User experience guarantee method and device and medium
By combining wireless network element resource usage data and user history data with the 5QI adaptive configuration module, the network configuration of live streaming users and background users is identified and optimized, solving the problem of insufficient real-time identification of live streaming users and improving user experience and operator revenue.
Patent Information
- Application Number
- CN202411131965.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-16
- Publication Date
- 2024-11-22
AI Technical Summary
In existing technologies, the real-time performance of live streaming user identification is insufficient, making it difficult to accurately identify live streaming users and failing to coordinate and guarantee the service experience of live streaming users and background users, thus affecting user experience and operator revenue.
By using wireless network element resource usage data and user history data, suspected live streaming users can be identified. Combined with the 5QI adaptive configuration module, network resources are dynamically adjusted to identify suppressed cells and optimize network configuration to ensure user experience.
It improves the real-time performance of live user identification, coordinates the protection of business performance for both live users and background users, enhances user experience, and increases operator revenue.
Smart Images

Figure CN119012287A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure at least relates to the field of communication technology, and in particular to a user experience assurance method, a user experience assurance device, and a computer-readable storage medium. Background Art
[0002] The wireless scenarios in which live broadcast users are located are complex and changeable, and their service needs are diverse. How to efficiently and accurately improve network performance and ensure the service quality of live broadcast users based on the distribution of live broadcast users in the community, changes in traffic distribution, and the status of the community is a common challenge faced by operators and equipment manufacturers.
[0003] The current solutions for ensuring the service experience of live broadcast users have the following shortcomings:
[0004] 1) The real-time identification of live broadcast users is insufficient. It is possible to identify whether a user is a live broadcast user based on the user's historical record data, but using historical data makes it difficult to identify live broadcast users in real time;
[0005] 2) The service perception guarantee needs of background users and live broadcast users are not considered as a whole, resulting in the traffic demand of background users and / or live broadcast users in the cell being suppressed, which affects both user experience and operator revenue. Summary of the invention
[0006] The technical problem to be solved by the present disclosure is to provide a user experience guarantee method, a user experience guarantee device and a computer-readable storage medium in view of the above-mentioned deficiencies, so as to solve the problem of how to improve the real-time performance of live user identification and collaboratively guarantee the user experience of live users and non-live users.
[0007] In a first aspect, the present disclosure provides a method for ensuring user experience, the method comprising:
[0008] According to the current resource occupancy data on the wireless network element side, the current suspected uplink live broadcast user is identified;
[0009] Identify the current live streaming user based on the historical record data of the current suspected live streaming user;
[0010] Identify whether the cell where the current uplink live broadcast user is located is a suppressed cell. A suppressed cell refers to a cell where some users have given up some network services because the network performance does not meet their needs.
[0011] Adjust the network configuration of users in the depressed cells to ensure the user experience of the depressed cells.
[0012] Furthermore, according to the current wireless network element side resource occupancy data, the current suspected uplink live broadcast user is identified, specifically including:
[0013] Acquire and differentiate the current performance indicators of a wireless network element by user, so as to obtain the connection duration of a user's current service connected to a wireless network element and the forward throughput, reverse throughput, forward to reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate and reverse effective rate provided by the wireless network element for the user's current service during the connection duration;
[0014] Determine whether a user is a currently suspected uplink live broadcast user based on the user's connection duration, forward throughput, reverse throughput, forward to reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate, and reverse effective rate.
[0015] Furthermore, according to the connection duration, forward throughput, reverse throughput, forward-reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate and reverse effective rate of a certain user, it is determined whether a certain user is a current suspected uplink live broadcast user, specifically including:
[0016] Obtaining a plurality of preset user service type determination models, each user service type determination model including connection duration, forward throughput, reverse throughput, forward to reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate and reverse effective rate, at least three of which should be greater than a determination threshold;
[0017] In response to the connection duration, forward throughput, reverse throughput, forward-reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate and reverse effective rate of a certain user satisfying a certain user service type determination model, the type of service currently in use by the certain user is determined, and the certain user is determined to be a current suspected uplink live broadcast user.
[0018] Further, according to the historical record data of the current suspected uplink live broadcast user, the current uplink live broadcast user is identified, specifically including:
[0019] Get the historical XDR data of the current suspected uplink live broadcast user;
[0020] Obtain the historical live broadcast data of the current suspected uplink live broadcast user based on the historical XDR data;
[0021] Determine the historical live broadcast type of the current suspected uplink live broadcast user based on historical live broadcast data;
[0022] In response to the type of the currently used service of the current suspected uplink live broadcast user belonging to its historical live broadcast type, the current suspected uplink live broadcast user is determined to be the current uplink live broadcast user.
[0023] Among them, XDR is extended detail record.
[0024] Further, identifying whether the cell where the current uplink live broadcast user is located is a suppressed cell specifically includes:
[0025] Identify the cell where the current uplink live broadcast user is located;
[0026] According to the historical performance indicators of the cell, the ideal traffic and / or ideal rate of each user in the cell where the current uplink live broadcast user is located is obtained;
[0027] According to the current performance indicators of the cell, obtain the current traffic and / or current rate of each user in the cell where the current uplink live broadcast user is located;
[0028] In response to a user in the cell having a current flow and / or a current rate that is less than the user's ideal flow and / or ideal rate, the cell where the current uplink live broadcast user is located is identified as a suppressed cell.
[0029] Further, identifying whether the cell where the current uplink live broadcast user is located is a suppressed cell specifically includes:
[0030] Identify current background users in the cell where the current uplink live broadcast user is located and other than the current uplink live broadcast user;
[0031] Obtaining a first ideal flow rate and / or a first ideal rate of the current background user according to a first historical flow rate and / or a first historical rate of the current background user when the historical time period has the same characteristics as the current time period and the current uplink live broadcast user has not performed any historical live broadcast behavior;
[0032] Obtaining a first current flow and / or a first current rate of a current background user in a current period;
[0033] In response to the first current traffic and / or the first current rate being less than the first ideal traffic and / or the first ideal rate, it is determined that the cell where the current uplink live broadcast user is located is a suppressed cell.
