sending an RRC Setup Complete (or similar) signaling. This signaling contains specific requirements for AI services, such as requiring the response time of AI services to be no more than 100ms, the accuracy to be no less than 90%, and the AI model to support a maximum of 200M parameters.

[0085] S2, in the RRC Setup Complete signaling, a first extension field is added to specifically describe the AI service requirements. This field may contain one or more subfields, each representing a specific requirement, such as response time, accuracy requirement, model size, etc.

[0086] S3, after the base station receives the signaling, it first needs to decode the original data in the signaling. Then, the base station will find and parse the first extension field in the signaling to extract all important information about the AI service request. These information will be used to generate the corresponding AI QoS measurement configuration to ensure that the subsequent performance monitoring and necessary base station switching can be targeted at the specific needs of the autonomous vehicle.

[0087] S4, the base station generates and sends measurement configuration information to the autonomous vehicle according to the parsed service request, guiding it to measure the AI QoS of the current serving base station and surrounding base stations.

[0088] The autonomous vehicle measures and reports relevant performance indicators according to the received measurement configuration information, either periodically or under certain circumstances. If the monitoring results show that the AI service quality has decreased and cannot meet its business needs, the base station can start the switching process to migrate the connection to another base station that can provide better AI services, to ensure the safe driving and service experience of the autonomous vehicle.

[0089] This embodiment can accurately understand the target device's needs by parsing specific business requirements through the first signaling and its extension field, avoiding waste of resources.

[0090] In an example embodiment, in step S204, the measurement configuration information is generated based on the service request, including: determining the resource of the base station and the capability information of the target device in response to the service request, wherein the resource of the base station and the capability information of the target device both correspond to the target service, and the resource of the base station includes at least one of the following: computing resource, model resource, communication resource, and data resource; and generating the measurement configuration information based on the resource of the base station, the capability information of the target device, and the processing requirement.

[0091] Optionally, the evaluation of the base station resource in the embodiment includes but is not limited to: computing resource: refers to the available computing power and current utilization rate of hardware resources such as central processing unit (CPU), graphics processing unit (GPU), and tensor processing unit (TPU) in the base station for executing AI services, as well as the maximum number of concurrent AI services that can be supported; model resource: refers to the AI model set stored in the base station, including the types, versions, and corresponding processing capabilities of the models, as well as the storage status and calling efficiency of the models; communication resource: refers to the data transmission capability between the base station and the UE, including bandwidth, latency, and communication quality, to ensure efficient transmission of AI service requests and responses; and data resource: refers to the size and quality of the data set accessible by the base station when processing AI services, as well as the data preprocessing and management capabilities.

[0092] Optionally, the capability information of the target device in the embodiment mainly includes: computing capability: the processor performance of the UE, including the model and computing power of CPU and GPU, as well as the supported AI computing types and performance levels; memory and storage: the available memory size and storage space of the UE, which determines the AI model size that the UE can support and the feasibility of its operation; battery life: especially for mobile devices, the battery life affects the ability of the UE to continuously perform AI services and communicate with the network; AI model compatibility: the types and versions of AI models that the UE can run, as well as the related software environment and framework support.

[0093] Optionally, the network side (such as gNB) generates specific measurement configuration information based on the comprehensive consideration of the above-mentioned resource and capability information and processing requirement. This configuration aims to set reasonable monitoring standards and triggering conditions to ensure the quality of AI services.

[0094] Optionally, the step of generating the measurement configuration includes:

[0095] Analyzing the processing requirement: determining the requirements of the AI service on response time, precision, and resource consumption, etc.

[0096] Matching resources and capabilities: Check if the base station and UE each meet the processing needs, such as whether the computing resources are sufficient, whether the communication latency is below the threshold, etc.

[0097] Setting measurement indicators: According to the analysis results, select the key indicators that need to be monitored, such as end-to-end response latency, model inference accuracy, base station computing resource utilization, etc.

[0098] Defining trigger events: Set event trigger conditions, such as triggering handover measurement when base station computing resource utilization exceeds 70%, or starting neighbor search when UE end-to-end response latency exceeds 100ms.

[0099] Configuring measurement frequency and duration: Specify the period and duration of the measurement to balance the real-time monitoring and network burden.

[0100] Signaling interaction: Send the measurement configuration to the UE through RRC (Radio Resource Control) signaling to ensure that the UE can perform the corresponding monitoring tasks according to the configuration.

[0101] For example, in one specific embodiment, assume a scenario of remote medical diagnosis, where a doctor (through a target device, such as a mobile medical workstation with AI image analysis capabilities) needs to obtain the AI analysis results of high-definition pathological images on the patient's site immediately to make preliminary disease diagnosis. The specific implementation steps include:

[0102] S1, the target device sends the first signaling (such as RRC Setup Complete) to the serving base station, which contains the first extension field, explicitly indicating that the service requirement is remote medical image analysis, requiring AI service response latency ≤100ms, accuracy ≥95%, and specifying the use of a specific type of AI model (such as a pathological image recognition model).

[0103] S2, after receiving the service request, the base station first parses the first extension field in the first signaling to understand the specific requirements. Next, the base station evaluates its computing resources (availability of CPU, GPU, etc.) and inquires the device parameters of the target device (through RRC signaling or SIB broadcast) to understand the device's computing power, memory capacity, network interface speed, etc.

[0104] S3, according to the service request, computing resource availability and device parameters, the base station generates a series of measurement configuration information. For example, the base station will define the measurement indicator configuration of AI service quality indicators (AI service response latency, accuracy), and set the measurement object configuration (current base station, neighboring base station), while configuring the time interval and trigger event of measurement result reporting (such as immediately reporting when the response latency exceeds 100ms).

[0105] S4, the base station sends the measurement configuration information to the target device through the RRC Connection Reconfiguration message, guiding the device on how to perform measurements and report results.

[0106] S5, the target device regularly measures the AI service quality (such as response time, accuracy) and other wireless signal indicators (such as RSRP, RSRQ) of the serving base station according to the received measurement configuration information, and sends the first measurement results to the base station when the reporting configuration time interval or trigger condition is reached.

[0107] S6, after receiving the first measurement results, the base station analyzes whether the current AI QoS meets the key parameters in the service request. If it finds that the AI service quality has decreased (such as the response time exceeding 100ms), the base station will evaluate the performance of neighboring base stations, looking for better computing resource availability and AI service capabilities. Based on the comprehensive evaluation, the base station may decide to trigger base station switching to ensure that the target device can continuously obtain high-quality AI services.

[0108] This embodiment can achieve fine management of resources by determining the computing resources of the base station and the device parameters of the target device. This means that computing resources can be intelligently allocated according to the specific needs and capabilities of each device, ensuring efficient use of resources and avoiding waste.

[0109] In an example embodiment, in step S206, the measurement configuration information is sent to the target device, and the first measurement result sent by the target device is received, including: sending second signaling to the target device, wherein the measurement configuration information is included in the second extension field in the second signaling; receiving third signaling sent by the target device; and parsing the first measurement result from the third extension field in the third signaling.

[0110] Optionally, the types of the second signaling, the third signaling and the first signaling in the above embodiment are similar, and will not be described here.

