A communication method and apparatus

Abstract

A communication method and device are used to make the analysis service of a data analysis network element meet the requirements of an analysis request network element. The method comprises: the data analysis network element receiving a request message from the analysis request network element, the request message being used to request recommended network parameters, the request message comprising network parameters required by the analysis request network element and network indexes expected by the analysis request network element. The data analysis network element can also determine the recommended network parameters according to the required network parameters and the expected network indexes. The data analysis network element can also send the recommended network parameters to the analysis request network element. The recommended network parameters correspond to the required network parameters and the expected network indexes of the analysis request network element, so the analysis result of the data analysis network element meets the requirements of the analysis request network element.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a communication method and apparatus. Background Technology

[0002] The rapid development of artificial intelligence and big data analytics has provided foundational technologies for network intelligence. To achieve 5G mobile network intelligence, a network data analytics function (NWDAF) network element and a management data analytics system (MDAS) are defined. NWDAF or MDAS can be used to provide intelligent analysis services for the network, supporting anomaly analysis, optimization, and service level agreement (SLA) assurance. In this application, NWDAF and / or MDAS may be referred to as data analytics network elements.

[0003] Taking NWDAF (Network Data Analyzer) as an example, NWDAF can predict the changing trends of network indicators (i.e., indicators representing the network's operational status) through intelligent analysis services, and the business network elements processing services can adjust the network based on these trends. However, the analysis services of current data analysis network elements cannot meet business requirements. Summary of the Invention

[0004] This application provides a communication method and apparatus that enables the analysis services of data analysis network elements to meet business requirements.

[0005] In a first aspect, embodiments of this application provide a communication method, which can be executed by a data analysis network element or a module (such as a chip) applied in the data analysis network element. Taking the execution of the method by a data analysis network element as an example, the method includes: the data analysis network element receiving a request message from an analysis request network element, the request message being used to request recommended network parameters, the request message including the network parameters required by the analysis request network element and the network indicators expected by the analysis request network element. The data analysis network element can also determine the recommended network parameters based on the required network parameters and the expected network indicators. The data analysis network element can also send the recommended network parameters to the analysis request network element.

[0006] According to the above scheme, the data analysis network element can determine the recommended network parameters based on the network parameters requested and the expected network indicators from the analysis requesting network element, and send the recommended network parameters to the analysis requesting network element to realize the recommendation of network parameters. The recommended network parameters correspond to the network parameters requested and the expected network indicators from the analysis requesting network element, thus enabling the analysis service of the data analysis network element to meet the business requirements of the analysis requesting network element.

[0007] In one possible design, the recommended network parameters are within the required network parameter range, and the predicted network metric corresponding to the recommended network parameters is outside the range of the desired network metric. This design improves the efficiency of recommending network parameters. For example, if no network parameter within the acceptable range has a predicted network metric within the desired range, then recommended network parameters are determined from the required range, and the predicted network metric corresponding to these recommended parameters is within the acceptable range of the network metric, which is greater than the expected range of the network metric.

[0008] In one possible design, the data analysis network element may also send the predicted network metrics corresponding to the recommended network parameters to the analysis request network element. If the predicted network metrics corresponding to the recommended network parameters are not within the range of expected network metrics, the data analysis network element may send the predicted network metrics corresponding to the recommended network parameters so that the analysis request network element can decide whether to accept the recommended network parameters based on the network metrics.

[0009] In one possible design, the recommended network parameters are not within the range of the required network parameters, but the predicted network metric corresponding to the recommended network parameters is within the range of the desired network metric. This design improves the efficiency of recommending network parameters. For example, if no network parameter within the required network parameter range has a predicted network metric within the range of the desired network metric, then recommended network parameters are determined from the acceptable range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is within the range of the desired network metric.

[0010] In one possible design, the data analysis network element can also send indication information to the analysis request network element, which is used by the analysis request network element to decide whether to accept the recommended network parameters. This design allows the data analysis network element to trigger the decision of whether the analysis request network element accepts the recommended network parameters, improving system coordination. In another possible implementation, if the analysis request network element accepts the recommended network parameters, it can also send a response information to the data analysis network element, indicating acceptance of the recommended network parameters.

[0011] In one possible design, the recommended network parameters are within the required network parameters, and the predicted network metrics corresponding to the recommended network parameters are within the expected network metrics. With this design, the recommended network parameters meet the requirements of the analysis requesting network element for network parameters and its expectations for network metrics, further enabling the analysis services of the data analysis network element to meet the requirements of the analysis requesting network element.

[0012] In one possible design, the data analysis network element can also send the tolerable range of the network parameters to the analysis request network element, whereby the recommended network parameters fall within the tolerable range. This design further ensures that the analysis service of the data analysis network element meets the requirements of the analysis request network element. Here, the tolerable range is the acceptable numerical range of the recommended network parameters; the service request network element can adjust the network parameters according to this tolerable range, and the network indicators corresponding to the adjusted network parameters will meet expectations.

[0013] In one possible design, the data analysis network element can also send a guarantee rate to the analysis request network element. The guarantee rate is the probability that the predicted network metric corresponding to the network parameters within the tolerance range can satisfy the predicted network metric corresponding to the recommended network parameters. With this design, the analysis request network element can decide whether to adjust the network parameters within the tolerance range based on the guarantee rate, further ensuring that the analysis service of the data analysis network element meets the requirements of the analysis request network element. For example, if the analysis request network element determines that the guarantee rate is too low, it can refuse to adjust the network parameters within the tolerance range. In this case, the analysis request network element can adjust the network parameters based on the recommended network parameters, or adjust the network parameters within a range smaller than the tolerance range but including the recommended network parameters.

[0014] In one possible design, the request message further includes requirement information indicating the recommended network parameters. The data analysis network element can then determine the recommended network parameters based on the requirement information, within the range of the required network parameters and / or within the range of network parameters corresponding to the desired network metric. With this design, the recommended network parameters better match the network parameter requirements of the service requesting network element, further enabling the data analysis network element's analysis services to meet the requirements of the analysis requesting network element.

[0015] In one possible design, the requirement information includes a cost function. For example, the cost function could be the objective function for finding the optimal solution using a trained model, and the requirement information could be used to indicate that the cost function of the recommended network parameters is minimized, thus minimizing the system overhead corresponding to the recommended network parameters.

[0016] In one possible design, the requirement information indicates that the recommended network parameters are the maximum or minimum values ​​within the range of the required network parameters, or the requirement information indicates that the recommended network parameters are the maximum or minimum values ​​within the range of the network parameters that satisfy the desired network metric.

[0017] Secondly, embodiments of this application provide a communication method that can be executed by an analysis request network element or a module (such as a chip) applied in the analysis request. Taking the execution of this method in an analysis request as an example, the method includes: the analysis request network element sending a request message to a data analysis network element, the request message being used to request recommended network parameters, the request message including the network parameters required by the analysis request network element and the network indicators expected by the analysis request network element. The analysis request network element can also receive recommended network parameters from the data analysis network element, the recommended network parameters being determined based on the required network parameters and the network indicators expected by the analysis request network element. The analysis request network element can also adjust the network parameters according to the recommended network parameters.

[0018] In one possible design, the recommended network parameters are within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is not within the range of the desired network metric.

[0019] In one possible design, the recommended network parameters are not within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is within the range of the desired network metric.

[0020] In one possible design, the analysis requesting network element may also receive instruction information from the data analysis network element; the analysis requesting network element may also determine whether to accept the recommended network parameters based on the instruction information.

[0021] In one possible design, the recommended network parameters are within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is not within the range of the desired network metric. The analysis requesting network element can also receive the predicted network metric corresponding to the recommended network parameters from the data analysis network element. The analysis requesting network element can determine whether to accept the recommended network parameters based on the predicted network metric corresponding to the recommended network parameters and the indication information.

[0022] In one possible design, the recommended network parameters are within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is within the range of the desired network metric.

[0023] In one possible design, the analysis requesting network element may also receive a tolerance range of the network parameters from the data analysis network element, wherein the recommended network parameters fall within the tolerance range. Specifically, the analysis requesting network element may adjust the network parameters within the tolerance range.

[0024] In one possible design, the analysis requesting network element may also receive a tolerance range for the network parameters from the data analysis network element, wherein the recommended network parameters fall within the tolerance range. The analysis requesting network element may also receive a guarantee rate from the data analysis network element, where the guarantee rate is the probability that the predicted network metric corresponding to the network parameters within the tolerance range can satisfy the predicted network metric corresponding to the recommended network parameters. Specifically, the analysis requesting network element may determine whether to adjust the network parameters within the tolerance range based on the guarantee rate.

[0025] In one possible design, the request message may also include requirement information for indicating the recommended network parameters.

[0026] In one possible design, the required information includes a cost function.

[0027] In one possible design, the requirement information indicates that the recommended network parameters are the maximum or minimum values ​​within the range of the required network parameters, or the requirement information indicates that the recommended network parameters are the maximum or minimum values ​​within the range of the network parameters that satisfy the desired network metric.

[0028] Thirdly, embodiments of this application provide a communication device, which may be a data analysis network element or a module (such as a chip) applied in a data analysis network element. This device has the function of implementing the first aspect described above and any possible design thereof. This function can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above-described function.

[0029] Fourthly, embodiments of this application provide a communication device, which may be an analysis request network element or a module (such as a chip) applied in an analysis request network element. This device has the function of implementing the second aspect described above and any possible design thereof. This function can be implemented in hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above-described function.