[0034] Further, identifying whether the cell where the current uplink live broadcast user is located is a suppressed cell specifically includes:
[0035] Acquire a second ideal traffic volume and / or a second ideal rate of the current uplink live broadcast user according to the second historical traffic volume and / or the second historical rate of multiple historical live broadcast behaviors of the current uplink live broadcast user;
[0036] Obtain the second current flow and / or the second current rate of the current uplink live broadcast user in the current period;
[0037] In response to the second current traffic and / or the second current rate being less than the second ideal traffic and / or the second ideal rate, it is determined that the cell where the current uplink live broadcast user is located is a suppressed cell.
[0038] Furthermore, the network configuration of users in the depressed cell is adjusted to ensure the user experience of the depressed cell, including:
[0039] Based on the initialization settings of the iterative algorithm model, the following iterative operations are performed:
[0040] Obtain the 5QI configuration parameters of this iteration for each user in the suppressed cell;
[0041] Based on the 5QI configuration parameters and cell capacity of this iteration, obtain the predicted traffic and / or predicted rate of each user in this iteration;
[0042] According to the difference between the predicted flow and / or predicted rate of this iteration and the ideal flow and / or ideal rate, adjust the 5QI configuration parameters of this iteration to obtain the 5QI configuration parameters of the next iteration;
[0043] Repeat the above iterative operation until a preset end condition is reached, wherein the preset end condition includes that the difference between the predicted flow rate and / or predicted rate and the ideal flow rate and / or ideal rate is less than a preset difference threshold;
[0044] Output the 5QI configuration parameters corresponding to the time when the difference between the predicted flow rate and / or predicted rate and the ideal flow rate and / or ideal rate is the smallest;
[0045] Among them, the initialization settings include setting the initial 5QI configuration parameters, and 5QI is the 5G service quality indicator.
[0046] In a second aspect, the present disclosure provides a user experience guarantee device, the device comprising:
[0047] The first identification module is used to identify the current suspected uplink live broadcast user according to the current wireless network element side resource occupancy data;
[0048] A second identification module, connected to the first identification module, is used to identify the current uplink live broadcast user according to the historical record data of the current suspected uplink live broadcast user;
[0049] The third identification module is connected to the second identification module and is used to identify whether the cell where the current uplink live broadcast user is located is a suppressed cell. A suppressed cell refers to a cell in which some users have given up some network services because the network performance does not meet their own needs;
[0050] The adjustment module is connected to the third identification module and is used to adjust the network configuration of the users in the suppressed cell to ensure the user experience of the suppressed cell.
[0051] In a third aspect, the present disclosure provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the user experience assurance method as described above is implemented.
[0052] The present disclosure provides a user experience guarantee method, a user experience guarantee device and a computer-readable storage medium, which identify uplink live broadcast users through current wireless network element side resource occupancy data combined with user historical record data, thereby improving the real-time performance of live broadcast user identification. For cells with uplink live broadcast users, it is determined whether there is suppressed user service demand in the cell, and the network configuration of the user in the suppressed cell is adjusted, which helps to release the suppressed user demand and coordinately guarantee the service performance of live broadcast users and non-live broadcast users in the cell, thereby improving user experience and increasing operator revenue. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] Figure 1 is a flow chart of a method for ensuring user experience according to an embodiment of the present disclosure;
[0054] Figure 2 is a flow chart of a method for identifying an uplink live broadcast user according to an embodiment of the present disclosure;
[0055] Figure 3 is a flow chart of a method for configuring a suppressed cell network according to an embodiment of the present disclosure;
[0056] Figure 4 It is a structural schematic diagram of a user experience assurance device according to an embodiment of the present disclosure. DETAILED DESCRIPTION
[0057] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the embodiments of the present disclosure will be further described in detail below with reference to the accompanying drawings.
[0058] It should be understood that the specific embodiments and drawings described herein are only used to explain the present disclosure rather than to limit the present disclosure.
[0059] It can be understood that, in the absence of conflict, the various embodiments of the present disclosure and the various features in the embodiments can be combined with each other.
[0060] It can be understood that, for the convenience of description, the drawings of the present disclosure only show parts related to the present disclosure, while parts irrelevant to the present disclosure are not shown in the drawings.
[0061] It can be understood that each unit or unit involved in the embodiments of the present disclosure may correspond to only one physical structure, or may be composed of multiple physical structures, or multiple units or units may be integrated into one physical structure.
[0062] It will be understood that, in the absence of conflict, the functions and steps marked in the flowcharts and block diagrams of the present disclosure may occur in an order different from that marked in the drawings.
[0063] It is understood that the flowcharts and block diagrams of the present disclosure illustrate the possible architectures, functions, and operations of the systems, devices, equipment, and methods according to the embodiments of the present disclosure. Each box in the flowchart or block diagram may represent a unit, unit, program segment, or code, which contains executable instructions for implementing the specified functions. Moreover, each box or combination of boxes in the block diagram and flowchart may be implemented by a hardware-based system that implements the specified functions, or may be implemented by a combination of hardware and computer instructions.
[0064] It can be understood that the units and units involved in the embodiments of the present disclosure can be implemented by software or hardware, for example, the units and units can be located in a processor.
[0065] Embodiment 1:
[0066] like Figure 1 As shown, the present disclosure provides a method for ensuring user experience, the method comprising:
[0067] S1. Identify the current suspected uplink live broadcast user based on the current wireless network element side resource occupancy data;
[0068] S2. Identify the current uplink live broadcast user according to the historical record data of the current suspected uplink live broadcast user;
[0069] S3, identifying whether the cell where the current uplink live broadcast user is located is a suppressed cell. A suppressed cell refers to a cell in which some users have given up some network services because the network performance does not meet their needs;
[0070] S4. Adjust the network configuration of users in the depressed cell to ensure the user experience of the depressed cell.
[0071] In this embodiment, the uplink live broadcast users are identified by combining the current wireless network element side resource occupancy data with the user's historical record data, thereby improving the real-time performance of live broadcast user identification. For cells with uplink live broadcast users, it is determined whether there is suppressed user service demand in the cell, and the user's network configuration is adjusted in the suppressed cell, which helps to release the user's suppressed demand and coordinately guarantee the service performance of live broadcast users and non-live broadcast users in the cell, thereby improving user experience and increasing operator revenue.
[0072] Specifically, this embodiment is a collaborative guarantee method for background user and live broadcast user experience based on 5QI (5G QoS Identifier, 5G (fifth generation mobile communication technology, 5th Generation Mobile Communication Technology) service quality indicator) adaptive configuration. It mainly solves the problem of real-time identification of live broadcast users and collaborative guarantee of service experience of live broadcast users and background users.