[0111] Optionally, the second extension field in the embodiment is an additional field in the second signaling that is specifically used to describe and carry the measurement configuration information, which extends the function of the traditional signaling to support complex target service quality monitoring requirements. For example, this field may contain detailed configurations on how to measure AI service performance, which indicators to use, and when to report measurement results.

[0112] Optionally, the third extension field in this embodiment is an additional field in the third signaling, which contains the first measurement result. For example, the measurement data of the UE on the current serving cell AI service quality, such as response time, computing resource availability, accuracy, etc., and the measurement result on the neighboring cell, for evaluating potential handover targets.

[0113] For example, in a specific embodiment, assuming that in an intelligent traffic management system, an autonomous vehicle as a target device needs to perform real-time road environment perception and prediction, which involves a large amount of AI computing. The vehicle requests AI service from the network through the first signaling (such as RRC Setup Complete), especially requiring low response time and high accuracy performance. The specific implementation steps include:

[0114] S1, the base station (gNB) receives and parses the service request of the vehicle, determines the computing resource (such as CPU / GPU load) of the base station and the device parameters (such as maximum AI model carrying capacity) of the vehicle. Based on these information, the gNB generates the measurement configuration information for AI QoS.

[0115] S2, the gNB sends the measurement configuration information to the vehicle through the second signaling (such as RRC Reconfiguration message), which is encoded in the second extension field. The configuration may specify the AI QoS indicators that need to be measured periodically, the information of the neighboring base stations and the conditions for triggering the report (for example, the response time exceeds the preset threshold or the accuracy drops to a certain level).

[0116] S3, the autonomous vehicle starts to monitor the AI service quality of the current base station according to the measurement configuration information, and collects relevant information of the neighboring base stations. Once the AI service quality is monitored to be degraded or the reporting condition is reached, the vehicle sends the first measurement result back to the base station through the third signaling (such as Measurement Report message). The detailed measurement data, such as end-to-end response time, AI service accuracy, computing resource availability of the neighboring base stations, etc., are contained in the third extension field.

[0117] This embodiment ensures the accuracy of the measurement by including the second extension field in the second signaling, and the base station explicitly indicates to the target device which specific service quality measurement needs to be performed, avoiding the waste of resources caused by irrelevant measurement. The target device performs the measurement and stores the result in the third extension field, and reports it to the base station through the third signaling, ensuring that the base station can obtain the service quality data of the location where the target device is located in real time, and respond in time to meet the business needs of the target device.

[0118] In an example embodiment, switching the connection between the target device and the base station based on the first measurement result comprises: in a case where it is determined from the first measurement result that the processing of the target service by the base station does not meet the processing requirement, switching the target device to another base station, wherein the another base station is a base station whose processing of the target service meets the processing requirement.

[0119] Optionally, the first measurement result in the embodiment includes measurement data of the target device on the AI service quality of the current serving base station (hereinafter referred to as the original base station) and the neighboring base station. This includes but is not limited to response delay, model accuracy, and computing resource utilization. These data are used to evaluate the ability of the original base station in processing the specific AI service, whether it can meet the requirements of the target service on response time and accuracy, etc.

[0120] Optionally, the base station switching in the embodiment is also called cell switching, which is used to ensure seamless service connection of the target device in mobile communication. For example, in the AI QoS-oriented scenario, the switching decision also considers the measurement result of the AI service quality to ensure that the UE can continuously obtain the service that meets its AI service requirements.

[0121] For example, in a specific embodiment, it is assumed that in a smart city scenario, a drone as a target device is performing a high-definition video analysis task, and its AI service requirements include low-delay video streaming and high-precision target recognition. During the movement of the drone, through the execution of the measurement, it is found that the AI service response delay of the original base station is too long, the accuracy is reduced, and it fails to meet the processing requirement. The specific implementation steps include:

[0122] S1. The drone periodically performs measurement and reports its measurement result to the network through the third signaling (such as Measurement Report) when the triggering condition (such as the response delay exceeding the preset threshold) is met. The measurement result includes the decline of the AI service quality of the original base station and the potential service capability of the neighboring base station.

[0123] S2. The network receives and analyzes the first measurement result to determine whether the original base station meets the processing requirement in processing the high-definition video analysis task of the drone. If the analysis result shows that the current AI service quality is lower than the required threshold, the network will start the switching process.

[0124] S3. The network selects a neighboring base station whose AI service quality (such as response time, model accuracy) is sufficient to meet the service requirements of the drone. The network then guides the drone to switch from the original base station to the new neighboring base station to ensure the continuity and quality of the service. The switching process may include steps such as re-negotiating the wireless resource configuration, updating the AI QoS parameters, etc.

[0125] The base station switching mechanism based on the first measurement result in the embodiment can actively monitor and respond to fluctuations in the quality of service of the base station, ensuring that the target device can continuously obtain high-quality task services during movement.

[0126] In another embodiment, the sending method of the measurement result in the embodiment is executed by the target device, Figure 3 is a flowchart of an optional measurement result sending method according to an embodiment of the application, as Figure 3 shown, the flow of the method can include the following steps:

[0127] Step S302, sending a service request to a base station, wherein the service request includes the processing requirements of a target device for a target service, and the target service includes an artificial intelligence (AI) service;

[0128] Step S304, receiving measurement configuration information sent by the base station, wherein the measurement configuration information is information generated by the base station based on the service request, and the measurement configuration information includes measurement object configuration and measurement result configuration, the measurement object configuration is used to indicate a network entity corresponding to the target service to be measured, and the measurement result configuration is used to indicate a time and a manner of sending a measurement result by the target device;

[0129] Step S306, measuring the quality of service of the target service based on the measurement configuration information to obtain a first measurement result;

[0130] Step S308, sending the first measurement result to the base station, wherein the base station is used to generate switching decision information based on the first measurement result, and the switching decision information is used to indicate whether to switch the connection between the target device and the base station.

[0131] The measurement result sending method in the embodiment can be applied to the field of communication between the base station and the target device, and can be applied to a scenario that relies on real-time and high-precision AI services. For example, in an autonomous vehicle, the vehicle (target device) may need to exchange data with the base station in real time to process AI services such as target detection, path planning, and environment perception. The switching method can ensure that when the quality of service (QoS) of the AI service is reduced due to changes in the computing power or communication conditions of the base station, the vehicle can quickly switch to a base station that can provide better AI services, maintaining the stability and safety of the autonomous driving system. For another example, in remote medical services, AI technology is used for image recognition, disease diagnosis, and health monitoring, which has very high requirements for response delay and accuracy. The switching method can immediately switch to a better base station when the AI service QoS is damaged, reducing potential risks in the remote medical process and improving the accuracy and timeliness of medical diagnosis.

[0132] Optionally, the target device in this embodiment refers to a device that needs to communicate with the base station (such as gNB) and relies on the base station to provide certain services, for example, user equipment (User Equipment, referred to as UE), specifically including but not limited to: smart phones, wearable devices, autonomous vehicles, drones, smart cameras, remote medical devices.

[0133] Optionally, the service request in this embodiment refers to a request sent by the target device to the base station, which contains the target device's processing requirements for a specific service (such as AI model inference, training tasks, etc.), such as the required maximum response time, minimum accuracy, supported model size, and other QoS parameters.