[0030] Fifthly, embodiments of this application provide a communication device, including a processor and a memory; the memory is used to store computer instructions, and when the device is running, the processor executes the computer instructions stored in the memory to cause the device to perform any implementation method in the first to second aspects and any possible designs described above.

[0031] In a sixth aspect, embodiments of this application provide a communication device including units or means for performing the steps of the first to second aspects and any possible designs thereof.

[0032] In a seventh aspect, embodiments of this application provide a communication device, including a processor and an interface circuit. The processor is configured to communicate with other devices via the interface circuit and execute the methods described in the first to second aspects and any possible designs thereof. The processor may include one or more devices.

[0033] Eighthly, embodiments of this application provide a communication device including a processor coupled to a memory, the processor being configured to invoke a program stored in the memory to execute the methods described in the first to second aspects and any possible designs thereof. The memory may be located within or outside the device. Furthermore, there may be one or more processors.

[0034] In a ninth aspect, embodiments of this application also provide a computer-readable storage medium storing instructions that, when executed on a communication device, cause the methods described in the first to second aspects and any possible designs thereof to be performed.

[0035] In a tenth aspect, embodiments of this application also provide a computer program product, which includes a computer program or instructions that, when executed by a communication device, cause the methods in the first to second aspects and any possible designs thereof to be performed.

[0036] Eleventhly, embodiments of this application also provide a chip system, including: a processor for executing the methods described in the first to second aspects and any possible designs thereof.

[0037] In a twelfth aspect, embodiments of this application also provide a communication system, including a data analysis network element for performing the methods described in the first aspect and any possible design thereof, and an analysis request network element for performing the methods described in the second aspect and any possible design thereof.

[0038] The beneficial effects of aspects two through twelfth and any possible designs above can be found in the description of the beneficial effects of aspect one and any possible designs, and will not be elaborated further. Attached Figure Description

[0039] Figure 1 This application provides a schematic diagram of the architecture of a communication system.

[0040] Figure 2 A schematic diagram of a machine learning model provided in an embodiment of this application;

[0041] Figure 3 This is a schematic diagram of the architecture of another communication system provided in an embodiment of this application;

[0042] Figure 4This application provides a flowchart illustrating a communication method.

[0043] Figure 5 A flowchart illustrating another communication method provided in an embodiment of this application;

[0044] Figure 6 A flowchart illustrating another communication method provided in an embodiment of this application;

[0045] Figure 7 A flowchart illustrating another communication method provided in an embodiment of this application;

[0046] Figure 8 This is a schematic diagram of the structure of a communication device provided in an embodiment of this application;

[0047] Figure 9 This is a schematic diagram of another communication device provided in an embodiment of this application. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the application will now be described in further detail with reference to the accompanying drawings.

[0049] Figure 1 This is a schematic diagram of a 5G network architecture based on a service-oriented architecture. Figure 1The 5G network architecture shown may include terminal devices, access network (AN) devices, and core network devices. Terminal devices access the data network (DN) through access network devices and core network devices. The core network devices include some or all of the following network functions (NFs): unified data management (UDM) network elements, network exposure function (NEF) network elements (not shown in the figure), application function (AF) network elements, policy control function (PCF) network elements, access and mobility management function (AMF) network elements, network slice selection function (NSSF) network elements, session management function (SMF) network elements, user plane function (UPF) network elements, network data analytics function (NWDAF) network elements, and network repository function (NRF) network elements (not shown in the figure).

[0050] Access network equipment can be radio access network (RAN) equipment. Examples include: base stations, evolved NodeBs (eNodeBs), transmission reception points (TRPs), next-generation NodeBs (gNBs) in 5G mobile communication systems, next-generation base stations in the 6th generation (6G) mobile communication systems, base stations in future mobile communication systems, or access nodes in wireless fidelity (WiFi) systems. It can also be a module or unit that performs some of the functions of a base station; for example, it can be a central unit (CU) or a distributed unit (DU). RAN equipment can be macro base stations, micro base stations, indoor stations, relay nodes, or donor nodes. The embodiments of this application do not limit the specific technologies or equipment forms used in the RAN equipment.

[0051] Terminal devices can be user equipment (UE), mobile stations, mobile terminals, etc. They can be widely used in various scenarios, such as device-to-device (D2D), vehicle-to-everything (V2X) communication, machine-type communication (MTC), the Internet of Things (IoT), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, smart transportation, and smart cities. Terminal devices can include mobile phones, tablets, computers with wireless transceiver capabilities, wearable devices, vehicles, urban air mobility vehicles (such as drones and helicopters), ships, robots, robotic arms, and smart home devices.

[0052] Access network equipment and terminal equipment can be fixed or mobile. They can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; and they can be deployed in the air on aircraft, balloons, and satellites. The embodiments of this application do not limit the application scenarios of the access network equipment and terminal equipment.

[0053] The access management network element (AMF) includes functions such as mobility management and access authentication / authorization. It is primarily used for terminal attachment, mobility management, and tracking area update procedures in mobile networks. The AMF terminates non-access stratum (NAS) messages, completes registration management, connection management, and reachability management, allocates tracking area lists (TAlists), and performs mobility management. It also transparently routes session management (SM) messages to the session management network element. In 5G communication systems, the AMF can be an AMF (Advanced Management Function) element. Furthermore, the AMF is responsible for transmitting user policies between terminal devices and the PCF (Programmable Component Provider).

[0054] Session management network elements are primarily used for session management in mobile networks, such as session establishment, modification, and release. Specific functions include assigning Internet Protocol (IP) addresses to terminals and selecting user plane network elements that provide packet forwarding capabilities. In 5G communication systems, the session management network element can be an SMF (Signature Management Function) element.

[0055] Network slicing selection elements are primarily used to select appropriate network slices for terminal services. In 5G communication systems, network slicing selection elements can be NSSF elements.

[0056] User plane network elements are primarily responsible for processing user packets, such as forwarding, billing, and lawful interception. User plane network elements can serve as Protocol Data Unit (PDU) session anchors (PSAs). In 5G communication systems, user plane network elements can be UPF (User Plane Filtering) elements. UPFs can communicate directly with the NWDAF (Network Window Filtering) through service-like interfaces, or through other means, such as through the SMF (Software Filtering) or a private or internal interface with the NWDAF.

[0057] Unified Data Management Network Element: Responsible for managing the terminal's subscription information. In a 5G communication system, the Unified Data Management Network Element can be a UDM network element (hereinafter referred to as UDM).

[0058] Network capability open elements are used to support the opening of capabilities and events. In 5G communication systems, network capability open elements can be NEF elements (hereinafter referred to as NEF).

[0059] Application function network elements are used to convey application-side requests to the network side, such as QoS requirements or user state event subscriptions. Application function network elements can be third-party functional entities or application servers deployed by the operator. In 5G communication systems, application function network elements can be AF network elements (hereinafter referred to as AF).

[0060] Policy control network elements include user subscription data management functions, policy control functions, billing policy control functions, and quality of service (QoS) control. In 5G communication systems, policy control network elements can be PCF network elements (hereinafter referred to as PCF). It should be noted that in actual networks, PCFs may also be divided into multiple entities according to hierarchy or function, such as global PCFs and PCFs within slices, or session management PCFs (SM-PCF) and access management PCFs (AM-PCF).

[0061] Network repository elements provide network element discovery functionality, offering network element information corresponding to the network element type based on requests from other network elements. They also provide network element management services, such as network element registration, updates, deregistration, and network element status subscription and push notifications. In 5G communication systems, network repository elements can be NRF elements (hereinafter referred to as NRFs).

[0062] A data analysis network element can be used to collect, analyze, and predict data. Data collection includes, but is not limited to, at least one of the following: data collected from various other network elements (NFs), such as data collected through the AMF, SMF, or PCF; data collected through the NEF or directly from the AF; or data collected from the operation, administration, and maintenance (OAM) system. This data can be data from terminal devices, access network devices, core network elements, or third-party application devices, or data from terminal devices on the access network device, core network element, or third-party application device. The collected data is then intelligently analyzed, and the analysis results are output. In a 5G communication system, the data analysis network element can be an NWDAF network element (hereinafter referred to as NWDAF). Intelligent analysis refers to the analysis of collected data using intelligent technologies such as artificial intelligence (AI). In this application, intelligent analysis includes, but is not limited to, predicting network indicators and recommending network parameters.

[0063] In this application, the NWDAF can utilize machine learning models for intelligent analysis. The NWDAF can also output recommended values ​​to the various NFs, AFs, or OAMs mentioned above, for use in policy decision-making. In 3GPP Release 17, the training and inference functions of the NWDAF are separated. An NWDAF can support only model training, only data inference, or both. The NWDAF supporting model training can also be called a training NWDAF, or an NWDAF supporting the model training logical function (MTLF) (abbreviated as NWDAF(MTLF)). The training NWDAF can train a model based on acquired data to obtain a trained model. The NWDAF supporting data inference can also be called an inference NWDAF, or an NWDAF supporting the analytics logical function (AnLF) (abbreviated as NWDAF(AnLF)). The inference NWDAF can input input data into the trained model to obtain analysis results or inference data. In this embodiment, a training NWDAF refers to an NWDAF that at least supports model training functionality. As a possible implementation, a training NWDAF can also support data inference functionality. An inference NWDAF refers to an NWDAF that at least supports data inference functionality. As a possible implementation, an inference NWDAF can also support model training functionality. If an NWDAF supports both model training and data inference functionality, then the NWDAF can be called a training NWDAF, an inference NWDAF, a training-inference NWDAF, or simply an NWDAF. In this embodiment, an NWDAF can be a separate network element or can be co-located with other network elements, such as being placed in a PCF network element or an AMF network element.