[0073] 1) Real-time identification of live broadcast users: XDR (eXtended Detail Record) data can accurately determine the service type of live broadcast users. However, due to the large amount of XDR data and untimely data processing, it is impossible to determine the specific service type of the current user in real time, and thus it is impossible to provide timely and effective network security measures for live broadcast users.
[0074] This embodiment can fully describe the resource occupancy characteristics based on the performance data of the wireless side, and determines the service type of the application in real time through the wireless network data. Then, the service subclass and characteristic information of the live broadcast service are further derived through the historical XDR data and the resource occupancy characteristics of the service, wherein the characteristic information includes public information, protocol-specific information, and specific service information. Public information refers to the field information that must be recorded when a user accesses any Internet service. Protocol-specific information is defined for common protocols, including HTTP (Hypertext Transfer Protocol), DNS (Domain Name System), etc. Various protocol call records are mutually exclusive, that is, the same traffic belongs to only one specific protocol information call record, and HTTP has the highest priority. Specific service information is defined for services of special concern, and is filled in at the granularity of a service request / login, and multiple related data streams need to be synthesized. Specific service information repeatedly counts the traffic of a single request xDR, that is, if a data stream is HTTP video, HTTP xDR and Video xDR are generated at the same time.
[0075] It should be noted that XDR data is one type of user history record data. If other types of history record data can determine whether the user has performed live broadcast services in the past and the type of live broadcast services, it is feasible to identify live broadcast users by combining this data with the current wireless network element side data, which is within the scope of protection requested in this application.
[0076] 2) Collaboratively guarantee the service experience of live broadcast users and background users: In the 5G wireless communication system, as the number of users in the cell increases, the total traffic in the cell gradually increases. However, when the number of users reaches a certain value, the actual traffic no longer increases due to the limited capacity of the cell. As the number of users increases further, the perceived rate of a single user will decrease, which may cause users to abandon certain traffic services, resulting in traffic suppression. Traffic grows nonlinearly with the number of users, which limits the operator's revenue to a certain extent, so it is necessary to release the suppressed traffic. Taking into account the requirements of live broadcast users and live broadcast services for network performance, a single solution for live broadcast protection of live broadcast users by scheduling network resources lacks flexibility. At the same time, when the network capacity cannot meet the live broadcast needs, the live broadcast perception of background users cannot be guaranteed, resulting in the risk of abnormal experience perception of both background users and live broadcast users.
[0077] This embodiment is based on the collaborative guarantee method of background users and live broadcast users' user experience guarantee with adaptive configuration of 5QI (5G QoS (Quality of Service) level quantitative index), considers the service perception guarantee needs of background users and live broadcast users as a whole, obtains the network user characteristics of live broadcast users without increasing network nodes, and provides the user big data information in the network to the 5QI adaptive configuration module. The 5QI adaptive configuration module makes decisions based on the network's traffic model (service characteristics, user distribution) and traffic load forecast information to dynamically determine what kind of service capabilities the network provides for background users and live broadcast users, what kind of 5QI parameters to configure, and realizes dynamic 5QI adaptive configuration management, so as to achieve the purpose of improving user experience and reducing operating costs.
[0078] It should be noted that 5QI configuration is a means of ensuring user network performance for 5G networks. In other types of networks, the corresponding network configuration parameters can be adaptively selected according to the method provided in this application.
[0079] In one implementation, S1, identifying a current suspected uplink live broadcast user according to current wireless network element side resource occupancy data, specifically includes:
[0080] Acquire and differentiate the current performance indicators of a wireless network element by user, so as to obtain the connection duration of a user's current service connected to a wireless network element and the forward throughput, reverse throughput, forward to reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate and reverse effective rate provided by the wireless network element for the user's current service during the connection duration;
[0081] Determine whether a user is a currently suspected uplink live broadcast user based on the user's connection duration, forward throughput, reverse throughput, forward to reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate, and reverse effective rate.
[0082] In this embodiment, steps S1 and S2 are mainly for real-time and accurate identification of live broadcast users. Figure 2 As shown in the figure, the process includes: collecting performance indicators related to resource occupancy of wireless network elements; judging the service type by combining the data collected by wireless network elements with the service judgment model; and identifying the uplink live broadcast users in the network by combining the user's historical CDR data and the user portrait of the service classification model. That is, firstly, based on the resource occupancy data on the wireless network element side, the user service type that meets the specific service judgment model is identified, and these users are regarded as suspected live broadcast users, and then based on the historical live broadcast behavior of these users, they are determined to be live broadcast users.
[0083] Specifically, the wireless network element side can collect the following resource occupancy information:
[0084] Air interface connection duration: the duration between the establishment of the air interface connection and the expiration time, corresponding to the application connection duration;
[0085] Physical layer throughput: forward / reverse throughput of the physical layer, which can be connected to users, network elements and other dimensions for collection;
[0086] RLP (Radio Link Protocol) layer throughput: RLP layer throughput, like physical layer throughput, can represent the traffic size of the application;
[0087] Forward / reverse duty cycle: the ratio of high-speed duration to total connection duration. Generally speaking, forward rates above 50 kbit / s or reverse rates above 20 kbit / s are considered high-speed.
[0088] Forward / reverse effective rate: forward and reverse rates during data activation;
[0089] Forward / reverse activation duration: the ratio of wireless network element service duration to total connection duration;
[0090] The above data can be collected from wireless network elements to judge the application service of a certain connection. For example, the above data can be obtained through the network service performance indicators provided by the base station to the user. If at least three of the above data are greater than the preset judgment threshold, it can be known that the user is using a service with a large resource occupancy rate, and therefore the user is judged as a suspected uplink live broadcast user.
[0091] In one implementation, whether a user is a currently suspected uplink live broadcast user is determined based on the connection duration, forward throughput, reverse throughput, forward to reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate, and reverse effective rate of a user, specifically including:
[0092] Obtaining a plurality of preset user service type determination models, each user service type determination model including connection duration, forward throughput, reverse throughput, forward to reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate and reverse effective rate, at least three of which should be greater than a determination threshold;
[0093] In response to the connection duration, forward throughput, reverse throughput, forward-reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate and reverse effective rate of a certain user satisfying a certain user service type determination model, the type of service currently in use by the certain user is determined, and the certain user is determined to be a current suspected uplink live broadcast user.