[0134] Optionally, the processing requirement in this embodiment refers to the specific performance requirements of the target device for the target service (AI service), such as response time, service accuracy, model size limit, etc. These requirements are the minimum standards of the AI service quality that the target device expects to obtain from the base station. For example, the processing requirement includes at least one of the following: processing latency requirement of the target service, processing accuracy requirement of the target service, computing resource requirement of the base station for processing the target service, computing capability requirement of a first network model for processing the target service, processing latency requirement of the first network model for processing the target service, computing resource requirement of the first network model for processing the target service, computing resource requirement of a second network model for processing the target service, processing latency requirement of the second network model for processing the target service, memory requirement of the target device, computing capability requirement of the target device, memory requirement of the first network model, memory requirement of the second network model, the first network model is set in the base station, and the second network model is set in the target device. Among them,

[0135] The processing latency requirement refers to the time delay in the target service processing process, that is, the time interval from sending the service request by the device to receiving the processing result. In real-time applications such as autonomous driving, remote surgery, online education interaction, etc., low latency is crucial because any delay can seriously affect user experience or safety.

[0136] The processing accuracy requirement is the requirement of the target device for the accuracy of the service processing result. For AI services such as image recognition, speech-to-text, data analysis, the target device may need to achieve a certain accuracy threshold, otherwise the service will be considered unacceptable.

[0137] The computing resource requirement of the base station is to carry out the execution of the AI model, and the base station needs sufficient computing resources. This may include the utilization rate of hardware resources such as CPU, GPU, and special accelerators, as well as specific AI service processing capabilities such as the maximum number of concurrent tasks.

[0138] The first network model is usually set in the base station to support AI inference, training, etc. The computing power requirement involves the basic computing power required for model execution, including FLOPS (floating point operations per second) and other indicators.

[0139] The processing delay requirement of the first network model is the constraint on the specific AI business delay of the first network model execution, ensuring that the calculation is completed within a limited time and does not affect the response speed of the entire AI service chain.

[0140] The computing resource requirement of the first network model is the specific requirement of the base station computing resource when the first network model executes, including the occupied memory, CPU / GPU core number, etc.

[0141] The second network model is usually resident in the UE (user equipment), such as lightweight AI models on edge devices. Its computing resource requirement reflects the local computing capability required by the UE to complete AI business.

[0142] The processing delay requirement of the second network model is the maximum acceptable delay when the UE executes local AI business, ensuring smooth AI computing process coordination with the base station.

[0143] The UE needs a certain amount of memory space to run specific AI applications or models. The memory requirement specifies the minimum memory capacity required by the UE to execute AI business.

[0144] In addition to memory, the computing capability of the UE is also a key factor in evaluating whether it can effectively execute AI business. This includes the performance of CPU, GPU, etc. processors, and whether it has a dedicated AI computing unit.

[0145] The first network model and the second network model run on the base station and the UE respectively, and their respective memory resource requirements are different, but they are both necessary conditions for the normal operation of the model. The first network model may be dynamically adjusted and optimized according to different business requirements and the capabilities of the UE to adapt to changing network conditions and the requirements of the UE. Its performance requirement refers to the processing capacity, resource consumption, scalability, etc. of the model.

[0146] Combining the above requirements, the target device will explicitly indicate its expectations for the quality of service of the target business in the business request, for example, requiring a response delay of less than 100ms, an accuracy of at least 95%, or requesting a specific type of AI model (such as YOLOv7 for object detection). The base station generates measurement configuration according to these requirements to monitor the performance of its AI service, and when the performance drops to the point where it cannot meet the UE's requirements, it triggers the switching process to connect the target device to the base station that can provide higher QoS, ensuring the continuity and high quality of AI service.

[0147] Optionally, the measurement configuration information in this embodiment is a set of parameters and instructions issued by the base station (such as gNB) to ensure that the target device (UE) can accurately and effectively measure the service processing quality. These configuration information guides the UE to make necessary measurements to evaluate the AI QoS performance of the current serving base station and neighboring base stations, and decides whether to perform base station switching according to the results. The measurement configuration information includes but is not limited to specific indicators that need to be measured. The measurement object configuration specifies which network entities the UE needs to perform performance measurement on. This can be the current processing target service base station (or cell), neighboring base station (or cell), or even inter-system base station (such as from 5G to 4G). Through the measurement object configuration, the UE can determine the range and type of base stations that need to be focused on in order to more accurately evaluate the service processing quality.

[0148] The measurement result configuration describes the rules of how and when the UE reports the measurement results to the base station. This includes the reporting period (such as reporting once every 120ms), event-triggered reporting (such as reporting when the response time exceeds 100ms), and the content format of the report. The report configuration ensures the effectiveness and timeliness of communication between the UE and the base station, enabling the base station to make switching decisions quickly when necessary.

[0149] The measurement configuration information further includes at least one of: a measurement indicator configuration for indicating performance indicators of the target service quality to be measured; a measurement object configuration for indicating network entities corresponding to the target service to be measured; a measurement result configuration for indicating the time and manner of sending the measurement results by the target device; and a measurement identifier configuration for identifying the measurement configuration information. The measurement indicator configuration is used to indicate the specific performance indicators that the UE needs to measure. In an AI QoS-oriented network, these indicators may include but are not limited to the response time, accuracy, computing resource usage, maximum AI model size supported by the UE, and traditional network performance indicators such as signal strength (RSRP, RSRQ) of the AI service. For example, the base station may require the UE to measure the end-to-end AI service response time (E2E_Response_Latency) and AI service accuracy (Accuracy). The measurement identifier configuration is used to uniquely identify a measurement configuration information, facilitating the tracking and management of different measurement configurations issued at different times between the UE and the base station. The measurement identifier may be a simple number or string, ensuring that the base station can identify when the UE reports the results corresponding to which configuration, thereby avoiding confusion and incorrect decisions.

[0150] The components of the measurement configuration information work together to provide the UE with a detailed measurement guidance scheme, ensuring the accuracy and timeliness of the measurement data, and thus helping the base station and network management system to make AI QoS-based optimization and handover decisions. For example, the UE can be configured to report measurement results every 120 ms, while also reporting an event when it finds that the AI service response time of a neighboring base station is 10 ms lower than that of the current serving base station. The base station evaluates network quality based on this information, and when it detects a decrease in AI service quality or a neighboring base station that can provide better service, it can quickly perform handover to maintain the UE's AI service experience.

[0151] Optionally, the first measurement result in this embodiment is a result generated by the target device after measuring the AI service performance of the current serving base station according to the measurement configuration information issued by the base station, which can include key indicators such as response time, accuracy, and computing resource availability, as well as performance measurement data of neighboring base stations.

[0152] Optionally, the handover of the base station in this embodiment refers to the process of switching from the current serving base station to a new serving base station, with the purpose of meeting the AI service quality requirements of the UE and improving user experience. The handover decision is based on the first measurement result reported by the UE, and if the AI service quality of the current base station does not meet the UE's requirements, the handover mechanism will identify a neighboring base station with better service quality and perform the corresponding connection handover.

