5G video call quality management system based on network analysis

Abstract

The invention discloses a 5G video call quality management system based on network analysis, relates to the technical field of communication network quality monitoring and optimization, and is used for solving the problems that network resources cannot be efficiently distributed and dynamically optimized according to actual demands, and the stability of video call service and the user experience quality are reduced. The method comprises the following steps of: acquiring bandwidth utilization rate, time delay fluctuation and packet loss rate in real time in a communication link, calculating a node load balancing index, identifying a high-load node, setting a marking time period, acquiring network jitter frequency, a flow peak value and a terminal performance parameter of the node in a corresponding time period, and calculating a first optimization coefficient and a second optimization coefficient; and constructing a support vector machine model to predict optimization execution time, collecting performance data again at the time point, generating an optimization strategy coefficient in combination with a random forest algorithm, comparing the optimization strategy coefficient with a user experience threshold, generating a recommendation optimization strategy, and realizing adaptive optimization of network configuration.

Description

Technical Field

[0001] The present invention relates to the technical field of communication network quality monitoring and optimization. More specifically, the present invention relates to a 5G video call quality management system based on network analysis. Background Art

[0002] With the wide deployment and application of 5G communication technology, high-definition video calls have become an important way for mobile terminal users' daily communication. While 5G networks provide high bandwidth and low latency, their complex link structures and highly dynamic network states also make video call quality vulnerable to various factors, such as uneven base station loads, frequent link jitters, strong traffic bursts, and differences in terminal device performance. These factors may cause problems in user experiences such as call stuttering, blurred images, and voice delays, thus affecting the overall quality of 5G video call services.

[0003] The existing technologies have the following deficiencies:

[0004] Currently, most management methods only record static parameters, lacking dynamic analysis and joint optimization in dimensions such as network volatility, burst performance consumption, and user experience perception. They cannot achieve intelligent prediction and real-time adjustment of video call quality in complex network environments, resulting in the inability to efficiently allocate and dynamically optimize network resources according to actual needs, reducing the stability of video call services and the quality of user experiences. Therefore, a 5G video call quality management system based on network analysis is proposed.

[0005] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] To overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a 5G video call quality management system based on network analysis, which solves the problems raised in the above background art by using real-time acquisition of multi-source link parameters, an optimization execution timing determination method based on support vector machines, and an intelligent strategy generation mechanism integrating a random forest algorithm.

[0007] To achieve the above object, the present invention provides the following technical solution. A 5G video call quality management system based on network analysis includes the following modules:

[0008] The network status acquisition module real-time acquires 5G link performance parameters, and after receiving the marked nodes and marked time periods passed in by the data preprocessing module, as well as the optimization execution time passed in by the dynamic optimization module, further acquires the network jitter frequency, traffic peak distribution, and terminal device performance parameters of the marked nodes within the corresponding time period, and passes them to the dynamic optimization module;

[0009] The data preprocessing module receives link performance parameters, analyzes the node load situation, divides the nodes according to the load, marks the high-load nodes, sets multiple time periods as the marking time periods, and passes them to the network status collection module and the dynamic optimization module;

[0010] The dynamic optimization module calculates the first and second optimization coefficients based on the network jitter, traffic peak, and device performance of the marked nodes, constructs a support vector machine model to determine the optimization execution time, and feeds it back to the network status collection module; combines the network jitter frequency, traffic peak distribution, and terminal device performance parameters within the optimization execution time, and uses the random forest algorithm to generate optimization strategy coefficients, which are passed to the user feedback module;

[0011] The user feedback module receives the optimization strategy coefficients, generates recommended optimization strategies for the user side to select and adjust the network configuration of the marked nodes.

[0012] In a preferred embodiment, a fixed time period traced back from the current sampling moment is used as the acquisition time window;

[0013] Calculate the ratio of the bandwidth used by a certain network node within the acquisition time window to the available bandwidth of the node as the bandwidth utilization rate;

[0014] Calculate the arrival delay difference between consecutive transmitted data packets to obtain the delay fluctuation;

[0015] Statistically calculate the proportion of data packets lost during data transmission to obtain the packet loss rate.

[0016] In a preferred embodiment, standardize the bandwidth utilization rate, delay fluctuation, and packet loss rate to obtain a standardized feature vector;

[0017] Define the node load status variable and construct a sample set;

[0018] Statistically calculate the frequency of each load category in the sample set and calculate the prior probability of the node load status;

[0019] Use one-dimensional Gaussian distribution modeling to calculate the conditional probability distributions of the bandwidth utilization rate, delay fluctuation, and packet loss rate under different load states;

[0020] Combine the feature vector to calculate the posterior probabilities of the node belonging to the high-load and low-load states;

[0021] If the posterior probability of the node belonging to the high load is greater than the preset determination threshold, then determine that the node is a high-load node and use it as a marked node.