[0094] In this embodiment, data can be collected through wireless network elements to obtain the occupation of different network resources by user services, and the application service type of a certain connection can be judged. The judgment models of various service types are described as follows:
[0095] Download category: The judgment model consists of connection duration, forward duty cycle, and forward flow rate judgment thresholds. For example, the connection duration is greater than 200 seconds, the forward duty cycle is greater than 70%, and the forward flow rate is greater than 10 Mbps.
[0096] Large rate burst class: The judgment model consists of connection duration, forward duty cycle, and forward effective rate;
[0097] Smaller rate burst type: The judgment model consists of connection duration, forward duty cycle, and forward effective rate. The forward effective rate of smaller rate burst services is lower than that of larger rate burst services.
[0098] Forward-reverse symmetry: The judgment model consists of the forward-reverse throughput ratio and the forward-reverse duty cycle;
[0099] Upload category: The judgment model consists of reverse throughput, reverse duty cycle, and reverse effective rate.
[0100] In one implementation, S2, identifying the current uplink live broadcast user according to the historical record data of the current suspected uplink live broadcast user, specifically includes:
[0101] Get the historical XDR data of the current suspected uplink live broadcast user;
[0102] Obtain the historical live broadcast data of the current suspected uplink live broadcast user based on the historical XDR data;
[0103] Determine the historical live broadcast type of the current suspected uplink live broadcast user based on historical live broadcast data;
[0104] In response to the type of the currently used service of the current suspected uplink live broadcast user belonging to its historical live broadcast type, the current suspected uplink live broadcast user is determined to be the current uplink live broadcast user.
[0105] Among them, XDR is extended detail record.
[0106] In this embodiment, the classification of user services can be completed through the air interface call data collected by the wireless network element, without the need to additionally monitor the core network side data and then perform in-depth unpacking (XDR) and other operations.
[0107] The complete business classification steps are as follows:
[0108] The wireless network element side collects the user's air interface call information;
[0109] Calculate characteristic key performance indicators (KPIs) for each call, such as call time, duty cycle, etc.
[0110] Compare the calculation results with the business model to determine the business types of all calls;
[0111] Merge the classified single calls to obtain business behaviors that are closer to user applications.
[0112] By establishing the corresponding relationship between wireless call KPI and network application through business model, wireless network elements collect call indicators and match them with business models to achieve the classification of business types. All information involved in the classification step can be collected under the existing system functions, without adding additional module design or deep unpacking of the core network, with low overhead, high data reliability and timeliness.
[0113] Then, combined with the user's historical XDR data, the live broadcast user profile built with performance is verified and related parameters are tuned. For example, by profiling the users of the service classification model, the uplink live broadcast service users in the network are identified. The interaction between live broadcast users and the network is a large-volume uplink data request, so they are called uplink live broadcast users.
[0114] In one implementation, in S3, identifying whether the cell where the current uplink live broadcast user is located is a suppressed cell specifically includes:
[0115] Identify the cell where the current uplink live broadcast user is located;
[0116] According to the historical performance indicators of the cell, the ideal traffic and / or ideal rate of each user in the cell where the current uplink live broadcast user is located is obtained;
[0117] According to the current performance indicators of the cell, obtain the current traffic and / or current rate of each user in the cell where the current uplink live broadcast user is located;
[0118] In response to a user in the cell having a current flow and / or a current rate that is less than the user's ideal flow and / or ideal rate, the cell where the current uplink live broadcast user is located is identified as a suppressed cell.
[0119] In this embodiment, the process of identifying a suppressed cell and performing network configuration on the suppressed cell is as follows: Figure 3As shown, it includes: data input related to the number of live broadcast users; through model training, fitting characteristic curves, and outputting a live broadcast user traffic suppression model; obtaining suppressed traffic and traffic corresponding to the suppression point; and generating a 5QI adaptive configuration strategy for live broadcast users based on whether the cell occupied by the live broadcast user is suppressed. A similar method is used for background users to obtain the result of whether it is suppressed. Model training mainly predicts the user's ideal traffic data based on the user's historical traffic data. The traffic data can be expressed in traffic / rate, and can identify users with suppressed traffic demand and the amount of suppressed traffic. Based on this, the 5QI configuration can be adjusted for all users in the cell to obtain the best network configuration result for all users in the cell.
[0120] Traffic suppression analysis extracts data such as traffic, number of users, and resource utilization within the network management system, combines big data analysis and machine learning algorithms, and outputs the results of cell traffic suppression analysis.
[0121] Specifically, the ideal traffic and / or ideal rate of the user can be obtained by referring to the time series prediction method. The time series prediction method extracts historical user numbers, historical traffic, historical resource utilization and historical rate data for training and modeling, provides cell-level capacity for forecasting and analysis, and predicts future traffic, number of users, resource utilization and single-user perceived rate. These indicators can be combined as needed as the basis for 5QI configuration. Specifically, the respective advantages of ARIMA and LSTM can be combined with the LGB algorithm to achieve time series prediction. The prediction process can include: selecting the most important features for the prediction task through LightGBML to generate new samples and features; using the ARIMA model to preprocess the time series, including differentiating the time series, determining the order of the ARIMA model, etc.; using the preprocessed time series as the input of the LSTM model, training the LSTM model to predict the time series; combining the prediction results of the ARIMA model and the LSTM model to obtain the final time series prediction result. This embodiment mainly utilizes the historical traffic / rate of each user to predict the current traffic / rate demand of the user. If the actual traffic / rate of the current user is less than the theoretically predicted traffic / rate (ideal traffic / ideal rate), it is considered that the user demand is suppressed.
[0122] In one implementation, in S3, identifying whether the cell where the current uplink live broadcast user is located is a suppressed cell specifically includes:
[0123] Identify current background users in the cell where the current uplink live broadcast user is located and other than the current uplink live broadcast user;
[0124] Obtaining a first ideal flow rate and / or a first ideal rate of the current background user according to a first historical flow rate and / or a first historical rate of the current background user when the historical time period has the same characteristics as the current time period and the current uplink live broadcast user has not performed any historical live broadcast behavior;
[0125] Obtaining a first current flow and / or a first current rate of a current background user in a current period;
[0126] In response to the first current traffic and / or the first current rate being less than the first ideal traffic and / or the first ideal rate, it is determined that the cell where the current uplink live broadcast user is located is a suppressed cell.