[0153] Through the embodiments provided in this application, since the target device sends a service request to the base station, the base station can immediately understand the specific processing requirements of the target device for the target service, and accordingly generate corresponding measurement configurations, which enables the network to dynamically adjust to better adapt to changes in service requirements; the measurement configuration information provides precise measurement instructions for the target device, which can ensure the accuracy and relevance of the measurement results; thereby the measurement results can be more accurately determined; based on the first measurement result received from the target device, the base station can timely determine whether the processing of the target service meets the service requirements of the target device, so that the target device can be switched to a more suitable base station in a timely manner when the processing of the target service by the base station does not meet the service quality requirements of the device, ensuring service continuity and service quality, and promoting efficient use of network resources. Therefore, the technical problem in the related art that the AI service quality requirements of the target device for the AI service cannot be accurately measured, resulting in the inability to timely switch the base station to process the AI service, can be solved, and the effect that the processing of the AI service by the base station can be accurately measured, and the base station can be timely triggered to switch when the processing of the AI service by the base station does not meet the service quality requirements of the device.

[0154] In one example embodiment, sending a service request to a base station includes sending a first signaling to the base station to send a service request to the base station, wherein a first extension field in the first signaling includes the service request.

[0155] Optionally, the first signaling in this embodiment refers to the signaling when the target device first interacts with the base station, requesting AI services or updating its service requirements. The first signaling includes but is not limited to RRC Setup Request signaling, RRC Setup Complete signaling, dedicated RRC signaling, NAS (Non-Access Stratum) signaling, SIB (System Information Block) type signaling. For example, the UE sends the QoS request of its AI service to the base station. The UE can send the QoS request of its AI service to the network through UE Assistance Information, dedicated RRC signaling (Reconfiguration Complete), or in the access process, the UE's own computing / communication capabilities, etc.

[0156] Optionally, the first extension field in this embodiment refers to the extension field added on the basis of the original signaling format to adapt to emerging technologies and specific needs. These fields can carry additional information, such as detailed requirements of AI services, device capabilities, etc., so that the network can better understand and respond to the needs of the device.

[0157] This embodiment can accurately understand the target device requirements through the first signaling and its extension field, avoiding waste of resources.

[0158] In one example embodiment, receiving the measurement configuration information sent by the base station includes: receiving the second signaling sent by the base station; and parsing the measurement configuration information from the second extension field in the second signaling.

[0159] Optionally, the second signaling in this embodiment is similar in type to the first signaling described above, and will not be repeated here.

[0160] Optionally, the second signaling, the third signaling, and the first signaling described above in this embodiment are similar in type, and will not be repeated here.

[0161] Optionally, the second extension field in this embodiment is an additional field in the second signaling that is specifically used to describe and carry measurement configuration information. It extends the function of the traditional signaling to support complex target service quality monitoring requirements. For example, this field may contain detailed configurations on how to measure AI service performance, which indicators to use, and when to report measurement results.

[0162] The second signaling in this embodiment contains the second extension field, which allows the base station to explicitly indicate to the target device which specific service quality measurements are needed, ensuring the accuracy of the measurements and avoiding waste of resources caused by irrelevant measurements.

[0163] In an example embodiment, the measuring the processing of the target service by the base station based on the measurement configuration information to obtain a first measurement result comprises: measuring the processing of the target service by the base station based on the measurement configuration information to obtain the first measurement result, wherein the measuring the processing of the target service by the base station comprises at least one of the following: measuring a response delay of the base station in response to the service request, measuring a processing accuracy of the base station in processing the target service, measuring a communication quality between the base station and the target device, and measuring a computing resource used by the base station in processing the target service.

[0164] Optionally, the first measurement result in the embodiment includes measurement data of the AI service quality of the original base station and the neighboring base station by the target device. This includes but is not limited to response delay, model accuracy, computing resource utilization, and other key indicators. These data are used to evaluate the ability of the original base station in processing specific AI service, whether it can meet the requirements of the target service on response time and accuracy, etc.

[0165] The base station switching mechanism based on the first measurement result in the embodiment can actively monitor and respond to the fluctuation of the base station service quality, and ensure that the target device can continuously obtain high-quality task service in the movement.

[0166] In an example embodiment, the sending the first measurement result to the base station comprises: sending third signaling to the base station to send the first measurement result to the base station, wherein the first measurement result is included in a third extension field in the third signaling.

[0167] Optionally, the third signaling in the embodiment is similar to the first signaling in the above embodiment, and will not be described here.

[0168] Optionally, the third extension field in the embodiment is an additional field in the third signaling, which contains the first measurement result. For example, the measurement data of the AI service quality of the current serving cell by the UE, such as response delay, computing resource availability, accuracy, etc., and the measurement result of the neighboring cell, are used to evaluate the potential switching target.

[0169] The first measurement result is sent to the base station through the third extension field of the third signaling in the embodiment, so that the base station can timely perform the switching of the base station according to the first measurement result, and ensure that the service quality always meets the service requirements.

[0170] The determination method of the measurement result in the embodiments of the present application will be explained in combination with optional examples. In the optional examples, an AI QoS-oriented switching measurement mechanism is proposed, which measures AI service quality key indicators (such as service response delay, accuracy, etc.) in combination with user service demand, to ensure that the base station or cell switching can be triggered in time and accurately when the AI service quality of the UE is impaired, and to ensure that the UE continuously obtains high-quality AI services.

[0171] In the embodiments, before triggering the base station or cell switching, the measurement indicators of AI QoS can be defined first to quantify and evaluate the AI service performance. The measurement indicators include but are not limited to:

[0172] 1) AI service end-to-end response delay (E2E_Response_Latency): indicating the end-to-end delay (ms) from initiating the AI service request from the UE to the base station to receiving the AI service response, the end-to-end response delay including the AI model calculation delay and the communication delay. The time stamps of AI service request sending and response receiving can be recorded, and the difference between the two generates the end-to-end response delay, and the calculation reference formula is AI service response receiving time - request sending time.

[0173] 2) AI service accuracy (Accuracy): mainly indicating the accuracy of the AI model inference result. The representation and calculation method of AI service accuracy depend on the specific application scenario and task type. For example, in the classification task, the AI service accuracy is usually represented in the form of percentage (%), and the calculation reference formula is the number of correct samples in the inference data set ÷ the total number of samples in the inference data set; while in the time series prediction, regression analysis and other tasks, the accuracy may be measured by mean square error (MSE), mean absolute error (MAE), root mean square error (RMSE) and other indicators. In specific applications, the appropriate accuracy representation and calculation method should be selected according to the task type and demand.

[0174] 3) Base station computing resource availability (Comp_Available): which can be used to measure the utilization of the current computing resources of the base station (such as the utilization rate of CPU, GPU, TPU and other computing units), and is an important indicator to evaluate whether the base station can undertake AI computing tasks. The base station computing resource availability can be defined as the proportion (%) of the idle computing resources available for AI services on the base station side to the total computing resources. The calculation reference formula is 1 - occupied computing resources ÷ total computing resources.

[0175] 4) Base Station Latency (BS_Latency): indicates the computing latency (ms) of the base station side AI model, which is determined by the computing capability (such as the floating point operation speed FLOPS of the computing chip) of the base station and the computing complexity (such as the number of floating point operations FLOPs) of the base station side model. The time stamps of the start and completion of the AI model computing task of the base station can be recorded, and the difference between the two generates the base station computing latency. The calculation reference formula is the completion time of the base station side AI computing task minus the start time of the base station side AI computing task. In theory, the calculation reference formula can also be the computing complexity of the base station side model ÷ the base station computing capability.