[0022] In a preferred embodiment, when determining that the node is a high-load node and using it as a marked node, trace back a preset marked time period from the current moment;

[0023] Collect the network jitter frequency and traffic peak distribution of the labeled nodes within the marked time period;

[0024] Collect the CPU utilization rate, memory occupancy rate, and image encoding and compression frame rate as the performance parameters of the terminal device.

[0025] In a preferred embodiment, a weighted average of the network jitter frequency and traffic peak distribution is obtained to get the first optimization coefficient;

[0026] Construct a linear logistic regression for the CPU utilization rate, memory occupancy rate, and image encoding and compression frame rate to get the second optimization coefficient.

[0027] In a preferred embodiment, a support vector machine model is constructed by combining the first optimization coefficient and the second optimization coefficient to obtain the optimized execution time;

[0028] Combine the first and second optimization coefficients into a coefficient feature vector;

[0029] Use the radial basis kernel function for the coefficient feature vector, and calculate the optimal hyperplane that maximizes the classification boundary by minimizing the Lagrangian objective function;

[0030] Obtain the support vector set and corresponding coefficients through training;

[0031] Calculate the classification scores at each time point using the support vector set and corresponding coefficients, and select the time point with the highest score as the optimized execution time;

[0032] Collect the network jitter frequency, traffic peak distribution, and terminal device performance parameters of the labeled nodes at the optimized execution time.

[0033] In a preferred embodiment, after normalizing the network jitter frequency, traffic peak distribution, and terminal device performance parameters of the labeled nodes, construct an optimized feature vector;

[0034] Use the feature vector as the input and calculate the optimized strategy coefficient using a random forest model;

[0035] Use the bootstrap sampling method to extract multiple groups of training samples from the optimized feature vector, and build a decision tree for each group;

[0036] Each decision tree recursively selects the feature with the largest Gini gain to partition the data and forms a complete decision tree;

[0037] Input the new node features into all decision trees simultaneously and collect the prediction scores of each tree;

[0038] Take the average of all the prediction scores to obtain the optimized strategy coefficient.

[0039] In a preferred embodiment, the optimization strategy coefficient of each marked node is compared with the user experience threshold to obtain the optimization recommendation level;

[0040] If the optimization strategy coefficient is greater than or equal to the user experience threshold, the optimization recommendation level is strong recommendation for optimization;

[0041] If the optimization strategy coefficient is less than the user experience threshold, the optimization recommendation level is strong recommendation for optimization.

[0042] In a preferred embodiment, according to the optimization recommendation level, the corresponding network configuration parameter adjustment suggestions are extracted to generate a recommended optimization strategy;

[0043] The recommended optimization strategy is sent to the user terminal system to adjust the access configuration and scheduling strategy of the network environment where the current node is located.

[0044] Technical effects and advantages of the present invention:

[0045] The present invention collects link performance parameters in real time in the communication link, including bandwidth utilization rate, delay fluctuation and packet loss rate, calculates the load balancing index of each network node, and identifies high-load nodes. Subsequently, a time window is set as the marked time period, and the network jitter frequency, traffic peak distribution and terminal device performance parameters of the high-load nodes are collected within the marked time period. Based on these data, the first optimization coefficient and the second optimization coefficient are calculated, and a support vector machine model is constructed to predict the best optimization execution time. At this time point, relevant performance parameters are further collected, and a random forest algorithm is used to generate an optimization strategy coefficient. Finally, the optimization strategy coefficient is compared with the preset user experience threshold to generate a recommended optimization strategy for the user to select and adjust the network configuration, so as to realize the adaptive optimization of the network configuration. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flowchart of the implementation of a 5G video call quality management system based on network analysis according to the present invention.

[0047] Figure 2 It is a schematic diagram of the steps of a 5G video call quality management system based on network analysis according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0049] Embodiment 1, a 5G video call quality management system based on network analysis, as Figure 1 shown, includes a network status collection module, a data preprocessing module, a dynamic optimization module, and a user feedback module, and the modules are connected by electrical signals;

[0050] The network status collection module collects real-time link performance parameters in the 5G network environment and transmits the collected data to the data preprocessing module; after detecting the marked nodes and marked time periods transmitted by the data preprocessing module and the optimization execution time transmitted by the dynamic optimization module, it further obtains the network jitter frequency, traffic peak distribution, and terminal device performance parameters of the marked nodes within the marked time period and the optimization execution time and transmits them to the dynamic optimization module;