[0127] In this embodiment, to identify whether the background user demand in the cell is suppressed, the flow / rate of the background users who are currently broadcasting live is compared with the flow / rate of the background users when there is no live broadcast in history. If the flow / rate of the current background users is significantly lower than the flow / rate of the same period in history, it is considered that the background user flow demand is suppressed.
[0128] In one implementation, in S3, identifying whether the cell where the current uplink live broadcast user is located is a suppressed cell specifically includes:
[0129] Acquire a second ideal traffic volume and / or a second ideal rate of the current uplink live broadcast user according to the second historical traffic volume and / or the second historical rate of multiple historical live broadcast behaviors of the current uplink live broadcast user;
[0130] Obtain the second current flow and / or the second current rate of the current uplink live broadcast user in the current period;
[0131] In response to the second current traffic and / or the second current rate being less than the second ideal traffic and / or the second ideal rate, it is determined that the cell where the current uplink live broadcast user is located is a suppressed cell.
[0132] In this embodiment, to identify whether the demand of live broadcast users in a cell is suppressed, the main method is to compare the traffic / rate of live broadcast users predicted by history with the current traffic / rate. If the current traffic / rate is significantly less than the predicted traffic / rate, it is considered that the traffic demand of live broadcast users is suppressed. Considering the characteristics of the live broadcast services of live broadcast users changing over time, the second ideal traffic and / or the second ideal rate can be obtained by using a time series prediction method.
[0133] Specifically, the SVM (Support Vector Machine) algorithm can be used to identify suppressed cells and non-suppressed cells, extract the average number of activated users, downlink PRB (Physical Resource Block) utilization, C / I (Carrier-to-Interference ratio) and SINR (Signal to Interference plus Noise Ratio) and other data, and through training and learning modeling, fit the characteristic curve and output the traffic suppression model. When the prediction result meets the decision threshold of the suppressed cell, it is necessary to pre-start the 5QI adaptive method to protect the live broadcast users.
[0134] The SVM support vector machine algorithm is used to identify whether the cell occupied by the live broadcast user is a suppressed cell or a non-suppressed cell. The user's ideal traffic hyperplane equation in the rectangular coordinate system can be defined as Formula 1:
[0135] W T *x+b=0 Formula 1
[0136] In formula 1, W = (W1, W2…W n ), W is the weight and b is the intercept.
[0137] Let P(x1,x2…x n ) is the point of single user perception rate, where x i is the characteristic variable of the perceived rate of the i-th user, then the expression of the distance d from the user to the ideal traffic hyperplane is as follows:
[0138]
[0139] Among them, ‖W‖ is the function of the user's ideal traffic hyperplane, and the constant b is the intercept of the straight line equation corresponding to the perceived rate and ideal traffic of the plane rectangular coordinate system to the number of users.
[0140] Find the maximum intercept corresponding to the perceived rate of all users and the ideal traffic, so as to determine the hyperplane of the ideal traffic corresponding to the maximum perceived rate of all users in the 5G cell, as shown in Formula 3:
[0141]
[0142] Among them, y represents the data label, and the distance is y(W Tx+b), y takes -1 or 1. If the distance is in the positive direction, it is a positive number. If it appears in the negative direction, it is still a positive number. When the half that its normal vector points to is its front side, y takes 1, and when the other side is its back side, y takes -1. Note that when W and b are enlarged in equal proportions, the result of d will not change. Let u of the support vector be 1, and other points where u is greater than 1 can be solved by adjusting W and b. Therefore, Equation 3 can be simplified to Equation 4:
[0143]
[0144] For the convenience of calculation, the objective function is equivalently replaced by Formula 5:
[0145]
[0146] Use the Lagrange multiplier method (Equation 6) to solve:
[0147]
[0148] After solving α by taking the partial derivative of L using Formula 6, we can get W and b to get the function of the user's ideal traffic hyperplane as shown in Formula 7, where α is the Lagrange multiplier:
[0149]
[0150] Equation 7 needs to satisfy the Karush-Kuhn-Tucker condition of Equation 8 (a necessary and sufficient condition for a nonlinear programming problem to have an optimal solution under certain regular conditions):
[0151]
[0152] From formula 1-8, we can obtain the service perception rate and traffic classification decision function of the user accessing the network. When the decision function value (i.e. formula 1) is negative, it indicates that the cell accessed by the service user has experienced service rate suppression due to capacity issues. If the live broadcast user currently occupies a suppressed cell, the corresponding 5QI adaptive configuration should be generated to ensure the user's service perception.
[0153] In one implementation, S4, adjusting the network configuration of the user in the suppressed cell to ensure the user experience of the suppressed cell, specifically includes:
[0154] Based on the initialization settings of the iterative algorithm model, the following iterative operations are performed:
[0155] Obtain the 5QI configuration parameters of this iteration for each user in the suppressed cell;
[0156] Based on the 5QI configuration parameters and cell capacity of this iteration, obtain the predicted traffic and / or predicted rate of each user in this iteration;
[0157] According to the difference between the predicted flow and / or predicted rate of this iteration and the ideal flow and / or ideal rate, adjust the 5QI configuration parameters of this iteration to obtain the 5QI configuration parameters of the next iteration;
[0158] Repeat the above iterative operation until a preset end condition is reached, wherein the preset end condition includes that the difference between the predicted flow rate and / or predicted rate and the ideal flow rate and / or ideal rate is less than a preset difference threshold;
[0159] Output the 5QI configuration parameters corresponding to the time when the difference between the predicted flow rate and / or predicted rate and the ideal flow rate and / or ideal rate is the smallest;
[0160] Among them, the initialization settings include setting the initial 5QI configuration parameters, and 5QI is the 5G service quality indicator.