[0176] 5) UE Supported AI Model Size (UE_Model_Size): can be defined as the maximum number of parameters of the AI model that the UE can support, which is an important indicator of UE capability and helps the base station to make resource matching and switching decisions. This indicator mainly depends on the memory of the UE, and the calculation reference formula is the memory resources available for AI models on the UE side ÷ the size of each parameter. For example, the parameter type of the model is float32, that is, each parameter occupies 4 bytes, and if the maximum memory resources available for AI models on the current UE is 6000MB, then the maximum number of parameters of the AI model that the current UE can support is 6000x1024 ÷ 4 ≈ 1.5M parameters.

[0177] 6) Measurement indicators based on AI service quality, new definitions consider AI service quality measurement events, including but not limited to:

[0178] Event XA1 (Event XA1): the AI service quality of the serving cell is higher than an absolute threshold value. This event can be used to turn off the measurement of the adjacent cell when the UE measures that the AI service quality of the serving cell is higher than the set threshold.

[0179] Event XA2 (Event XA2): the AI service quality of the serving cell is lower than an absolute threshold value.

[0180] Event XA3 (Event XA3): the AI service quality of the adjacent cell is higher than the AI service quality of the serving cell by a bias value.

[0181] Event XA4 (Event XA4): the AI service quality of the adjacent cell is higher than an absolute threshold value.

[0182] Event XA5 (Event XA5): the AI service quality of the serving cell is lower than an absolute threshold value, and the AI service quality of the adjacent cell is higher than another absolute threshold value.

[0183] Event XA6 (Event XA6): the AI service quality of the adjacent cell is higher than the AI service quality of the secondary cell by a bias value.

[0184] Event XB1 (Event XB1): AI service quality of a non-UE-associated cell is higher than an absolute threshold. Non-UE-associated cell refers to a cell of a different system (e.g., a 4G communication system, a 5G communication system).

[0185] Event XB2 (Event XB2): AI service quality of a serving cell is lower than an absolute threshold, and AI service quality of a non-UE-associated cell is higher than another absolute threshold.

[0186] In this embodiment, the measurement events (XA1-XA6, XB1-XB2) of AI service quality can be triggered by a single index or multiple indexes in combination, and the threshold values can be dynamically configured through signaling. When the events use different trigger quantities (e.g., E2E_Response_Latency, Accuracy, Comp_Available, etc.), the "AI service quality" in the above event definitions can be modified to the meaning represented by the corresponding trigger quantity, so that the definition of the event is more accurate.

[0187] In addition, the measurement indexes and measurement events of AI service quality can be combined with the conventional wireless signal quality (e.g., RSRP, SINR) indexes and events (A1-A6, B1-B2) for joint measurement and analysis, to form a comprehensive decision basis for handover triggering.

[0188] In one specific embodiment, assume that a vehicle (UE) equipped with an automatic driving system is using a real-time target detection AI service (model: YOLOv7) provided by a base station (gNB). The AI model (i.e., YOLOv7) is divided into a UE-side model and a base station-side model. The UE and the base station cooperatively calculate a final inference result. The UE requires that the end-to-end response latency of the AI service be ≤100 ms, the accuracy be ≥90%, and the AI model size received by the UE be ≤200M. During the movement of the vehicle, the computing load of the current serving base station increases and the wireless signal quality between the base station and the UE decreases, resulting in an end-to-end response latency of the AI service greater than 100 ms, so that the AI service quality requirement of the UE cannot be met, triggering base station handover. The specific process is shown in Figure 4 and includes the following steps:

[0189] S401, the UE establishes an RRC connection with the base station. During the process of accessing the base station, the UE reports the QoS service request of the AI service through RRC SetupComplete, for example, the maximum end-to-end response latency of the AI service, the minimum accuracy, the maximum AI model size that the UE can support, the computing or communication capability of the UE itself, etc. The specific content in the QoS service request is shown in Table 1.

[0190] Table 1:

[0191]

[0192] S402, the gNB configures the measurement configuration information (including the measurement parameters of the AI service quality and the traditional wireless signal quality) by RRC Reconfiguration, specifies the triggering of the XA2 event (the end-to-end response time of the current serving base station is greater than 100 ms or the AI service accuracy is less than 90%) and the A3 event (the signal quality of the neighboring cell is better than the serving cell by a certain threshold), as shown in Table 2.

[0193] Table 2:

[0194]

[0195] S403, the UE and the base station perform measurement according to the measurement configuration information. The UE measures the AI service quality indicators (end-to-end response time, accuracy, maximum AI model size supported by the UE, etc.) and the wireless signal quality indicators (RSRP, RSRQ). In addition, the gNB can also measure the computing delay of the base station side AI model, the current computing resource availability, etc. For the acquisition of the base station AI capability information, the UE can be through SIB broadcast or RRC signaling, etc. The base stations can interact through the XnAP interface.

[0196] S404, the UE detects that the AI service end-to-end response time of the current base station rises to 115 ms, or the signal quality of the neighboring cell is better than the serving cell by a certain threshold, triggering the XA2 event or the A3 event. The UE reports the first measurement result (MeasurementReport), including the AI service quality indicators and the wireless signal quality indicators of the serving base station and the neighboring cell, as shown in Table 3.

[0197] Table 3:

[0198]

[0199] S405, the handover process is performed. The current serving gNB generates a handover decision, comprehensively considers the AI service quality indicators (end-to-end response time, accuracy, base station computing resource availability, etc.) and the wireless signal quality indicators (RSRP, RSRQ, etc.), determines the target base station to be switched and initiates the handover process, and then performs the handover process.

[0200] The embodiment ensures that the UE can continuously obtain high-quality AI services in movement, i.e., with low delay and high accuracy, improving the user experience. By monitoring the availability of computing resources and the processing efficiency of AI services, the network can intelligently schedule resources, avoiding resource waste and overload. Combined with the monitoring of AI service quality and wireless signal quality, the handover decision is more comprehensive and accurate, reducing unnecessary handover and improving service continuity and network efficiency.

[0201] It should be noted that, for each of the foregoing method embodiments, for the sake of simple description, each is described as a combination of a series of actions, but those skilled in the art should know that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0202] From the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and a necessary general hardware platform, and of course it can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as a Read-Only Memory (ROM) / Random Access Memory (RAM), a magnetic disk, an optical disk) and includes a number of instructions for causing an end device (which can be a mobile phone, a computer, a server, or a network device, etc.) to perform the methods described in the various embodiments of the present application.

[0203] According to another aspect of the embodiments of the present application, a measurement result determination apparatus is also provided, which can be used to implement the measurement result determination method provided in the above embodiments, which has been described and will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware or a combination of software and hardware is also possible and is contemplated.