[0064] A Domain Provider (DN) is a network located outside of the carrier's network. A carrier's network can connect to multiple DNs, and various services can be deployed on a DN, providing data and / or voice services to terminal devices. For example, a DN might be the private network of a smart factory. Sensors installed in the workshop can act as terminal devices, and a control server for these sensors is deployed within the DN. The control server provides services to the sensors. Sensors can communicate with the control server, receive instructions from it, and transmit the collected sensor data back to the control server accordingly. Another example is a DN serving as an internal office network for a company. Employees' mobile phones or computers can act as terminal devices, accessing information and data resources on the company's internal office network.

[0065] Figure 1 Npcf, Nnef, Namf, Nudm, Nsmf, Naf, Nnssf, and Nnwdaf are the service interfaces provided by PCF, NEF, AMF, UDM, SMF, AF, NSSF, and NWDAF, respectively, used to invoke the corresponding service operations. N1, N2, N3, N4, and N6 are interface sequence numbers, with the following meanings:

[0066] 1) N1: The interface between the AMF and the terminal device, which can be used to transmit non-access stratum (NAS) signaling (such as QoS rules from the AMF) to the terminal device.

[0067] 2) N2: The interface between the AMF and the access network equipment, which can be used to transmit radio bearer control information from the core network side to the access network equipment.

[0068] 3) N3: The interface between the access network device and the UPF, mainly used to transmit uplink and downlink user plane data between the access network device and the UPF.

[0069] 4) N4: The interface between SMF and UPF, which can be used to transmit information between the control plane and the user plane, including the distribution of forwarding rules, QoS rules, traffic statistics rules, etc. from the control plane to the user plane, as well as the reporting of information from the user plane.

[0070] 5) N6: The interface between UPF and DN, used to transmit uplink and downlink user data streams between UPF and DN.

[0071] Figure 1 The service-oriented architecture shown enables the 5G core network to form a flat architecture. Through the control plane signaling bus, control plane network function entities in the same network slice can discover each other through NRF, obtain each other's access address information, and then communicate directly with each other through the control plane signaling bus.

[0072] It is understood that the aforementioned network element or function can be a network component in a hardware device, a software function running on dedicated hardware, or a virtualization function instantiated on a platform (e.g., a cloud platform). As one possible implementation method, the aforementioned network element or function can be implemented by a single device, multiple devices working together, or a functional module within a single device; this application does not specifically limit this.

[0073] As one implementation method, the data analysis network element in this application embodiment can be the aforementioned NWDAF, or a network element with the aforementioned NWDAF function in future communications such as 6G networks. The data analysis network element can also be MDAS. MDAS is a data analysis system deployed on the network management plane, which can be used to collect and analyze management data such as performance statistics, alarms, and operation configurations, and can also output suggestions for resource allocation or configuration optimization. MDAS also has training and inference functions. Compared with NWDAF, MDAS is part of the network management system, often operating offline and not in real-time, providing operators with resource and deployment adjustment and optimization suggestions, and focusing on longer-term trend analysis and optimization suggestions. For ease of explanation, the following description uses NWDAF as an example of the data analysis network element; actions performed by NWDAF in this application can also be performed by MDAS.

[0074] The following explains the intelligent analysis process of NWDAF. NWDAF can collect data from multiple sources and dimensions, perform correlation analysis, output historical statistics, or train and fit a model. Based on the model, it outputs predicted values ​​of network indicators to guide service network elements in adjusting network parameters to optimize network indicators. It should be understood that different network indicators correspond to different network parameters, and network indicators are related to the network operating status. Among them, network indicators include network service evaluation values ​​(hereinafter referred to as service experience), key network performance indicators, or network overhead indicators, etc. Network parameters may include factors such as time, UE location, application location, service flow bit rate, packet latency, and the number of transmitted and retransmitted packets.

[0075] This example uses network metrics for intelligent analysis of business experience (hereinafter referred to as business experience analysis). Business experience refers to the user's evaluation of the experience of accessing services through the network. In this case, network metrics can be a quantitative assessment by the user. This process may include the following steps:

[0076] Step 1, NWDAF first collects the following data:

[0077] (1) Collect service experience scores, the percentage of UEs that achieve the experience score (e.g., the percentage of UEs with excellent service experience quality is not less than 90%), UE IP addresses, and application location information (such as data network access identifier (DNAI)) from the AF. Among them, the experience score is, for example, the mean opinions score (MOS).

[0078] (2) Collect the UE's subscription permanent identifier (SUPI) and UE's location information (such as global cell identifier (GCI)) through the AMF;

[0079] (3) Collect UE's SUPI, PDU session network slice identifier (such as single-network slice selection assistance information (S-NSSAI)), UPF information (such as UPF identifier (ID)), IP filtering information and service flow identifier (QFI) from SMF;

[0080] (4) Collect parameters such as bit rate, end-to-end delay (or packet delay), number of transmitted and retransmitted messages, etc. from the UPF.

[0081] Step 2: NWDAF uses IP filtering information and the UE's IP address to associate data collected from the AF by a UE with data collected from the SMF by the same UE. Then, based on SUPI, it associates location data collected from the AMF by the same UE with session data collected from the SMF. QFI is then used to further associate data collected from the UPF by the same UE with the above data. Similarly, NWDAF performs correlation analysis on data from a large number of UEs.

[0082] Step 3: NWDAF trains and fits a model based on the data described above. For example, a deep learning network can be trained using the data. This deep learning network is, for example... Figure 2 As shown.

[0083] For example, in the training process, NWDAF uses the training function to treat network parameters such as UE location, application location, time, QoS Flow bit rate, packet latency, and the number of transmitted and retransmitted packets as independent variables, and network metrics such as service experience and the percentage of UEs achieving the corresponding service experience as dependent variables. This data, after being processed and correlated, is used to train the deep learning network, resulting in a deep learning model. In other words, during training, the independent variables are network parameters, and the dependent variables are network metrics.

[0084] Step 4: NWDAF sets the trained deep learning model to inference mode (i.e., uses inference function), predicts the most likely range of values ​​for each independent variable in the future based on the historical statistical trends of each independent variable, and then calculates the predicted result of the dependent variable in the future based on the predicted values ​​of each independent variable through the trained deep learning model.

[0085] Accordingly, through the service experience analysis process, NWDAF can predict the predicted values ​​of network metrics corresponding to network parameters. NWDAF can also send these predicted values ​​to service network elements (or service processing network elements), so that the service network elements can adjust the network parameters according to the predicted values, thereby optimizing the network metrics after the adjustment.

[0086] Specifically, in the service experience analysis process, the service network element may include the Service Management Function (SMF), network parameters may include QoS parameters, and network metrics may include the experience score. The Network Window Function (NWDAF) can output a predicted value of the experience score to the SMF. The SMF can determine adjusted QoS parameters based on the predicted experience score, which may specifically include adjusted bit rate and / or adjusted packet delay. The SMF can also execute the adjusted QoS parameters through the Upgraded Service Function (UPF), thereby improving the service score through QoS parameter optimization.

[0087] However, in the above business experience analysis process, NWDAF outputs predicted network metrics. Business network elements adjust network parameters based on these predicted values. But in some cases, the adjustment method determined by the business network element based on these predicted values ​​may not meet its requirements for network parameter adjustments. Taking the business experience analysis process as an example, if the NWDAF output predicts that the network parameter adjustment method corresponds to adjusting the end-to-end latency of the network slice, but the SMF (Service Flow Management) may not support this adjustment or may not want to perform it, or may want to adjust other network parameters such as bit rate, then the SMF may not be able to adjust the network parameters based on the predicted values, and therefore the predicted values ​​have no practical value. Therefore, NWDAF needs to analyze according to the requirements of the business network element, so that the analysis service of the data analysis network element meets the requirements of the network element requesting the analysis.

[0088] To ensure that the analysis services of the data analysis network element meet the requirements of the analysis requesting network element, embodiments of this application provide a communication method. This communication method can be executed by both the data analysis network element and the analysis requesting network element. For example... Figure 3 As shown, a data analysis network element can be used to perform intelligent analysis of the network based on a request message (or analysis request) from an analysis request network element, and send the analysis results (or the response message corresponding to the request message, or simply the response message) to the analysis request network element. For example, data analysis network elements include NWDAF or MDAS. The analysis request network element can be a network element within the network to be analyzed, or a network element outside the network. The analysis request network element can include service network elements used to adjust network parameters of the network based on the analysis results, or it can include other network elements besides service network elements. For example, the analysis request network element can be, for example, AMF or SMF, etc., without specific limitation. The network can include at least one network element, for example, including... Figure 1 At least one NF in the architecture shown.

[0089] The following is combined with Figure 4 This application provides a communication method that may include the following steps:

[0090] S101: The analysis requesting network element sends a request message to the data analysis network element. This request message can be used to request recommended network parameters. The request message may include at least one of the network parameters requested by the analysis requesting network element and the network metrics expected by the analysis requesting network element.

[0091] In this application, at least one of the network parameters required by the requesting network element and the network indicators expected by the requesting network element can be used to determine the analysis results, such that the number of analysis results meets the requirements of the requesting network element. The analysis results may include recommended network parameters, enabling the requesting network element to adjust its network parameters according to these recommended parameters to achieve better network optimization results.