[0161] In this embodiment, the 5QI adaptive configuration combines network big data information, evaluates the current network capacity, service characteristics, traffic model, etc., performs traffic model and traffic load prediction, and combines the respective advantages of the ARIMA (AutoRegressive Integrated Moving Average) model of time prediction and the LSTM (Long Short-Term Memory) model of the long short-term memory neural network. The ARIMA model is used to preprocess the time series, including differentiating the time series and determining the order of the ARIMA model. The preprocessed time series is used as the input of the LSTM model, and the LSTM model is trained to predict the time series. The decision tree LGB (Light Gradient Boosting, based on gradient enhancement decision tree) algorithm is iteratively trained to obtain the optimal indicator prediction model related to cell capacity, provide cell-level capacity for predictive analysis, and predict future traffic, number of users, resource utilization and single-user perceived rate. These indicators can be combined as needed as the basis for 5QI configuration.
[0162] In order to manage 5QI dynamically, in real time and in a refined manner, step S1 obtains network user characteristics and provides the big data information of users in the network to the 5QI adaptive configuration module. The 5QI adaptive configuration combines the business model of the user's traffic constructed in steps S1 and S2, and the network depression situation, to perform traffic business model and traffic load prediction; decisions are made based on the traffic model (business characteristics, user distribution, where user distribution is obtained based on the user's current main service cell) and traffic load prediction information to dynamically determine what kind of service capabilities the network provides for background users and live users, and what kind of 5QI parameters are configured, so as to realize dynamic 5QI adaptive configuration management and achieve the purpose of improving user experience and reducing operating costs.
[0163] The 5QI adaptive configuration module can use the artificial intelligence ant colony algorithm (or other iterative optimization algorithms) to optimize the solution space (the user's 5QI optimal configuration) based on the balance between optimization objectives and search time. Through simulation learning, the 5QI configuration group of users with insignificant effects is discarded, and the maximum number of iterations is set to reduce the requirements for computing power, while also greatly reducing the search time for the optimal solution. The 5QI adaptive configuration gives the user's 5QI optimal configuration according to different scenarios, user distribution and optimization objectives. After being sent to the network element for execution, the optimization effect is verified through the UE's reported data, and the next optimization iteration is performed. If the KPI deteriorates, the previous 5QI configuration is directly rolled back.
[0164] The steps to implement 5QI adaptive configuration for users based on the adaptive ant colony algorithm are as follows:
[0165] 1) The mapping function between the service perception and network performance of live users and background users (users in the same primary service cell but with non-live services) is constructed as follows:
[0166] RANK=f(SSB,SRS,H) Formula 9
[0167] Among them, SSB (Synchronization Signal Block) is the key parameter of 5QI design, SRS (Sounding Reference Signal) is the detection reference signal, H (Half Rate) is half rate (a new wireless network optimization technology that reasonably adjusts the configuration rate and expands the network capacity), and the RANK function is the actual perceived rate of the user's corresponding 5QI parameter configuration in the wireless network environment;
[0168] 2) Initialize the service user population, maximum number of iterations, pheromone volatilization factor, transition probability constant, and local search step size;
[0169] 3) Randomly generate the initial configuration of 5QI of live broadcast users and background users, calculate the fitness function value, use the least squares method to select the residual sum of squares as the individual fitness function, set the initial pheromone, and calculate the probability of user service status change. The calculation formula is as follows:
[0170]
[0171] Among them, max(Tau) represents the maximum value of pheromone, Tau(i) represents the pheromone of user i, and P(iter,i) represents the transformation probability value of service user i in the iter-th iteration; each individual makes relevant selections under the corresponding probability; in order to make the genes of the fitness individual be inherited, the best individual in the group is directly selected; Formula 10 is the probability of the 5QI key parameters in Formula 9 being adapted to user perception by live users and background users;
[0172] 4) Update the 5QI configurations of live users and background users. The update rule is to replace the 5QI key parameter configurations with the worst adaptability affecting user perception in the existing network with relevant parameter configurations suitable for ensuring user perception;
[0173] 5) Calculate the updated fitness value, which is the calculation result of formula 10, determine whether the user service has changed, and update the pheromone;
[0174] 6) Determine whether the termination condition is met: when the maximum number of iterations is reached, the maximum duration of the algorithm is reached, and the residual sum of squares of the best individual in the population is less than or equal to the threshold, the iteration is terminated;
[0175] 7) If the condition is satisfied, the search process is terminated and the optimal result of 5QI configuration is output. The criterion for judging the optimal result is that the perception of background users and live users is not affected when the traffic of the primary service cell is suppressed. If the condition is not satisfied, iterative optimization is continued;
[0176] Based on the ant colony algorithm to obtain the 5QI configuration results, combined with the user's ideal traffic hyperplane equation in the plane rectangular coordinate system, the configuration results of the optimal 5QI related parameters of background users and live users are obtained to ensure the user's perceived rate.
[0177] Embodiment 2:
[0178] like Figure 4 As shown, the present disclosure provides a user experience guarantee device, the device comprising:
[0179] The first identification module 1 is used to identify the current suspected uplink live broadcast user according to the current wireless network element side resource occupancy data;
[0180] The second identification module 2 is connected to the first identification module 1 and is used to identify the current uplink live broadcast user according to the historical record data of the current suspected uplink live broadcast user;
[0181] The third identification module 3 is connected to the second identification module 2 and is used to identify whether the cell where the current uplink live broadcast user is located is a suppressed cell. A suppressed cell refers to a cell in which some users have given up some network services because the network performance does not meet their needs;
[0182] The adjustment module 4 is connected to the third identification module 3 and is used to adjust the network configuration of the users in the suppressed cell to ensure the user experience of the suppressed cell.
[0183] In one implementation, the first identification module 1 specifically includes:
[0184] A resource occupancy data acquisition unit is used to acquire and differentiate the current performance indicators of a wireless network element by user, so as to obtain the connection duration of a user's current service connected to a wireless network element and the forward throughput, reverse throughput, forward-to-reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate and reverse effective rate provided by a wireless network element for the user's current service during the connection duration;
[0185] The first determination unit is connected to the resource occupancy data acquisition unit, and is used to determine whether a user is a currently suspected uplink live broadcast user based on the user's connection duration, forward throughput, reverse throughput, forward to reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate and reverse effective rate.
[0186] In one implementation, the first determination unit specifically includes:
[0187] A determination model subunit is used to obtain a plurality of preset user service type determination models, each of which includes a determination threshold that at least three of the following items should be greater than: connection duration, forward throughput, reverse throughput, forward to reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate, and reverse effective rate;
[0188] The model corresponding subunit is connected to the judgment model subunit, and is used to respond to the connection duration, forward throughput, reverse throughput, forward-reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate and reverse effective rate of a user satisfying a user's service type judgment model, to judge the type of service currently in use by a user, and to judge a user as a currently suspected uplink live broadcast user.