[0204] Figure 5 is a structural block diagram of an optional measurement result determination apparatus according to an embodiment of the present application, as shown in Figure 5 The measurement result determination apparatus includes:

[0205] The first memory 502, the first processor 504, and the first computer program 5002 stored on the first memory 502 and executable on the first processor 504, the first processor implements the following operations when executing the first computer program:

[0206] receiving a service request sent by a target device, wherein the service request includes processing requirements of the target device for a target service, and the target service includes an artificial intelligence (AI) service;

[0207] generating measurement configuration information based on the service request, wherein the measurement configuration information is used to guide the target device to perform measurement of the service quality of the target service, the measurement configuration information comprises measurement object configuration and measurement result configuration, the measurement object configuration is used to represent a network entity corresponding to the target service to be measured, and the measurement result configuration is used to represent a time and a manner in which the target device sends a measurement result;

[0208] sending the measurement configuration information to the target device, and receiving the first measurement result sent by the target device, wherein the first measurement result is a result obtained by the target device based on the measurement configuration information and measurement of the service quality of the target service.

[0209] According to the application, since the base station receives the service request sent by the target device, the base station can immediately understand the specific processing requirement of the target device for the AI service, and accordingly generate corresponding measurement object configuration and measurement result configuration, which enables the network to be dynamically adjusted to better adapt to the change of the AI service requirement. The measurement configuration information is sent to the target device, and the target device measures the service quality of the AI service based on the measurement configuration information. The measurement configuration information provides accurate measurement instructions for the target device, and can ensure the accuracy and pertinence of the measurement result. Therefore, the measurement result can be more accurately determined, and whether to switch the connection between the target device and the base station can be determined in a timely manner, so as to ensure the continuity and service quality of the AI service and promote the efficient use of network resources. Therefore, the technical problem that the service quality requirement of the target device for the AI service cannot be accurately measured in the related art, and the base station cannot be switched in a timely manner to process the AI service can be solved, and the effect that the processing of the AI service by the base station can be accurately measured, and the base station can be switched in a timely manner when the processing of the AI service by the base station does not meet the service quality requirement of the device can be achieved.

[0210] In one example embodiment, the first processor implements the following operation when the first computer program is executed: after the measurement configuration information is generated based on the service request, the service quality of the target service by the base station is measured by using the measurement configuration information to obtain a second measurement result, wherein the measurement of the service quality of the target service by the base station comprises at least one of the following: measurement of processing delay of processing the target service, measurement of processing accuracy of processing the target service, and measurement of computing resources used for processing the target service.

[0211] In an example embodiment, the first processor, when executing the first computer program, implements the following operation: sending the measurement configuration information to the target device, and receiving the first measurement result sent by the target device, and generating handover decision information based on the measurement result, wherein the handover decision information is used to indicate whether to switch the connection between the target device and the base station.

[0212] In an example embodiment, the first processor, when executing the first computer program, implements the following operation: generating the handover decision information based on the first measurement result, and / or the second measurement result.

[0213] In an example embodiment, the first processor, when executing the first computer program, implements the following operation: generating the handover decision information in a case where it is determined from the first measurement result, and / or the second measurement result that the processing of the target service by the base station does not meet the processing requirement.

[0214] In an example embodiment, the first processor, when executing the first computer program, implements the following operation: receiving the first signaling sent by the target device; and parsing the service request from a first extension field in the first signaling.

[0215] In an example embodiment, the processing requirement includes at least one of the following: a processing delay requirement of the target service, a processing accuracy requirement of the target service, a computing resource requirement of a base station processing the target service, a computing capability requirement of a first network model processing the target service, a processing delay requirement of the first network model processing the target service, a computing resource requirement of the first network model processing the target service, a computing resource requirement of a second network model processing the target service, a processing delay requirement of the second network model processing the target service, a memory requirement of the target device, a computing capability requirement of the target device, a memory requirement of the first network model, a memory requirement of the second network model, the first network model being deployed in the base station, and the second network model being deployed in the target device.

[0216] In an example embodiment, the first processor, when executing the first computer program, implements the following operation: in response to the service request, determining resource of the base station and capability information of the target device, wherein the resource of the base station and the capability information of the target device both correspond to the target service, and the resource of the base station includes at least one of the following: a computing resource, a model resource, a communication resource, and a data resource; and generating the measurement configuration information based on the resource of the base station, the capability information of the target device, and the processing requirement.

[0217] In an example embodiment, the measurement configuration information comprises at least one of: a measurement index configuration, used to indicate a performance index of a target service quality of service to be measured; and a measurement identifier configuration, used to identify the measurement configuration information.

[0218] In an example embodiment, the first processor, when executing the first computer program, implements the following operation: sending second signaling to the target device, wherein a second extension field in the second signaling comprises the measurement configuration information; receiving third signaling sent by the target device; and parsing the first measurement result from a third extension field in the third signaling.

[0219] In an example embodiment, the first processor, when executing the first computer program, implements the following operation: in a case where it is determined from the first measurement result that the processing of the target service by the base station does not meet the processing requirement, switching the target device to another base station, wherein the another base station is a base station whose processing of the target service meets the processing requirement.

[0220] In an example embodiment, the first processor, when executing the first computer program, implements the following operation: after generating the measurement configuration information based on the service request, measuring the processing of the target service by the base station using the measurement configuration information to obtain a second measurement result, wherein measuring the processing of the target service by the base station comprises at least one of: measuring a processing delay of processing the target service, measuring a processing accuracy of processing the target service, and measuring a computing resource used for processing the target service; and switching the connection between the target device and the base station based on the second measurement result.

[0221] In an example embodiment, the first processor, when executing the first computer program, implements the following operation: in a case where it is determined from the second measurement result that the processing of the target service by the base station does not meet the processing requirement, and / or in a case where it is determined from the first measurement result that the processing of the target service by the base station does not meet the processing requirement, switching the target device to another base station, wherein the another base station is a base station whose processing of the target service meets the processing requirement.

[0222] Figure 6 is a structural block diagram of another optional measurement result sending device according to an embodiment of the present application, as shown in Figure 6 the measurement result sending device comprises:

[0223] The second memory 602, the second processor 604, and the second computer program 6002 stored on the second memory 602 and capable of running on the second processor 604, the second processor implements the following operations when executing the second computer program:

[0224] sending a service request to the base station, wherein the service request includes a processing requirement of a target device for a target service, and the target service includes an artificial intelligence (AI) service;

[0225] receiving measurement configuration information sent by the base station, wherein the measurement configuration information is generated by the base station based on the service request, and includes measurement object configuration and measurement result configuration, the measurement object configuration is used to indicate a network entity corresponding to the target service to be measured, and the measurement result configuration is used to indicate a time and a manner of sending a measurement result by the target device;

[0226] measuring a quality of service of the target service based on the measurement configuration information to obtain a first measurement result;

[0227] sending the first measurement result to the base station, wherein the base station is configured to generate handover decision information based on the first measurement result, and the handover decision information is used to indicate whether to switch a connection between the target device and the base station.

[0228] According to the embodiments provided in the present application, since the target device sends a service request to the base station, the base station can immediately understand the specific processing requirement of the target device for the target service, and accordingly generate corresponding measurement configuration, which enables the network to be dynamically adjusted to better adapt to the change of service requirement; the measurement configuration information provides accurate measurement instructions for the target device, and can ensure the accuracy and pertinence of the measurement result; so that the measurement result can be more accurately determined; based on the first measurement result received from the target device, the base station can timely judge whether the processing of the target service meets the service requirement of the target device, so that the target device can be timely switched to a more suitable base station in the case that the service requirement is not met, to ensure the service continuity and service quality, and promote the efficient use of network resources. Therefore, the technical problem that the service quality requirement of the target device for the AI service cannot be accurately measured in the related art, and the base station processing the AI service cannot be timely switched, can be solved, and the effect that the processing of the AI service by the base station can be accurately measured, and the base station can be timely switched when the processing of the AI service by the base station does not meet the service quality requirement of the device, can be achieved.