[0092] Specifically, the network parameters requested by the requesting network element can include the type of network parameters requested, or both the type and value of the requested network parameters. The network parameters requested by the requesting network element can be an acceptable adjustment range for the network parameters for the requesting network element, allowing the data analysis network element to determine recommended network parameters based on this acceptable range, thus avoiding the recommended network parameters exceeding the acceptable range for the requesting network element. For example, taking bit rate and end-to-end latency as network parameters, the requested network parameters could indicate an acceptable bit rate less than or equal to 20 megabits per second (Mbps) and an acceptable end-to-end latency greater than or equal to 20 milliseconds (ms). It should be understood that the request message can carry a list of network parameters including at least one requested network parameter.

[0093] The analysis of network metrics desired by the requesting network element can include the type of network metric desired by the requesting network element, or the type and value (or range) of the desired network metric. The desired network metric can be a value that the requesting network element hopes the network metric will reach. For example, if the requesting network element wants to adjust network parameters to achieve a certain network metric value, it can send this value to the data analysis network element, allowing the data analysis network element to predict recommended network parameters to achieve that value. Therefore, the data analysis network element can determine the network parameters that will achieve the desired network metric, and use this information to determine the recommended network parameters. Taking MOS as an example, if the requesting network element expects a MOS of at least 4.5, the data analysis network element can analyze the network parameters that will ensure a MOS of at least 4.5, and determine the recommended network parameters based on these parameters, where 0 ≤ MOS ≤ 5.

[0094] One possible implementation is that the request message may further include recommended network parameter requirement information. This requirement information indicates how to determine the recommended network parameters from the range of network parameters required by the requesting network element and / or the range of network parameters corresponding to the network indicators expected by the requesting network element. Specifically, the requirement information can be used to indicate the maximum or minimum network parameter within the required range of network parameters and / or the range of network parameters corresponding to the network indicators expected by the requesting network element as the recommended network parameter; alternatively, the requirement information can be used to indicate the network parameter corresponding to the maximum or minimum value of the predicted network indicator as the recommended network parameter. Furthermore, the requirement information can also be used in a cost function. The cost function is the objective function used to find the optimal solution using a trained model, used to determine the optimal network parameter as the recommended value from multiple network parameters corresponding to the network indicators expected by the requesting network element. Specifically, the requirement information can be used to indicate that the cost function of the recommended network parameter is minimized, at which point the system overhead corresponding to the recommended network parameter is minimized.

[0095] One possible implementation is that if the request message includes the network metric expected by the network element requesting analysis, the request message may also include the expected percentage of the network metric that the network element expects to achieve that expected metric. Taking the network metric MOS as an example, this expected percentage can indicate the expected percentage of the user's MOS that the network element expects to achieve after adjusting the network parameters according to the analysis results corresponding to the data request message. For example, this expected percentage may be no less than 90%.

[0096] One possible implementation is that the request message may also include the analysis type requested by the analysis requesting network element, such as carrying an analysis type identifier like business experience analysis.

[0097] In this application, the request message shown in S101 can be a request for the data analysis network element to provide intelligent analysis services, or a subscription request for subscribing to analysis services. If it is a request to provide intelligent analysis services, the data analysis network element outputs the analysis results to the analysis requesting network element all at once, according to the request. If it is a subscription request for analysis services, the data analysis network element outputs the analysis results to the analysis requesting network element multiple times, either periodically or triggered by events, according to the request, until the analysis requesting network element cancels the subscription. If the request message is a subscription request, it may also include a subscription identifier to identify the subscription, allowing the analysis requesting network element to distinguish between different analysis subscriptions.

[0098] Accordingly, the data analysis network element receives request messages from the analysis request network element.

[0099] S102: After receiving the request message, the data analysis network element can determine recommended network parameters based on the network parameters requested by the requesting network element and / or the network indicators expected by the requesting network element. In one possible implementation, the recommended network parameters determined in S102 are within the range of network parameters requested by the requesting network element, and / or the predicted network indicators corresponding to the recommended network parameters are within the range of expected network indicators.

[0100] In S102, the data analysis network element can determine the recommended network parameters using the trained model. If no pre-trained model exists, the data analysis network element must first enter the model training phase to obtain a trained model. During the model training phase, the input data consists of data collected by the data analysis network element, including the model's independent variables (i.e., network parameters) and corresponding dependent variables (i.e., network metrics). The output is the structure and internal parameters of the network model, i.e., the trained model. At this point, the data analysis network element obtains the trained model. If the data analysis network element already has a trained model—for example, obtained through a previous training phase or received from another network element or device—it can use this trained model for inference, prediction, or recommendation. When the model is used for inference and prediction, the input data can include independent variables, and the output can include dependent variables. When the trained model is used to determine the recommended independent variables, the input data can include the predicted dependent variables (i.e., one or more network metrics to be achieved, specifically, the network metrics expected by the analysis requesting network element in S102), and the output can include at least one type of recommended independent variable (i.e., one or more network parameters).

[0101] Furthermore, the data analysis network element can determine the type of network parameters requested by the analysis requesting network element based on the request message. Based on the type of requested network parameters, it determines one or more types of independent variables from multiple types of independent variables in the trained model that require recommended values, serving as recommended network parameters. In addition, the data analysis network element can also determine other types of recommended network parameters besides the requested type (hereinafter referred to as unrequired network parameters). These unrequired network parameters can take the current value or historical average of the network parameters of that type, or they can take the predicted value of the network parameter of that type with the highest probability of occurrence in the future. Then, based on the model and the one or more predicted network indicators to be achieved, the data analysis network element analyzes and outputs recommended values ​​for one or more corresponding network parameters. These recommended values ​​include the recommended values ​​for the requested network parameters and may also include the recommended values ​​for the unrequired network parameters. For example, if the type of the requested network parameter is bit rate, the data analysis network element can determine the recommended bit rate and also determine the recommended end-to-end delay, and send the recommended bit rate and recommended end-to-end delay to the analysis requesting network element.

[0102] In one possible implementation, the output of the data analysis network element may further include the predicted proportion of network indicators reaching the predicted network indicators. This predicted proportion indicates the percentage by which the actual network indicators reach the predicted network indicators corresponding to a certain network parameter after adjustment. The data analysis network element may then send the predicted proportion corresponding to the recommended network parameters to the analysis request network element. This predicted proportion may be equal to or greater than the expected proportion stated in the request message, or it may be less than the expected proportion stated in the request message. The predicted proportion helps the analysis request network element determine whether to accept the recommended network parameters and make adjustments.

[0103] It should be understood that the training process of the model described here can be carried out in the data analysis network element, or other network elements can obtain the trained model through training and then send the model to the data analysis network element. If the data analysis network element determines the model, it can collect data and train the model according to a certain period, so it is not necessary to retrain the model every time network intelligent analysis is performed.

[0104] In the implementation of S102, the data analysis network element can determine the network parameters and the predicted values ​​of the corresponding network indicators through the model, and determine the network parameters with better predicted values ​​of the corresponding network indicators (hereinafter referred to as candidate network parameters). These candidate network parameters can be used to determine the recommended network parameters. The predicted network indicator corresponding to the network parameter refers to the network indicator that is the output result of the model when the network parameter is used as input data (or as part of the input data). In one possible implementation, the candidate network parameters are within the range of network parameters required by the analysis requesting network element, and / or, the predicted network indicators corresponding to the candidate network parameters are within the range of desired network indicators.

[0105] Taking business experience analysis as an example, if the network metric is MOS (Mean Orientation of Memory), the data analysis network element can determine network parameters that ensure the MOS is not lower than a threshold (e.g., 4.5) as candidate network parameters, and further determine the recommended network parameters based on the candidate network parameters. In one possible implementation, this threshold is the network metric expected by the analysis requesting network element.

[0106] One possible implementation is that, during the process of determining recommended network parameters, if the request message includes network parameters required by the analysis requesting network element, then in S102, the recommended network parameters determined by the data analysis network element are within the range of the required network parameters. For example, the data analysis network element uses the required network parameters as given input data to determine the output results, and determines the output result with the higher value within the range of the obtained output results. The network parameters corresponding to the output result with the higher value can be used as candidate network parameters. Taking service experience analysis as an example, if the network parameters required by the analysis requesting network element are, for example, a bit rate less than or equal to 20 Mbps and an end-to-end latency greater than or equal to 20 ms, then the recommended network parameters determined by the data analysis network element include the bit rate and the end-to-end latency, and the bit rate is not higher than 20 Mbps, and the end-to-end latency is not lower than 20 ms.

[0107] Furthermore, if the request message includes the network metrics expected by the requesting network element, then during the process of determining candidate network parameters, the data analysis network element can determine the range of network parameters corresponding to the expected network metrics based on the model. For example, the data analysis network element uses the expected network metrics as a given output to determine the input data, and the range of the resulting input data is the range of network parameters corresponding to the expected network metrics. Then, the data analysis network element can determine candidate network parameters from the range of network parameters corresponding to the expected network metrics, and subsequently determine the recommended network parameters based on the candidate network parameters. Taking business experience analysis as an example, if the network metric expected by the requesting network element is a MOS of no less than 4.5, then the recommended network parameters determined by the data analysis network element correspond to a MOS of no less than 4.5.

[0108] If the request message includes the expected percentage of the network metric that reaches the desired network metric, the model's output can also include the predicted percentage of the network metric that reaches the predicted network metric. During the process of determining candidate network parameters, the data analysis network element can identify network parameters that ensure the predicted percentage included in the output is not lower than the expected percentage as candidate network parameters. For example, if the expected percentage is not lower than 95%, the data analysis network element can identify network parameters that ensure the predicted percentage is not lower than 95% as candidate network parameters, and further determine the recommended network parameters from among the candidate network parameters.