[0189] In one implementation, the second identification module 2 specifically includes:
[0190] An XDR data acquisition unit is used to acquire historical XDR data of a current suspected uplink live broadcast user;
[0191] A live broadcast data acquisition unit, connected to the XDR data acquisition unit, for acquiring historical live broadcast data of a current suspected uplink live broadcast user according to historical XDR data;
[0192] A live broadcast type determination unit, connected to the live broadcast data acquisition unit, for determining the historical live broadcast type of the current suspected uplink live broadcast user according to the historical live broadcast data;
[0193] The live broadcast type comparison unit is connected to the live broadcast type judgment unit, and is used to determine that the current suspected uplink live broadcast user is the current uplink live broadcast user in response to the type of the currently used service of the current suspected uplink live broadcast user belonging to its historical live broadcast type.
[0194] Among them, XDR is extended detail record.
[0195] In one implementation, the third identification module 3 specifically includes:
[0196] A cell identification unit is used to identify the cell where the current uplink live broadcast user is located;
[0197] An ideal data acquisition unit, connected to the cell identification unit, is used to obtain the ideal flow and / or ideal rate of each user in the cell where the current uplink live broadcast user is located according to the historical performance indicators of the cell;
[0198] A current data acquisition unit, connected to the cell identification unit, is used to obtain the current flow and / or current rate of each user in the cell where the current uplink live broadcast user is located according to the current performance index of the cell;
[0199] The suppression identification unit is connected to the ideal data unit and the current data unit, and is used to identify that the cell where the current uplink live broadcast user is located is a suppressed cell in response to the current traffic and / or current rate of a user in the cell being less than the ideal traffic and / or ideal rate.
[0200] In one embodiment, the third recognition module 3 includes a background user suppression recognition unit, which specifically includes:
[0201] A background user identification subunit is used to identify current background users in the cell where the current uplink live broadcast user is located and other than the current uplink live broadcast user;
[0202] The background user ideal data subunit is connected to the background user identification subunit and is used to obtain the first ideal flow and / or first ideal rate of the current background user according to the first historical flow and / or first historical rate of the current background user in the historical period with the same characteristics as the current period and when the current uplink live broadcast user does not perform any historical live broadcast behavior;
[0203] A background user current data subunit, connected to the background user identification subunit, for obtaining a first current flow and / or a first current rate of a current background user in a current period;
[0204] The background user suppression determination subunit is connected to the background user ideal data subunit and the background user current data subunit, and is used to determine that the cell where the current uplink live broadcast user is located is a suppression cell in response to the existence of a first current flow and / or a first current rate that is less than a first ideal flow and / or a first ideal rate.
[0205] In one embodiment, the third identification module 3 includes a live broadcast user suppression identification unit, which specifically includes:
[0206] The live broadcast user ideal data subunit is used to obtain the second ideal traffic and / or second ideal rate of the current uplink live broadcast user according to the second historical traffic and / or second historical rate of multiple historical live broadcast behaviors of the current uplink live broadcast user;
[0207] The live broadcast user current data subunit is used to obtain the second current flow and / or the second current rate of the current uplink live broadcast user in the current period;
[0208] The live user suppression determination subunit is connected to the live user ideal data subunit and the live user current data subunit, and is used to determine that the cell where the current uplink live user is located is a suppressed cell in response to the existence of a second current flow and / or a second current rate that is less than the second ideal flow and / or the second ideal rate.
[0209] In one embodiment, the adjustment module 4 includes an iterative operation unit, specifically including:
[0210] The initialization unit is used to perform the following iterative operations based on the initialization settings of the iterative algorithm model:
[0211] An input parameter unit is used to obtain the 5QI configuration parameters of this iteration for each user in the suppressed cell;
[0212] A result prediction unit, connected to the input parameter unit, is used to obtain the current iteration predicted traffic and / or predicted rate of each user based on the current iteration 5QI configuration parameters and cell capacity;
[0213] A parameter adjustment unit, connected to the result prediction unit, is used to adjust the 5QI configuration parameters of this iteration according to the difference between the predicted flow and / or predicted rate of this iteration and the ideal flow and / or ideal rate, so as to obtain the 5QI configuration parameters of the next iteration;
[0214] An iterative judgment unit, connected to the parameter adjustment unit, is used to repeat the above iterative operation until a preset end condition is reached, wherein the preset end condition includes that the difference between the predicted flow rate and / or predicted rate and the ideal flow rate and / or ideal rate is less than a preset difference threshold;
[0215] A result output unit, connected to the iteration judgment unit, is used to output the 5QI configuration parameter corresponding to the time when the difference between the predicted flow rate and / or predicted rate and the ideal flow rate and / or ideal rate is the smallest;
[0216] Among them, the initialization settings include setting the initial 5QI configuration parameters, and 5QI is the 5G service quality indicator.
[0217] Embodiment 3:
[0218] Embodiment 3 of the present disclosure provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the user experience assurance method as described in Embodiment 1 is implemented, or the user experience assurance device as described in Embodiment 2 is implemented.
[0219] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program elements or other data). Computer-readable storage media include, but are not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable read only memory), flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory), digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.
[0220] In addition, the present disclosure may also provide a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor runs the computer program stored in the memory, the processor executes the user experience guarantee method as described in Example 1. The computer device may be the user experience guarantee device as described in Example 2.
[0221] The memory is connected to the processor, the memory may be a flash memory or a read-only memory or other memory, and the processor may be a central processing unit or a single-chip microcomputer.
[0222] Embodiments 1-3 of the present disclosure provide a user experience assurance method, a user experience assurance device, and a computer-readable storage medium, which identify uplink live broadcast users through the current wireless network element side resource occupancy data combined with the user's historical record data, thereby improving the real-time performance of live broadcast user identification. For cells with uplink live broadcast users, it is determined whether there is suppressed user service demand in the cell, and the user's network configuration is adjusted in the suppressed cell, which helps to release the suppressed user demand and coordinately ensure the service performance of live broadcast users and non-live broadcast users in the cell, thereby improving user experience and increasing operator revenue.