[0051] The data preprocessing module receives the link performance parameters transmitted by the network status collection module, analyzes the load balancing situation of each network node according to the link performance parameters, divides the network nodes into high-load nodes and low-load nodes, and screens out the high-load nodes for marking; sets multiple time periods with different time intervals as marked time periods, and transmits the marked nodes and marked time periods to the network status collection module and the dynamic optimization module;

[0052] The dynamic optimization module receives the network jitter frequency, traffic peak distribution, and terminal device performance parameters of the marked nodes within the marked time period, calculates the first optimization coefficient and the second optimization coefficient, constructs a support vector machine model by synthesizing the first optimization coefficient and the second optimization coefficient, determines the optimization execution time within the marked time period, and sends the optimization execution time to the network status collection module; synthesizes the network jitter frequency, traffic peak distribution, and terminal device performance parameters of each marked node within the optimization execution time, generates an optimization strategy coefficient by using the random forest algorithm, and transmits the optimization strategy coefficient to the user feedback module;

[0053] The user feedback module receives the optimization strategy coefficient transmitted by the dynamic optimization module, performs an operation with a preset user experience threshold, generates a recommended optimization strategy, and sends the recommended optimization strategy to the user side for the user to select and adjust the network configuration of the corresponding marked node.

[0054] The specific implementation is as follows:

[0055] In the network status collection module, by monitoring and counting the data stream transmitted in real time in the communication link, with the current sampling moment as the benchmark, a fixed time period with a length of is used as the collection time window, and the link performance parameters in the 5G network environment are collected in real time within the collection time window. The link performance parameters include three indicators: bandwidth utilization rate, delay fluctuation, and packet loss rate.

[0056] The bandwidth utilization rate is the ratio of the bandwidth used by a certain network node within the acquisition time window to the available bandwidth of that node. The calculation formula is ;

[0057] Among them, is the bandwidth utilization rate, represents the actual bandwidth usage of node n at time point t, represents the total allocable bandwidth of node n at the same time point.

[0058] The delay fluctuation represents the difference in the arrival delay between continuously transmitted data packets. Its calculation formula is ;

[0059] Among them, is the delay fluctuation, represents the one-way transmission delay of the i-th data packet of node n at time point t, is the average transmission delay of node n within this time period, and M is the number of statistical samples.

[0060] The packet loss rate represents the proportion of data packets lost during data transmission. The calculation formula is ;

[0061] Among them, is the packet loss rate, represents the number of data packets lost by node n at time point t, represents the total number of data packets sent by this node within the same time period.

[0062] It should be noted that the communication link refers to the physical channel or logical path used to achieve data transmission in the communication system.

[0063] The network status acquisition module transmits the above link performance parameters to the data preprocessing module to provide the original support data for subsequent load analysis and network optimization strategy generation.

[0064] In the data preprocessing module, a Bayesian classification model is used to identify the load status of network nodes, and the network nodes are divided into high-load nodes and low-load nodes;

[0065] To eliminate the influence of dimensions and unify the value range, first, the bandwidth utilization rate, delay fluctuation, and packet loss rate are standardized to obtain the standardized feature vector:

[0066] ;

[0067] Among them, , , , are the mean values of the bandwidth utilization rate, delay fluctuation, and packet loss rate respectively, is its corresponding standard deviation.

[0068] Define the node load status variable , when the node is in a high load state, , when the node is in a low load state, ;

[0069] Construct a sample set by collecting historical data from the historical database , count the frequencies of each load category in the sample set, and calculate the prior probability P(y) of the node load status as follows:

[0070] ;

[0071] where, represents the number of high load nodes in the sample, represents the number of low load nodes, .

[0072] For each dimension feature , calculate its conditional probability distribution under different load states . For ease of calculation, a one-dimensional Gaussian distribution is used for modeling, and the expression is ;

[0073] where, represents the mean of the feature under the load state , is its standard deviation.

[0074] Based on Bayes' theorem, combined with the feature vector , calculate its posterior probabilities of belonging to high load and low load states. The calculation formula is as follows: ;

[0075] where, the numerator is the joint probability of the current load state , and the denominator is the sum of the joint probabilities under all possible states.

[0076] Compare the posterior probabilities of the two states. The decision rule is as follows:

[0077] If , then judge that the node is a high load node;

[0078] If , then judge that the node is a low load node;

[0079] where, is the discrimination threshold, which is obtained by professionals through experiments and will not be elaborated here.