[0109] One possible implementation is that if the model's output includes the predicted proportion of network metrics reaching the predicted network metrics, the data analysis network element can also determine the predicted proportion corresponding to the recommended network parameters. In this case, the model's input data includes the recommended network parameters, and the model's output includes the predicted proportion corresponding to the recommended network parameters. The data analysis network element can also send the predicted proportion corresponding to the recommended network parameters to the analysis request network element, indicating the proportion of the actual network metrics reaching the predicted network metrics after adjustments using the recommended network metrics.

[0110] In one possible implementation, the data analysis network element can also determine the tolerance range of network parameters, and the recommended network parameters fall within this tolerance range. The data analysis network element can also send this tolerance range to the analysis request network element. This tolerance range can be a numerical range including the recommended network parameters, representing acceptable actual network parameter values. It should be understood that, since the network indicators corresponding to the actual network parameters may deviate from the predicted network indicators corresponding to the recommended network parameters after the analysis request network element or other service network elements adjust the network parameters based on the analysis results, the data analysis network element can determine and indicate the acceptable range of actual network parameter values, i.e., the tolerance range, to the analysis request network element, enabling the analysis request network element to adjust the network parameters within this range. For example, the analysis request network element can ensure that the actual value of the adjusted network parameters does not exceed the tolerance range. In this application, the tolerance range can be represented by the distance between the center value and the boundary value of the tolerance range. The center value of the tolerance range can be the recommended network parameters. Taking bit rate as an example, if the recommended bit rate is 20 Mbps and the tolerance range is 19 Mbps to 21 Mbps, then the tolerance range can be represented by 20 ± 1 Mbps, where 1 Mbps is the radius of the tolerance range. Alternatively, a numerical range of 2 Mbps can be used, centered on the recommended bit rate. In this case, the network metrics corresponding to the actual bit rate will deviate to some extent from the predicted network metrics corresponding to the recommended network parameters.

[0111] One possible implementation is that the data analysis network element can determine the tolerance range of network parameters based on the expected proportion of network indicators. For example, if the expected proportion in the request message is not less than 95%, the data analysis network element can determine some values ​​for the network parameters. Values ​​greater than or equal to these values ​​can satisfy the following condition: compared with the predicted values ​​of the network indicators corresponding to the recommended network parameters, the predicted values ​​of the corresponding network indicators are controlled within a deviation range of no more than 5%. The data analysis network element uses this value as the boundary value of the tolerance range of the network parameters.

[0112] When the request message does not specify the expected proportion of network indicators, the data analysis network element can also infer the tolerable range of network parameters based on the pre-configured guarantee rate or the guarantee rate indicated by other network elements or devices. For example, if the pre-configured guarantee rate is not less than 90%, the data analysis network element can determine the value of the network parameter whose predicted value deviates from the recommended network parameter value by no more than 10%, and use this value as the boundary value of the tolerable range of the network parameter.

[0113] Furthermore, the data analysis network element can determine the guarantee rate, which is the probability that the actual network metric (or predicted network metric) corresponding to the network parameters within the tolerance range can meet the predicted network metric corresponding to the recommended network parameters. For example, the data analysis network element can determine that the guarantee rate is not less than 90% based on the pre-configuration, and then determine the corresponding network parameter tolerance range based on this guarantee rate. That is, according to the model, when the network parameters are within the tolerance range, there is a 90% probability that the predicted network metric will be achieved. For example, when the actual bit rate is the recommended bit rate of 20 Mbps, the predicted MOS value is greater than or equal to 4.5. When the network parameters are within the tolerance range of 20 ± 1 Mbps, there is a 90% probability that the predicted MOS value will be greater than or equal to 4.5. The analysis request network element can decide whether to strictly adjust according to the recommended network parameters or to relatively loosely control the network parameter values ​​within the tolerance range based on the guarantee rate.

[0114] It should be understood that the above tolerance ranges and guarantee rates are determined for a specific type of network parameters. For example, if the recommended network parameters include a recommended bit rate and a recommended end-to-end delay, then the tolerance range and guarantee rate can be determined for the recommended bit rate, and the tolerance range and guarantee rate can be determined separately for the recommended end-to-end delay.

[0115] One possible implementation is that if there are multiple candidate network parameters and the request message also includes requirement information, the data analysis network element can determine the recommended network parameter from the multiple candidate network parameters based on the requirement information. The requirement information can be found in the description in S101. For example, the data analysis network element can determine the recommended network parameter from the candidate network parameters based on the requirement information. For example, the data analysis network element can determine the maximum or minimum network parameter from the candidate network parameters as the recommended network parameter (e.g., the recommended network parameter is the network parameter with the lowest requirement), or determine the network parameter with the maximum or minimum corresponding predicted network metric from the candidate network parameters as the recommended network parameter (e.g., the recommended network parameter is the network parameter with the optimal corresponding predicted network metric). Furthermore, when the requirement information includes a cost function, the recommended network parameter is determined based on the cost function. If the requirement information requires the recommended network parameter to have the minimum cost function, the data analysis network element can determine the network parameter with the minimum cost function from the candidate network parameters as the recommended network parameter.

[0116] Specifically, the cost function can be an expression for calculating network overhead based on the required recommended network parameters. The data analysis network element can use this expression to determine the network parameter that minimizes the network overhead expression from multiple candidate network parameters as the recommended network parameter. Alternatively, the cost function can be an expression for calculating the service rate based on the required network parameters. The data analysis network element can use this expression to determine the network parameter that minimizes the billing rate from multiple candidate network parameters as the recommended network parameter.

[0117] It should be understood that the alternative network parameters mentioned here can also be replaced with network parameters within the range of network parameters required by the network element requesting the analysis and / or within the range of network parameters corresponding to the network indicators expected by the network element requesting the analysis. In other words, the data analysis network element can determine the recommended network parameters from the range of network parameters required by the network element requesting the analysis and / or within the range of network parameters corresponding to the network indicators expected by the network element requesting the analysis based on the requirement information. The specific method of determining the recommended network parameters based on the requirement information will not be elaborated here. Please refer to the explanation of determining the recommended network parameters from the alternative network parameters based on the requirement information.

[0118] S103: The data analysis network element sends the recommended network parameters to the analysis request network element.

[0119] In one possible implementation, the data analysis network element can send analysis results to the analysis requesting network element. These results may include recommended network parameters. Furthermore, the analysis results may also include at least one of the following: network parameters requested by the analysis requesting network element, network metrics expected by the analysis requesting network element, predicted network metrics corresponding to the recommended network parameters, prediction ratios corresponding to the recommended network parameters, the tolerance range of the network parameters, or the guarantee rate corresponding to the tolerance range of the network parameters.

[0120] Accordingly, the analysis requesting network element receives recommended network parameters (or analysis results containing recommended network parameters) from the data analysis network element. If the analysis requesting network element is a service network element, it can adjust the network parameters according to the recommended network parameters to optimize the network and improve network performance. If the analysis requesting network element is not a service network element, it can send the recommended network parameters to the service network element, or send a network adjustment strategy determined based on the recommended network parameters, enabling the service network element to adjust the network parameters. Taking service experience analysis as an example, the recommended network parameters include the recommended bit rate and / or the recommended end-to-end delay. The analysis requesting network element can be an SMF (Service Provider Function), and the service network element can be a UPF (User Provider Function). After receiving the analysis results, the SMF can determine the adjusted QoS parameters based on the recommended bit rate and / or the recommended end-to-end delay. The adjusted QoS parameters may include the adjusted bit rate and / or the adjusted end-to-end delay. The SMF can also send the adjusted QoS parameters to the UPF, causing the UPF to execute the adjusted QoS parameters.

[0121] As can be seen, the communication method provided in this application embodiment allows the analysis requesting network element to send required network parameters and expected network indicators to the data analysis network element. This enables the data analysis network element to perform its analysis process according to the network parameters and expected network indicators requested by the analysis requesting network element, and obtain analysis results, which may include recommended network parameters. Since the analysis results meet the analysis requesting network element's requirements for network parameters and expectations for network indicators, the analysis service of the data analysis network element can satisfy the requirements of the analysis requesting network element.

[0122] The following describes several implementation methods of S103 based on the relationship between the recommended network parameters and the required network parameter ranges, as well as the relationship between the predicted network metrics corresponding to the recommended network parameters and the expected network metrics ranges.

[0123] Method 1

[0124] The recommended network parameters are within the range of network parameters requested by the network element requesting the analysis, and the predicted network indicators corresponding to the recommended network parameters are within the range of the expected network indicators. In Method 1, the data analysis network element can send the recommended network parameters to the network element requesting the analysis.

[0125] Taking a service experience analysis process where NWDAF acts as the data analysis network element and SMF acts as the analysis request network element as an example, the communication method implemented according to method 1 provided in this application embodiment includes: Figure 5 The following steps are shown:

[0126] S201: SMF sends a request message to NWDAF.

[0127] The request message includes a service experience analysis type identifier, network parameters required by the Service Experience Analyzer (SMF), network metrics expected by the SMF, and the expected percentage of network metrics that meet those expectations. Specifically, the network parameters required by the SMF indicate a required bit rate of less than or equal to 20 Mbps and a required end-to-end latency of greater than or equal to 20 ms. The network metrics expected by the SMF indicate a MOS of not less than 4.5.

[0128] One possible implementation is that the request message may also include requirement information, which can be found in the description in S101.

[0129] Accordingly, NWDAF receives the request message.

[0130] S202: After recognizing the type identifier carried in the request message, NWDAF can determine the analysis result based on the request message, wherein the analysis result includes recommended network parameters. For example, the recommended network parameters include a recommended bit rate and a recommended end-to-end latency, such as a recommended bit rate of 15 Mbps and a recommended end-to-end latency of 30 ms.