[0223] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present disclosure, but the present disclosure is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and substance of the present disclosure, and these modifications and improvements are also considered to be within the scope of protection of the present disclosure.
Claims
1. A method for ensuring user experience, characterized in that: The method comprises: According to the current resource occupancy data on the wireless network element side, the current suspected uplink live broadcast user is identified; Identify the current live streaming user based on the historical record data of the current suspected live streaming user; Identify whether the cell where the current uplink live broadcast user is located is a suppressed cell. A suppressed cell refers to a cell where some users have given up some network services because the network performance does not meet their needs. Adjust the network configuration of users in the depressed cells to ensure the user experience of the depressed cells.
2. The method according to claim 1, characterized in that According to the current wireless network element side resource occupancy data, the current suspected uplink live broadcast user is identified, including: Acquire and differentiate the current performance indicators of a wireless network element by user, so as to obtain the connection duration of a user's current service connected to a wireless network element and the forward throughput, reverse throughput, forward to reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate and reverse effective rate provided by the wireless network element for the user's current service during the connection duration; Determine whether a user is a currently suspected uplink live broadcast user based on the user's connection duration, forward throughput, reverse throughput, forward to reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate, and reverse effective rate.
3. The method according to claim 2, characterized in that Determine whether a user is a suspected uplink live broadcast user based on the user's connection duration, forward throughput, reverse throughput, forward to reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate, and reverse effective rate. Specifically, the following are used to determine whether a user is a suspected uplink live broadcast user: Obtaining a plurality of preset user service type determination models, each user service type determination model including connection duration, forward throughput, reverse throughput, forward to reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate and reverse effective rate, at least three of which should be greater than a determination threshold; In response to the connection duration, forward throughput, reverse throughput, forward-reverse throughput ratio, forward duty cycle, reverse duty cycle, forward effective rate and reverse effective rate of a certain user satisfying a certain user service type determination model, the type of service currently in use by the certain user is determined, and the certain user is determined to be a current suspected uplink live broadcast user.
4. The method according to claim 3, characterized in that According to the historical record data of the current suspected uplink live broadcast user, the current uplink live broadcast user is identified, including: Get the historical XDR data of the current suspected uplink live broadcast user; Obtain the historical live broadcast data of the current suspected uplink live broadcast user based on the historical XDR data; Determine the historical live broadcast type of the current suspected uplink live broadcast user based on historical live broadcast data; In response to the type of the currently used service of the current suspected uplink live broadcast user belonging to its historical live broadcast type, the current suspected uplink live broadcast user is determined to be the current uplink live broadcast user. Among them, XDR is extended detail record.
5. The method according to any one of claims 1 to 4, characterized in that: Identify whether the cell where the current uplink live broadcast user is located is a suppressed cell, including: Identify the cell where the current uplink live broadcast user is located; According to the historical performance indicators of the cell, the ideal traffic and / or ideal rate of each user in the cell where the current uplink live broadcast user is located is obtained; According to the current performance indicators of the cell, obtain the current traffic and / or current rate of each user in the cell where the current uplink live broadcast user is located; In response to a user in the cell having a current flow and / or a current rate that is lower than the user's ideal flow and / or ideal rate, the cell where the current uplink live broadcast user is located is identified as a suppressed cell.
6. The method according to claim 5, characterized in that Identify whether the cell where the current uplink live broadcast user is located is a suppressed cell, including: Identify current background users in the cell where the current uplink live broadcast user is located and other than the current uplink live broadcast user; Obtaining a first ideal flow rate and / or a first ideal rate of the current background user according to a first historical flow rate and / or a first historical rate of the current background user when the historical time period has the same characteristics as the current time period and the current uplink live broadcast user has not performed any historical live broadcast behavior; Obtaining a first current flow and / or a first current rate of a current background user in a current period; In response to the first current traffic and / or the first current rate being less than the first ideal traffic and / or the first ideal rate, it is determined that the cell where the current uplink live broadcast user is located is a suppressed cell.
7. The method according to claim 5, characterized in that Identify whether the cell where the current uplink live broadcast user is located is a suppressed cell, including: Acquire a second ideal traffic volume and / or a second ideal rate of the current uplink live broadcast user according to the second historical traffic volume and / or the second historical rate of multiple historical live broadcast behaviors of the current uplink live broadcast user; Obtain the second current flow and / or the second current rate of the current uplink live broadcast user in the current period; In response to the second current traffic and / or the second current rate being less than the second ideal traffic and / or the second ideal rate, it is determined that the cell where the current uplink live broadcast user is located is a suppressed cell.
8. The method according to claim 5, characterized in that Adjust the network configuration of users in the depressed cells to ensure the user experience in the depressed cells, including: Based on the initialization settings of the iterative algorithm model, the following iterative operations are performed: Obtain the 5QI configuration parameters of this iteration for each user in the suppressed cell; Based on the 5QI configuration parameters and cell capacity of this iteration, obtain the predicted traffic and / or predicted rate of each user in this iteration; According to the difference between the predicted flow and / or predicted rate of this iteration and the ideal flow and / or ideal rate, adjust the 5QI configuration parameters of this iteration to obtain the 5QI configuration parameters of the next iteration; Repeat the above iterative operation until a preset end condition is reached, wherein the preset end condition includes that the difference between the predicted flow rate and / or predicted rate and the ideal flow rate and / or ideal rate is less than a preset difference threshold; Output the 5QI configuration parameters corresponding to the time when the difference between the predicted flow rate and / or predicted rate and the ideal flow rate and / or ideal rate is the smallest; Among them, the initialization settings include setting the initial 5QI configuration parameters, and 5QI is the 5G service quality indicator.
9. A user experience guarantee device, characterized in that: The device comprises: The first identification module is used to identify the current suspected uplink live broadcast user according to the current wireless network element side resource occupancy data; A second identification module, connected to the first identification module, is used to identify the current uplink live broadcast user according to the historical record data of the current suspected uplink live broadcast user; The third identification module is connected to the second identification module and is used to identify whether the cell where the current uplink live broadcast user is located is a suppressed cell. A suppressed cell refers to a cell in which some users have given up some network services because the network performance does not meet their own needs; The adjustment module is connected to the third identification module and is used to adjust the network configuration of the users in the suppressed cell to ensure the user experience of the suppressed cell.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the user experience assurance method according to any one of claims 1 to 8 is implemented.