[0229] In an example embodiment, the first processor, when executing the first computer program, implements the following operation: sending first signaling to the base station, so as to send a service request to the base station, wherein the service request is included in a first extension field in the first signaling.

[0230] In an example embodiment, the processing requirement comprises at least one of the following: a processing latency requirement of the target service, a processing precision requirement of the target service, a computing resource requirement of a base station processing the target service, a computing capability requirement of a first network model processing the target service, a processing latency requirement of the first network model processing the target service, a computing resource requirement of the first network model processing the target service, a computing resource requirement of a second network model processing the target service, a processing latency requirement of the second network model processing the target service, a memory requirement of the target device, a computing capability requirement of the target device, a memory requirement of the first network model, a memory requirement of the second network model, the first network model being arranged in the base station, and the second network model being arranged in the target device.

[0231] In an example embodiment, the measurement configuration information comprises at least one of the following: a measurement index configuration, used to indicate a performance index of a quality of service of a target service to be measured; and a measurement identifier configuration, used to identify the measurement configuration information.

[0232] In an example embodiment, the first processor, when executing the first computer program, implements the following operation: receiving second signaling sent by the base station; and parsing the measurement configuration information from a second extension field in the second signaling.

[0233] In an example embodiment, the first processor, when executing the first computer program, implements the following operation: measuring a quality of service of the target service of the base station by using the measurement configuration information, so as to obtain the first measurement result, wherein the measuring the quality of service of the target service of the base station comprises at least one of the following: measuring a response latency of the base station in response to the service request, measuring a processing precision of the base station processing the target service, measuring a communication quality between the base station and the target device, and measuring a computing resource used by the base station to process the target service.

[0234] In an example embodiment, the first processor, when executing the first computer program, implements the following operation: sending third signaling to the base station, so as to send the first measurement result to the base station, wherein the first measurement result is included in a third extension field in the third signaling.

[0235] It should be noted that the above various modules can be implemented by software or hardware, and for the latter, the implementation can be achieved by the following ways, but is not limited to: all the above modules are located in the same processor; or the above various modules are located in different processors in any combination.

[0236] According to another aspect of the embodiments of the present application, a computer readable storage medium is provided, which includes a stored program, wherein the program performs the steps in any of the above method embodiments when executed.

[0237] In an example embodiment, the above computer readable storage medium can include, but is not limited to: a U disk, a ROM, a RAM, a mobile hard disk, a magnetic disk or an optical disk, and various computer program storage media.

[0238] According to another aspect of the embodiments of the present application, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor is configured to execute the steps in any of the above method embodiments through the computer program. In an example embodiment, the above electronic device can further include a transmission device and an input and output device, wherein the transmission device is connected to the processor, and the input and output device is connected to the processor.

[0239] The specific examples in the present embodiment can refer to the examples described in the above embodiments and example implementations, which will not be described here again.

[0240] According to another aspect of the embodiments of the present application, a computer program product is also provided, which includes computer programs / instructions containing program codes for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from the network through the communication part 709, and / or installed from the detachable medium 711. When the computer program is executed by the central processing unit 701, various functions provided by the embodiments of the present application are executed. The above serial numbers of the embodiments of the present application only describe the embodiments, and do not represent the advantages or disadvantages of the embodiments.

[0241] Figure 7 The computer system structure block diagram of the electronic device for implementing the embodiments of the present application is schematically shown. As shown in FIG. 7, the computer system includes a central processing unit 701, a memory 702, a storage device 703, a communication part 704, an input and output device 705, and a display device 706. The central processing unit 701 is connected to the memory 702, the storage device 703, the communication part 704, the input and output device 705, and the display device 706. Figure 7As shown, the computer system 700 includes a central processing unit (CPU) 701 which can perform various appropriate actions and processes according to programs stored in a ROM 702 or loaded into a RAM 703 from a storage section 708. In the random access memory 703, various programs and data required for the operation of the system are also stored. The central processing unit 701, the read only memory 702, and the random access memory 703 are connected to each other through a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0242] Connected to the I / O interface 705 are an input section 706 including a keyboard, a mouse, etc.; an output section 707 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a local area network card, a modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the input / output interface 705 as necessary. A removable recording medium 711 such as a magnetic disk, an optical disc, a magneto-optical disc, a semiconductor memory, etc. is attached to the drive 710 as necessary, so that a computer program read therefrom is installed into the storage section 708 as necessary.

[0243] In particular, according to embodiments of the present application, the processes described in the various method flowcharts can be implemented as a computer software program. For example, embodiments of the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via the communication section 709, and / or installed from the removable recording medium 711. When the computer program is executed by the central processing unit 701, various functions defined in the systems of the present application are performed.

[0244] It should be noted that, Figure 7 The computer system 700 of the electronic device shown is merely an example and should not impose any limitation on the functions and the range of use of embodiments of the present application.

[0245] It is apparent that those skilled in the art can modify and / or change the above-described modules or steps of the present application with general computing devices, which can be centralized on a single computing device or distributed on a network composed of multiple computing devices, which can be implemented by program codes executable by the computing devices, so that they can be stored in storage devices and executed by the computing devices, and in some cases, the steps shown or described can be executed in different orders, or they can be made into individual integrated circuit modules, or multiple modules or steps can be made into a single integrated circuit module. Thus, the present application is not limited to any particular combination of hardware and software.

[0246] The above is only the preferred embodiment of the present application, and is not intended to limit the present application. Those skilled in the art can make various modifications and changes to the present application. Any modification, equivalent replacement, improvement, etc. within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method of determining a measurement result, characterized by, The method comprises: receiving a service request sent by a target device, wherein the service request comprises a processing requirement of the target device for a target service; generating measurement configuration information based on the service request, wherein the measurement configuration information is used to guide measurement of service quality of the target service; determining a measurement result based on the measurement configuration information, wherein the measurement result is obtained by measuring the service quality of the target service; wherein the generating of the measurement configuration information based on the service request comprises: determining resources of a base station and capability information of the target device in response to the service request, wherein the resources of the base station and the capability information of the target device both correspond to the target service, the resources of the base station comprise at least one of the following: computing resources, model resources, communication resources, and data resources; and generating the measurement configuration information based on the resources of the base station, the capability information of the target device, and the processing requirement.

2. The method of claim 1, wherein, The determining of the measurement result based on the measurement configuration information comprises at least one of the following: sending the measurement configuration information to the target device and receiving a first measurement result sent by the target device, wherein the first measurement result is obtained by the target device based on the measurement configuration information by measuring the service quality of the target service; measuring service quality of the target service of the base station by using the measurement configuration information to obtain a second measurement result, wherein the measurement of the service quality of the target service of the base station comprises at least one of the following: measuring processing delay of processing the target service, measuring processing accuracy of processing the target service, and measuring computing resources used for processing the target service.