[0131] In implementation, NWDAF can use the network metric expected by SMF as the dependent variable of the model, and determine the independent variables of the model based on this dependent variable, which are then used as the network parameters that satisfy the network metric expected by SMF. NWDAF can obtain candidate network parameters by taking the intersection of the network parameters that satisfy the network metric expected by SMF and the network parameters required by SMF. Furthermore, when there are multiple candidate network parameters and the request message includes requirement information, NWDAF can also determine the recommended network parameters from the candidate network parameters based on the requirement information. The implementation method can be referred to the description in S102 when determining the recommended network parameters based on the requirement information.

[0132] Furthermore, the analysis results may include the recommended tolerance range for network parameters, or the recommended tolerance range and guarantee rate. The meaning and determination method of the tolerance range and guarantee rate can be found in the explanation in S102. For example, the analysis results may indicate that the radius of the tolerance range corresponding to the recommended bit rate is 1 Mbps and the corresponding guarantee rate is 90%, and that the radius of the tolerance range corresponding to the recommended end-to-end delay is 5 ms and the corresponding guarantee rate is 95%. It can be assumed that the center value of the tolerance range corresponding to the recommended bit rate is the recommended bit rate, and the center value of the tolerance range corresponding to the recommended end-to-end delay is the recommended end-to-end delay.

[0133] S203: NWDAF sends the analysis results to SMF.

[0134] One possible implementation is that the analysis results may also include the network parameters required by SMF and / or the network metrics expected by SMF.

[0135] Accordingly, the SMF receive analysis results.

[0136] S204: SMF determines the adjusted QoS parameters based on the analysis results.

[0137] S205: SMF sends the adjusted QoS parameters to UPF.

[0138] Accordingly, the UPF receives and executes the adjusted QoS parameters.

[0139] Method 2

[0140] The recommended network parameters are within the range of network parameters requested by the analysis requesting network element, and the predicted network metric corresponding to the recommended network parameters is not within the range of the expected network metric. In Method 2, if the data analysis network element determines that within the required network parameter range, there is no network parameter that can make the predicted network metric fall within the range of the expected network metric, but within the required network parameter range there exists at least one network parameter that makes the corresponding predicted network metric not significantly different from the range of the expected network metric, or in other words, the network parameter makes the corresponding predicted network metric close to the range of the expected network metric, then the data analysis network element can use this at least one network parameter as the recommended network parameter, and the data analysis network element can send the recommended network parameter to the analysis requesting network element in S103. Furthermore, the data analysis network element can also send the predicted network metric corresponding to the recommended network parameter to the analysis requesting network element, which can then decide whether to accept the recommended network parameter based on the predicted network metric. For example, if the analysis requesting network element believes that the predicted network metric corresponding to the recommended network parameter does not meet the requirements, such as being too far from the expected network metric, then it decides not to accept the recommended network parameter; if the analysis requesting network element believes that the predicted network metric corresponding to the recommended network parameter is acceptable, then it can decide to accept the recommended network parameter.

[0141] The statement that the predicted network metric corresponding to the network parameter is not significantly different from the expected network metric means that the distance between the predicted and expected network metric ranges is within a first threshold. One possible implementation is that the first threshold can be determined based on the expected network metric range, for example, by a certain proportion or size of the expected network metric range. Alternatively, the first threshold can be pre-configured in the data analysis network element, or indicated by the analysis request network element, other network elements, or devices.

[0142] In one possible implementation, Method 2, the data analysis network element can also send indication information to the analysis request network element. This first indication information can be used to indicate whether the analysis request network element should accept the recommended network parameters. Specifically, the indication information can be used to indicate that the recommended network parameters do not meet the expected network performance range, or to indicate whether the service request network element should accept the recommended network parameters. For example, the indication information can be an identifier carried in specific bits.

[0143] If the network element requesting the analysis accepts the recommended network parameters, that is, adjusts the network parameters using the recommended network parameters, then the network element requesting the analysis can send a response message corresponding to the instruction message to the data analysis network element, instructing the network element requesting the analysis to accept the recommended network parameters.

[0144] Taking a service experience analysis process where NWDAF is used as the data analysis network element and SMF is used as the analysis request network element as an example, the communication method implemented according to method 2 provided in this application embodiment includes: Figure 6 The following steps are shown:

[0145] S301: The SMF sends a request message to the NWDAF, which is a subscription request. The request message may carry a subscription identifier.

[0146] The request message includes a service experience analysis type identifier, network parameters required by the Service Experience Analyzer (SMF), network metrics expected by the SMF, and the expected percentage of network metrics that meet those expectations. Specifically, the network parameters required by the SMF indicate a bit rate less than or equal to 10 Mbps, and the network metrics expected by the SMF indicate a MOS (Mean Offset Standard) of not less than 4.5.

[0147] One possible implementation is that the request message may also include requirement information, which can be found in the description in S101.

[0148] Accordingly, NWDAF receives the request message.

[0149] S302: NWDAF sends a response message to SMF regarding the subscription request.

[0150] This response message can be used to indicate a successful subscription. The response message for the subscription request may include the subscription identifier from S301.

[0151] Accordingly, SMF receives a response message for the subscription request.

[0152] S303: NWDAF can determine the analysis results based on the request message after recognizing the type identifier carried in the request message. The analysis results include recommended network parameters and the predicted network metric corresponding to the recommended network parameters. If the recommended network parameters do not meet the expected range of the network metric, NWDAF will select the network parameter values ​​within the required range that make the predicted network metric closest to the expected network metric as the recommended network parameters.

[0153] In implementation, NWDAF can use the network parameters required by the SMF (Self-Management Function) as the independent variables of the model, and the network metric expected by the SMF as the dependent variable. It determines that within the required network parameter range, no network parameter can make the predicted dependent variable (i.e., the network metric) fall within the range of the expected network metric. For example, recommended network parameters might include a recommended bit rate of 10 Mbps. This recommended value indicates that within the SMF-required range of less than or equal to 10 Mbps, there is no bit rate that satisfies the expected MOS (Mean Offset of Motion) value greater than or equal to 4.5, but the MOS value is optimal (e.g., equal to 4.3) when the bit rate is equal to 10 Mbps. In this case, NWDAF will use a bit rate of 10 Mbps as the recommended network parameter. Simultaneously, it will use a MOS of 4.3 as the predicted network metric corresponding to the recommended network parameter.

[0154] The method by which NWDAF determines the recommended network parameters from the candidate network parameters can be found in the description in this application, and will not be elaborated here.

[0155] Furthermore, the analysis results may include the recommended tolerance range for network parameters, or the recommended tolerance range and guarantee rate. The meaning and determination method of the tolerance range and guarantee rate can be found in the explanation in S102. For example, the analysis results may indicate that the radius of the tolerance range corresponding to the recommended bit rate is 1 Mbps and the corresponding guarantee rate is 90%. It can be assumed that the center value of the tolerance range corresponding to the recommended bit rate is the recommended bit rate.

[0156] One possible implementation is to include indication information in the analysis results, which indicates whether the SMF should accept the recommended network parameters.

[0157] S304: NWDAF sends the analysis results to SMF. The analysis results may include the subscription identifier from S301.

[0158] One possible implementation is that the analysis results may also include the network parameters required by SMF and / or the network metrics expected by SMF.

[0159] Accordingly, the SMF receive analysis results.

[0160] If the SMF accepts the recommended network parameters, then execute S305; otherwise, if the SMF does not accept the recommended network parameters, then end the process or send a response message to the NWDAF indicating that it refuses to accept the recommended network parameters. The NWDAF can then re-determine the recommended network parameters.

[0161] S305: The SMF sends a response message to the NWDAF, indicating acceptance of the recommended network parameters. The response message may include the subscription identifier from S301.

[0162] Then execute S306-S307. S306 to S307 can be referred to as S204-S205.

[0163] Accordingly, NWDAF receives the response information.

[0164] S308: NWDAF stores the recommendation results. These recommendations include, but are not limited to, the analysis results, and may also include the results from the response information in S305 indicating whether the SMF accepts or does not accept the recommended network parameters. In subsequent iterations, NWDAF can use these stored recommendation results to determine whether the actual network metrics match the predicted network metrics, further train the model, and improve the accuracy of predictions and recommendations. It should be understood that this application does not impose any restrictions on the execution timing of S308 and S306.

[0165] Figure 6 The process shown is the same as Figure 5 The main difference in the illustrated process is that the recommended network parameters determined by NWDAF are within the range of network parameters required by the requesting network element, and the predicted network indicators corresponding to the recommended network parameters are not within the range of the expected network indicators. Therefore, the requesting network element can decide whether to accept the recommended network parameters based on the predicted network indicators corresponding to the recommended network parameters. If the recommended network parameters are accepted, steps S305 to S308 can be executed to adjust the network parameters according to the recommended network parameters.

[0166] Method 3

[0167] The recommended network parameters are not within the range of network parameters required by the analysis requesting network element, and the predicted network metric corresponding to the recommended network parameters is within the range of the expected network metric. In method 3, if the data analysis network element determines that no network parameter can make the predicted network metric within the range of the expected network metric, but at least one network parameter outside the required network parameter range can make the corresponding predicted network metric within the range of the expected network metric, and if the difference between the at least one network parameter and the required network parameter range is not significant, then the data analysis network element can use the at least one network parameter as the recommended network parameter. The data analysis network element can send the recommended network parameter and the predicted network metric corresponding to the recommended network parameter to the analysis requesting network element in S103. Here, the predicted network metric corresponding to the recommended network parameter is not within the range of the expected network metric.