[0080] Denote the set of nodes identified as high - load states as:

[0081] ;

[0082] Perform a marking process on each network node in the above - mentioned set of nodes;

[0083] To support multi - scale time - domain analysis, when a node is determined to be a high - load node and serves as a marked node, further preset K marking time periods with different durations ;

[0084] Among them, , each marking time period represents a time window that traces back from the current moment t.

[0085] It should be noted that the Bayesian classification model is a supervised learning model for probability inference based on Bayes' theorem, through the joint calculation of prior probability and conditional probability; the historical database is used to record the operation status data and load annotation information of network nodes in multiple historical time periods; the prior probability refers to the known probability of the occurrence of an event or category itself without observing sample data; one - dimensional Gaussian distribution modeling refers to the method of modeling a continuous variable in a certain dimension with a normal distribution; the posterior probability is the conditional probability of the occurrence of a certain category given the observed sample.

[0086] The data pre - processing module passes the marked nodes and marking time periods screened above to the network status acquisition module and the dynamic optimization module.

[0087] After receiving the marked nodes and marking time periods passed by the data pre - processing module, the network status acquisition module collects the network jitter frequency, traffic peak distribution, and terminal device performance parameters of the marked nodes within the marking time period.

[0088] The network jitter frequency is the delay change of consecutive video call data packets within the marking time period , and the calculation formula is , where is the network jitter frequency, represents the network delay of the i - th data packet, and N represents the total number of data packets collected within the marking time period.

[0089] The traffic peak distribution represents the maximum uplink or downlink throughput value within the marking time period , and the calculation expression is , where is the traffic peak distribution, represents the number of transmitted bytes at the sampling time point , represents the window time length.

[0090] The performance parameters of the terminal device, denoted as , collect the CPU utilization rate, memory occupancy rate, and image encoding and compression frame rate of the terminal device of this node, and form the following feature vector:

[0091] ;

[0092] Among them, is the CPU utilization rate, is the memory occupancy rate, is the image encoding and compression frame rate;

[0093] After the network status acquisition module completes the acquisition of the above three types of performance parameters, it will the data set is input to the dynamic optimization module.

[0094] It should be noted that the uplink throughput value refers to the amount of data successfully transmitted by the terminal device in the direction of the base station per unit time; the downlink throughput value refers to the amount of data successfully transmitted by the base station in the direction of the terminal device per unit time.

[0095] The dynamic optimization module calculates the first optimization coefficient and the second optimization coefficient respectively by weighted average based on the input feature set, which are used to characterize the short-term jitter adaptability and device bearing margin;

[0096] The first optimization coefficient is calculated from the network jitter frequency and traffic peak distribution, and the calculation formula is as follows:

[0097] ;

[0098] Among them, is the first optimization coefficient, is the network jitter frequency, is the traffic peak distribution, the network jitter frequency and the traffic peak distribution The larger the value, the larger the first optimization coefficient .

[0099] The second optimization coefficient is calculated from the performance parameters of the terminal device, and the calculation formula is as follows:

[0100] ;

[0101] Among them, is the second optimization coefficient, is the CPU utilization rate in the performance parameters of the terminal device, is the memory occupancy rate in the performance parameters of the terminal device, is the image encoding and compression frame rate in the performance parameters of the terminal device, is the maximum coding frame rate supported by the system, is the weight parameter.

[0102] The dynamic optimization module combines the first optimization coefficient and the second optimization coefficient to construct a support vector machine model, and uses supervised learning to train and predict the optimization execution time within the marked time period ;

[0103] First, the first optimization coefficient and the second optimization coefficient are combined to form a joint feature vector , defined as ;

[0104] Based on the above feature vector set , a support vector machine model is constructed to predict the optimization execution time within the given marked time period. The support vector machine model uses the radial basis kernel function as the kernel function form, defined as ;

[0105] Among them, is the kernel width parameter, represents any two training samples, and the optimal hyperplane is solved by minimizing the Lagrangian objective function during the training process. The calculation formula is ;

[0106] The constraint conditions are: ;

[0107] Among them, is the weight vector, is the bias term, is the feature mapping function, is the soft margin variable, is the penalty factor, represents the sample classification label. In this case, the classification label is optimization adaptation or non - adaptation.

[0108] Based on the support vector set and the corresponding coefficients obtained after training, the final discriminant function is constructed as , among which, is the Lagrangian multiplier corresponding to the support vector.

[0109] The dynamic optimization module inputs the discrete time points within the time period into the corresponding one by one, and obtains the optimization execution time . The is the time point with the highest score among the positive class discriminant results, and its calculation formula is ;

[0110] It should be noted that the support vector machine model is a supervised learning method for classification and regression tasks; the radial basis kernel function is a commonly used support vector machine kernel function for controlling the scale of non-linear mapping; the Lagrangian objective function is used to solve the Lagrange multipliers, and the minimization of this function is the optimization objective for model training.