3. The method of claim 2, wherein, After the determining of the measurement result based on the measurement configuration information, the method further comprises: generating handover decision information based on the measurement result, wherein the handover decision information is used to indicate whether to hand over a connection between the target device and the base station.

4. The method of claim 3, wherein, The generating of the handover decision information based on the measurement result comprises: generating the handover decision information based on the first measurement result and / or the second measurement result.

5. The method of claim 4, wherein, The generating of the handover decision information based on the first measurement result and / or the second measurement result comprises: generating the handover decision information in a case where it is determined from the first measurement result and / or the second measurement result that the processing of the target service by the base station does not meet the processing requirement.

6. The method of claim 1, wherein, The processing requirements include at least one of the following: a processing delay requirement of the target service, a processing accuracy requirement of the target service, a computing resource requirement of a base station processing the target service, a computing capability requirement of a first network model processing the target service, a processing delay requirement of the first network model processing the target service, a computing resource requirement of the first network model processing the target service, a computing resource requirement of a second network model processing the target service, a processing delay requirement of the second network model processing the target service, a memory requirement of the target device, a computing capability requirement of the target device, a memory requirement of the first network model, a memory requirement of the second network model, the first network model being deployed in the base station, and the second network model being deployed in the target device.

7. The method of claim 1, wherein, Receiving a service request sent by a target device, including: Receiving first signaling sent by the target device; Parsing the service request from a first extension field in the first signaling.

8. The method of claim 1, wherein, The measurement configuration information includes at least one of the following: measurement quantity configuration, used to indicate a performance indicator of the target service quality to be measured; measurement object configuration, used to indicate a network entity corresponding to the target service to be measured; measurement result configuration, used to indicate a time and a manner of sending a measurement result by the target device; and measurement identification configuration, used to identify the measurement configuration information.

9. The method of claim 2, wherein, Sending the measurement configuration information to the target device and receiving first measurement results sent by the target device, including: Sending second signaling to the target device, wherein the measurement configuration information is included in a second extension field in the second signaling; Receiving third signaling sent by the target device; Parsing the first measurement results from a third extension field in the third signaling.

10. A method of transmitting measurement results, characterized by Including: Sending a service request to a base station, wherein the service request includes processing requirements of a target device for a target service; Receiving measurement configuration information sent by the base station, wherein the measurement configuration information is information generated by the base station based on the service request; Measuring the service quality of the target service based on the measurement configuration information to obtain first measurement results; Sending the first measurement results to the base station, wherein the base station is configured to generate handover decision information based on the first measurement results, wherein the handover decision information is used to indicate whether to switch a connection between the target device and the base station; The measurement configuration information is information generated by the base station based on the service request by: in response to the service request, determining resource of the base station and capability information of the target device, wherein the resource of the base station and the capability information of the target device both correspond to the target service, and the resource of the base station includes at least one of the following: computing resource, model resource, communication resource, and data resource; and generating the measurement configuration information based on the resource of the base station, the capability information of the target device, and the processing requirements.

11. The method of claim 10, wherein, The processing requirements include at least one of the following: a processing delay requirement of the target service, a processing accuracy requirement of the target service, a computing resource requirement of a base station processing the target service, a computing capability requirement of a first network model processing the target service, a processing delay requirement of the first network model processing the target service, a computing resource requirement of the first network model processing the target service, a computing resource requirement of a second network model processing the target service, a processing delay requirement of the second network model processing the target service, a memory requirement of the target device, a computing capability requirement of the target device, a memory requirement of the first network model, a memory requirement of the second network model, the first network model being arranged in the base station, and the second network model being arranged in the target device.

12. The method of claim 10, wherein, sending a service request to the base station, including: sending first signaling to the base station to send a service request to the base station, wherein a first extension field in the first signaling includes the service request.

13. The method of claim 10, wherein, The measurement configuration information includes at least one of the following: measurement quantity configuration, used to indicate the performance indicators of the target service quality to be measured; measurement object configuration, used to indicate the network entity corresponding to the target service to be measured; measurement result configuration, used to indicate the time and manner of sending the measurement result by the target device; and measurement identification configuration, used to identify the measurement configuration information.

14. The method of claim 10, wherein, receiving the measurement configuration information sent by the base station, including: receiving second signaling sent by the base station; parsing the measurement configuration information from a second extension field in the second signaling.

15. The method of claim 10, wherein, measuring the service quality of the target service based on the measurement configuration information to obtain a first measurement result, including: measuring the service quality of the target service of the base station using the measurement configuration information to obtain the first measurement result, wherein measuring the service quality of the target service of the base station includes at least one of the following: measuring the response delay of the base station in response to the service request, measuring the processing accuracy of the base station processing the target service, measuring the communication quality between the base station and the target device, and measuring the computing resources used by the base station to process the target service.

16. The method of claim 11, wherein, sending the first measurement result to the base station, including: sending third signaling to the base station to send the first measurement result to the base station, wherein a third extension field in the third signaling includes the first measurement result.

17. A determination apparatus of a measurement result, characterized by, including a first memory, a first processor, and a first computer program stored on the first memory and executable on the first processor, wherein the first processor executes the first computer program to implement the following operations: receiving a service request sent by a target device, wherein the service request includes processing requirements of the target device for a target service; generating measurement configuration information based on the service request, wherein the measurement configuration information is used to guide the target device to perform measurement of the service quality of the target service; determining a measurement result based on the measurement configuration information, wherein the measurement result is a result of measuring quality of service of the target service; The first processor further implements the following operation when executing the first computer program: determining resource of a base station and capability information of the target device in response to the service request, wherein the resource of the base station and the capability information of the target device both correspond to the target service, and the resource of the base station includes at least one of the following: computing resource, model resource, communication resource, and data resource; and generating the measurement configuration information based on the resource of the base station, the capability information of the target device, and the processing requirement.

18. A transmission apparatus of a measurement result, characterized by, The second processor implements the following operation when executing the second computer program: sending a service request to a base station, wherein the service request includes a processing requirement of a target device for a target service; receiving measurement configuration information sent by the base station, wherein the measurement configuration information is information generated by the base station based on the service request; measuring quality of service of the target service based on the measurement configuration information to obtain a first measurement result; sending the first measurement result to the base station, wherein the base station is configured to generate handover decision information based on the first measurement result, wherein the handover decision information is used to indicate whether to switch connection between the target device and the base station; The measurement configuration information is information generated by the base station based on the service request in the following manner: determining resource of the base station and capability information of the target device in response to the service request, wherein the resource of the base station and the capability information of the target device both correspond to the target service, and the resource of the base station includes at least one of the following: computing resource, model resource, communication resource, and data resource; and generating the measurement configuration information based on the resource of the base station, the capability information of the target device, and the processing requirement.

19. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instruction is executed by the processor to implement the steps of the method in any one of claims 1 to 9, or implement the steps of the method in any one of claims 10 to 16.

20. A computer-readable storage medium, characterized in that, The computer program / instruction is executed by the processor to implement the steps of the method in any one of claims 1 to 9, or implement the steps of the method in any one of claims 10 to 16.

21. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, The computer program / instruction is executed by the processor to implement the steps of the method in any one of claims 1 to 9, or implement the steps of the method in any one of claims 10 to 16.

Images