[0168] The statement that the network parameter is not significantly different from the required network parameter range means that the distance between the network parameter and the required network parameter range is within a second threshold. One possible implementation is that the second threshold can be determined based on the required network parameter range, for example, by a certain proportion or size of the required network parameter range. Alternatively, the second threshold can be pre-configured in the data analysis network element, or indicated by the analysis request network element, other network elements, or devices.

[0169] In one possible implementation, method 3, the data analysis network element can also send indication information to the analysis request network element. This first indication information can be used to indicate whether the analysis request network element decides to accept the recommended network parameters. Specifically, the indication information can be used to indicate the range of network parameters that the recommended network parameters cannot meet the requirements, or to indicate whether the service request network element decides to accept the recommended network parameters. For example, the indication information can be an identifier carried in specific bits.

[0170] If the network element requesting the analysis accepts the recommended network parameters, that is, adjusts the network parameters using the recommended network parameters, then the network element requesting the analysis can send a response message corresponding to the instruction message to the data analysis network element, instructing the network element requesting the analysis to accept the recommended network parameters.

[0171] Taking a service experience analysis process where NWDAF is used as the data analysis network element and SMF is used as the analysis request network element as an example, the communication method implemented according to method 3 provided in this application embodiment includes: Figure 7 The following steps are shown:

[0172] S401: The SMF sends a request message to the NWDAF, which is a subscription request. The request message may carry a subscription identifier.

[0173] The request message includes a service experience analysis type identifier, network parameters required by the Service Experience Analyzer (SMF), network metrics expected by the SMF, and the expected percentage of network metrics that meet those expectations. Specifically, the network parameters required by the SMF indicate a bit rate less than or equal to 10 Mbps, and the network metrics expected by the SMF indicate a MOS (Mean Offset Standard) of not less than 4.5.

[0174] One possible implementation is that the request message may also include requirement information, which can be found in the description in S101.

[0175] Accordingly, NWDAF receives the request message.

[0176] S402: NWDAF sends a response message to SMF regarding the subscription request.

[0177] This response message can be used to indicate a successful subscription. The response message for the subscription request may include the subscription identifier from S401.

[0178] Accordingly, SMF receives a response message for the subscription request.

[0179] S403: After recognizing the type identifier carried in the request message, NWDAF can determine the analysis result based on the request message. The analysis result includes recommended network parameters, and these recommended network parameters are not within the range of the required network parameters. NWDAF can determine the network parameters corresponding to the predicted network metric within the range of the expected network metric, and select the network parameters whose range is not significantly different from the required network parameter range as the recommended network parameters.

[0180] In implementation, NWDAF can use the network metric expected by the SMF as the dependent variable of the model and the network parameters required by the SMF as the independent variables of the model. It determines that within the range of required network parameters, no network parameter can make the predicted dependent variable (i.e., the network metric) fall within the range of the expected network metric. For example, recommended network parameters might include a recommended bit rate of 10 Mbps. This recommended value means that within the range of 10 Mbps or less required by the SMF, there is no bit rate that satisfies the expected MOS value greater than or equal to 4.5. However, the MOS is 4.5 at a bit rate of 12 Mbps. Therefore, a bit rate of 12 Mbps satisfies the MOS value greater than or equal to 4.5. In this case, NWDAF will use a bit rate of 12 Mbps as the recommended network parameter.

[0181] Furthermore, the analysis results may include the recommended tolerance range for network parameters, or the recommended tolerance range and guarantee rate. The meaning and determination method of the tolerance range and guarantee rate can be found in the explanation in S102. For example, the analysis results may indicate that the radius of the tolerance range corresponding to the recommended bit rate is 1 Mbps and the corresponding guarantee rate is 90%. It can be assumed that the center value of the tolerance range corresponding to the recommended bit rate is the recommended bit rate.

[0182] One possible implementation is to include indication information in the analysis results, which indicates whether the SMF should accept the recommended network parameters.

[0183] S404: NWDAF sends the analysis results to SMF. The analysis results may include the subscription identifier from S401.

[0184] One possible implementation is that the analysis results may also include the network parameters required by SMF and / or the network metrics expected by SMF.

[0185] Accordingly, the SMF receive analysis results.

[0186] If the SMF accepts the recommended network parameters, then execute S405; otherwise, if the SMF does not accept the recommended network parameters, then end the process or send a response message to the NWDAF indicating that it refuses to accept the recommended network parameters. The NWDAF can then re-determine the recommended network parameters.

[0187] S405: The SMF can send a response message to the NWDAF, indicating acceptance of the recommended network parameters. The response message may include the subscription identifier from S401. Then, S406-S407 are executed; S406 to S407 are similar to S204-S205.

[0188] Accordingly, NWDAF receives the response information.

[0189] S408: NWDAF stores the results of this recommendation.

[0190] For an example of how to implement S408, please refer to the description of S306.

[0191] Figure 7 The process shown is the same as Figure 6 The main difference in the illustrated process is that: if the recommended network parameters determined by NWDAF are not within the range of network parameters required by SMF, and the predicted network metric corresponding to the recommended network parameters is within the range of the expected network metric, then the analysis result carries the recommended network parameters, and the requesting network element decides whether to accept the recommended network parameters. If the recommended network parameters are accepted, steps S405 to S408 can be continued to adjust the network parameters according to the recommended network parameters.

[0192] Figure 8 and Figure 9 The diagram illustrates the possible communication devices provided in the embodiments of this application. These communication devices can be used to implement the functions of the data analysis network element or analysis request network element in the above method embodiments, and thus can also achieve the beneficial effects of the above method embodiments. In the embodiments of this application, the communication device can be a data analysis network element or analysis request network element, or it can be a module (such as a chip) applied to the data analysis network element or analysis request network element.

[0193] like Figure 8 As shown, the communication device 800 includes a processing unit 810 and a transceiver unit 820. The communication device 800 is used to implement the functions of the data analysis network element or the analysis request network element in the above method embodiments.

[0194] In the first embodiment, the communication device is used to implement the function of the data analysis network element in the above method embodiment. The transceiver unit 820 can be used to receive a request message from the analysis requesting network element. The request message is used to request recommended network parameters, and the request message includes the network parameters required by the analysis requesting network element and the network indicators expected by the analysis requesting network element. The processing unit 810 can be used to determine the recommended network parameters based on the required network parameters and the expected network indicators. The transceiver unit 820 can also be used to send the recommended network parameters to the analysis requesting network element.

[0195] As one possible implementation, the recommended network parameters are within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is not within the range of the desired network metric.

[0196] As one possible implementation, the transceiver unit 820 can also be used to send the predicted network indicators corresponding to the recommended network parameters to the analysis request network element.

[0197] As one possible implementation, the recommended network parameters are not within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is within the range of the desired network metric.

[0198] As one possible implementation, the transceiver unit 820 can also be used to send indication information to the analysis requesting network element, the indication information being used by the analysis requesting network element to decide whether to accept the recommended network parameters.

[0199] As one possible implementation, the recommended network parameters are within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is within the range of the desired network metric.

[0200] As one possible implementation, the transceiver unit 820 can also be used to send the tolerance range of the network parameters to the analysis request network element, wherein the recommended network parameters fall within the tolerance range.

[0201] As one possible implementation, the transceiver unit 820 can also be used to send a guarantee rate to the analysis request network element, wherein the guarantee rate is the probability that the predicted network indicators corresponding to the network parameters within the tolerance range can meet the predicted network indicators corresponding to the recommended network parameters.

[0202] As one possible implementation, the request message further includes requirement information indicating the recommended network parameters. The data analysis network element can also determine the recommended network parameters based on the requirement information, from the range of the required network parameters and / or the range of network parameters corresponding to the desired network metric. For example, the requirement information includes a cost function. Alternatively, the requirement information may indicate that the recommended network parameters are the maximum or minimum value within the range of the required network parameters, or that the requirement information indicates that the recommended network parameters satisfy the maximum or minimum value within the range of network parameters corresponding to the desired network metric.

[0203] In the second embodiment, the communication device is used to implement the function of the analysis request network element in the above method embodiment. The transceiver unit 820 can be used to send a request message to the data analysis network element. The request message requests recommended network parameters, including the network parameters required by the analysis request network element and the network indicators expected by the analysis request network element. The transceiver unit 820 can also be used to receive recommended network parameters from the data analysis network element, the recommended network parameters being determined based on the required network parameters and the network indicators expected by the analysis request network element. The processing unit 810 can be used to adjust the network parameters according to the recommended network parameters.

[0204] In one possible design, the recommended network parameters are within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is not within the range of the desired network metric.

[0205] In one possible design, the recommended network parameters are not within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is within the range of the desired network metric.

[0206] In one possible design, the transceiver unit 820 can also be used to receive indication information from the data analysis network element; the processing unit 810 can also be used to determine whether to accept the recommended network parameters based on the indication information.

[0207] In one possible design, the recommended network parameters are within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is not within the range of the desired network metric. The transceiver unit 820 can also be used to receive the predicted network metric corresponding to the recommended network parameters from the data analysis network element. The processing unit 810 can determine whether to accept the recommended network parameters based on the predicted network metric corresponding to the recommended network parameters and the indication information.

[0208] In one possible design, the recommended network parameters are within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is within the range of the desired network metric.

[0209] In one possible design, the transceiver unit 820 can also be used to receive the tolerance range of the network parameters from the data analysis network element, wherein the recommended network parameters fall within the tolerance range. Specifically, the processing unit 810 can be used to adjust the network parameters within the tolerance range.

[0210] In one possible design, the transceiver unit 820 can also be used to receive the tolerance range of the network parameters from the data analysis network element, wherein the recommended network parameters fall within the tolerance range. The transceiver unit 820 can also be used to receive a guarantee rate from the data analysis network element, where the guarantee rate is the probability that the predicted network metric corresponding to the network parameters within the tolerance range can satisfy the predicted network metric corresponding to the recommended network parameters. Specifically, the processing unit 810 can be used to determine whether to adjust the network parameters within the tolerance range based on the guarantee rate.

[0211] In one possible design, the request message may further include requirement information indicating the recommended network parameters. For example, the requirement information may include a cost function. Alternatively, the requirement information may indicate that the recommended network parameters are the maximum or minimum value within a range of required network parameters, or that the requirement information indicates that the recommended network parameters are the maximum or minimum value within a range of network parameters that satisfy the desired network metric.

[0212] A more detailed description of the processing unit 810 and the transceiver unit 820 can be obtained directly from the relevant descriptions in the above method embodiments, and will not be repeated here.

[0213] like Figure 9 As shown, the communication device 900 includes a processor 910. As one implementation, the communication device 900 also includes an interface circuit 920, which is coupled to the processor 910. It is understood that the interface circuit 920 can be a transceiver or an input / output interface. As another implementation, the communication device 900 may also include a memory 930 for storing instructions executed by the processor 910, or storing input data required by the processor 910 to execute instructions, or storing data generated after the processor 910 executes instructions.

[0214] When the communication device 900 is used to implement the above method embodiment, the processor 910 is used to implement the function of the processing unit 810, and the interface circuit 920 is used to implement the function of the transceiver unit 820.

[0215] It is understood that the processor in the embodiments of this application may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor may be a microprocessor or any conventional processor.

[0216] The method steps in the embodiments of this application can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disks, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Alternatively, the ASIC can reside in a base station or terminal. Of course, the processor and storage medium can also exist as discrete components in the base station or terminal.

[0217] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of this application are performed entirely or partially. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a base station, a user equipment, or other programmable device. The computer program or instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; it can also be an optical medium, such as a digital video optical disc; or it can be a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or non-volatile storage medium, or may include both types of storage media.

[0218] In the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of different embodiments are consistent and can be referenced by each other. The technical features of different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0219] In this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. In the textual description of this application, the character " / " generally indicates an "or" relationship between the preceding and following related objects; in the formulas of this application, the character " / " indicates a "division" relationship between the preceding and following related objects.

[0220] It is understood that the various numerical designations used in the embodiments of this application are merely for descriptive convenience and are not intended to limit the scope of the embodiments of this application. The order of the process numbers described above does not imply the order of execution; the execution order of each process should be determined by its function and internal logic.

Claims

1. A communication method, characterized in that, include: The data analysis network element receives a request message from the analysis request network element. The request message is used to request recommended network parameters. The request message includes the network parameters required by the analysis request network element and / or the network indicators expected by the analysis request network element. The required network parameters include the type of the required network parameters, or the required network parameters include the type and value of the required network parameters. The expected network indicators include average subjective evaluation or service experience quality. The data analysis network element determines the recommended network parameters based on the required network parameters, wherein the recommended network parameters are within the range of the required network parameters; and / or, The data analysis network element determines the recommended network parameters based on the expected network indicators, and the predicted network indicators corresponding to the recommended network parameters are within the range of the expected network indicators. The data analysis network element sends the recommended network parameters to the analysis request network element.

2. The method as described in claim 1, characterized in that, The recommended network parameters are within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is not within the range of the expected network metric.

3. The method as described in claim 2, characterized in that, The method further includes: The data analysis network element sends the predicted network metrics corresponding to the recommended network parameters to the analysis request network element.

4. The method as described in claim 1, characterized in that, The recommended network parameters are not within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is within the range of the expected network metric.

5. The method as described in any one of claims 2-4, characterized in that, The method further includes: The data analysis network element sends an instruction to the analysis request network element, the instruction being used by the analysis request network element to decide whether to accept the recommended network parameters.

6. The method as described in claim 1, characterized in that, The method further includes: The data analysis network element sends the tolerance range of the network parameters to the analysis request network element, and the recommended network parameters fall within the tolerance range.

7. The method as described in claim 6, characterized in that, The method further includes: The data analysis network element sends a guarantee rate to the analysis request network element. The guarantee rate is the probability that the predicted network indicators corresponding to the network parameters within the tolerance range can meet the predicted network indicators corresponding to the recommended network parameters.

8. The method according to any one of claims 1-4 and 6-7, characterized in that, The request message also includes requirement information for indicating the recommended network parameters. The data analysis network element determines the recommended network parameters based on the required network parameters and the desired network indicators, including: The data analysis network element determines the recommended network parameters based on the required information, from the range of the required network parameters and / or from the range of the network parameters corresponding to the desired network indicators.

9. The method as described in claim 8, characterized in that, The required information includes a cost function.

10. The method as described in claim 8, characterized in that, The requirement information indicates that the recommended network parameters are the maximum or minimum values ​​within the range of the required network parameters, or that the requirement information indicates that the recommended network parameters are the maximum or minimum values ​​within the range of the network parameters that satisfy the desired network metric.

11. The method according to any one of claims 1-4, 6-7, and 9-10, characterized in that, The required network parameter is Quality of Service (QoS).

12. The method according to any one of claims 1-4, 6-7, and 9-10, characterized in that, The required network parameters also include at least one of the following: time, location of user equipment (UE), location of application, bit rate of service flow, packet delay, and number of transmitted and retransmitted messages.

13. A communication method, characterized in that, include: The analysis requesting network element sends a request message to the data analysis network element. The request message is used to request recommended network parameters. The request message includes the network parameters required by the analysis requesting network element and / or the network indicators expected by the analysis requesting network element. The required network parameters include the type of the required network parameters, or the required network parameters include the type and value of the required network parameters. The expected network indicators include average subjective evaluation or service experience quality. The analysis request network element receives recommended network parameters from the data analysis network element. These recommended network parameters are determined based on the required network parameters and are within the range of the required network parameters; and / or, The recommended network parameters are determined based on the network metrics expected by the network element requesting the analysis, and the predicted network metrics corresponding to the recommended network parameters are within the range of the expected network metrics. The analysis request network element adjusts the network parameters according to the recommended network parameters.

14. The method as described in claim 13, characterized in that, The recommended network parameters are within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is not within the range of the expected network metric.

15. The method as described in claim 13, characterized in that, The recommended network parameters are not within the range of the required network parameters, and the predicted network metric corresponding to the recommended network parameters is within the range of the expected network metric.

16. The method as described in claim 14 or 15, characterized in that, The method further includes: The analysis request network element receives instruction information from the data analysis network element; The analysis requesting network element determines whether to accept the recommended network parameters based on the indication information.

17. The method as described in claim 16, characterized in that, The recommended network parameters are within the required network parameters, and the predicted network metric corresponding to the recommended network parameters is not within the range of the desired network metric. The method further includes: The analysis request network element receives the predicted network metrics corresponding to the recommended network parameters from the data analysis network element. The analysis requesting network element determines whether to accept the recommended network parameters based on the indication information, including: The analysis requesting network element determines whether to accept the recommended network parameters based on the predicted network indicators corresponding to the recommended network parameters and the indication information.

18. The method as described in claim 13, characterized in that, The method further includes: The analysis request network element receives the tolerance range of the network parameters from the data analysis network element, and the recommended network parameters belong to the tolerance range; The analysis request network element adjusts network parameters according to the recommended network parameters, including: The analysis requests the network element to adjust the network parameters within the tolerance range.

19. The method as described in claim 18, characterized in that, The method further includes: The analysis request network element receives the tolerance range of the network parameters from the data analysis network element, and the recommended network parameters belong to the tolerance range; The analysis request network element receives a guarantee rate from the data analysis network element, whereby the guarantee rate is the probability that the predicted network indicators corresponding to the network parameters within the tolerance range can meet the predicted network indicators corresponding to the recommended network parameters. The analysis request network element adjusts network parameters according to the recommended network parameters, including: The analysis request network element determines whether to adjust the network parameters within the tolerance range based on the guarantee rate.

20. The method according to any one of claims 13-15 and 17-19, characterized in that, The request message also includes requirement information for indicating the recommended network parameters.

21. The method as described in claim 20, characterized in that, The required information includes a cost function.

22. The method as described in claim 20, characterized in that, The requirement information indicates that the recommended network parameters are the maximum or minimum values ​​within the range of the required network parameters, or that the requirement information indicates that the recommended network parameters are the maximum or minimum values ​​within the range of the network parameters that satisfy the desired network metric.

23. The method according to any one of claims 13-15, 17-19, and 21-22, characterized in that, The required network parameter is Quality of Service (QoS).

24. The method according to any one of claims 13-15, 17-19, and 21-22, characterized in that, The required network parameters also include at least one of the following: time, location of user equipment (UE), location of application, bit rate of service flow, packet delay, and number of transmitted and retransmitted messages.

25. A communication device, characterized in that, It includes a processor and a memory; the memory is used to store computer instructions, and the processor executes the computer instructions stored in the memory to cause the apparatus to perform the method according to any one of claims 1 to 24.

26. A communication system, characterized in that, include: A data analysis network element, used to perform the method described in any one of claims 1 to 12; as well as The analysis request network element is used to receive recommended network parameters from the data analysis network element and adjust the network parameters according to the recommended network parameters.

27. A computer-readable storage medium, characterized in that, The storage medium stores a computer program or instructions, which, when executed by a communication device, implement the method as described in any one of claims 1 to 24.

Images