[0111] The dynamic optimization module transmits the obtained optimized execution time to the network status collection module to guide the subsequent data collection enhancement and scheduling behaviors of this node.

[0112] After receiving the optimized execution time passed in by the dynamic optimization module, the network status collection module collects the network jitter frequency, traffic peak distribution, and terminal device performance parameters of the marked nodes at the optimized execution time.

[0113] After the dynamic optimization module receives the unified normalization processing of the said parameters, for each marked node n, a feature vector is constructed: ;

[0114] Subsequently, the random forest algorithm is used to calculate the optimization strategy coefficient. The random forest model consists of independent classification and regression trees to fit and predict the scores of the scheduling strategies for the node status. Each tree constructs a bootstrap sample set from the training set data by sampling with replacement during the training stage, and constructs a tree structure in the feature subspace. The specific division basis is to maximize the information gain: ;

[0115] Among them, is the current node sample set, represents the subset after division according to the value of the feature , is the proportion of samples belonging to the th class.

[0116] During the model prediction stage, the dynamic optimization module inputs the current feature vector into all subtrees to obtain the individual prediction results of each subtree. Finally, the optimization strategy coefficient is obtained from the average of the prediction values of each subtree: ;

[0117] Among them, represents the policy adjustment priority of node during the current optimized execution time period. The higher the optimization strategy coefficient value, the more the node needs to be scheduled and adjusted under network fluctuation and load scenarios, and the higher the priority.

[0118] It should be noted that random forest is an ensemble learning algorithm composed of multiple decision trees; sampling with replacement is a statistical resampling method that randomly selects samples from the original dataset to form a new sample set, and the samples can be drawn again after each sampling.

[0119] The user feedback module receives the optimization strategy coefficient passed in by the dynamic optimization module and then, for each marked node compares the optimization strategy coefficient with the user experience threshold and performs the operation of generating a recommended optimization strategy. The user experience threshold is a preset parameter that satisfies and is used to represent the lower limit of the minimum network service quality score acceptable to users, which is obtained by those skilled in the art through experiments.

[0120] Evaluates the optimization recommendation level for each node ; If

[0121] If , then it is a strong recommendation for optimization;

[0122] If , then it is a weak recommendation for optimization.

[0123] After completing the classification of the optimization recommendation levels for all nodes, the user feedback module extracts the corresponding network configuration parameter adjustment suggestions from the optimization strategy rule library according to each type of optimization recommendation level to generate a recommended optimization strategy ;

[0124] Among them, represents the recommended network parameter configuration of node n, which is determined by the optimization strategy function, that is , and the policy function calls the matching parameter template in the optimization strategy rule library according to the combined relationship between the node optimization strategy coefficient and the recommendation level to complete the generation of the recommended strategy content.

[0125] The user feedback module encapsulates the recommended optimization strategy and sends it to the user terminal system, and presents it to the user through the interface control interface. The user side can choose whether to apply the recommended configuration according to the actual perception. If it is confirmed to execute, the terminal system will adjust the access configuration and scheduling strategy of the network environment where the current node is located according to the recommended parameters to complete the process of adaptive adjustment of the network configuration for the marked node.

[0126] It should be noted that the optimization strategy rule library is a structured collection of strategy knowledge, which defines the network state characteristics and recommended configuration behaviors corresponding to each rule, and will not be elaborated here.

[0127] The above formulas are all dimensionless and only take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0128] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from a website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that the computer can access, or a data storage device such as a server or data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0129] It should be understood that in various embodiments of the present application, the magnitudes of the serial numbers of the above processes do not mean the order of execution, and the execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0130] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professionals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0131] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0132] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0133] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0134] In addition, in each embodiment of the present application, the functional units can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0135] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or a part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0136] The above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims described above.

Claims

1. A 5G video call quality management system based on network analysis, characterized in that: It includes the following modules: The network status collection module collects 5G link performance parameters in real time. After receiving the marked nodes, marked time periods passed in by the data preprocessing module, and the optimization execution time passed in by the dynamic optimization module, it further collects the network jitter frequency, traffic peak distribution, and terminal device performance parameters of the marked nodes within the corresponding time period, and passes them to the dynamic optimization module; The data preprocessing module receives the link performance parameters, analyzes the node load situation, divides the nodes according to the load and marks the high-load nodes, sets multiple time periods as the marked time periods, and passes them to the network status collection module and the dynamic optimization module; The dynamic optimization module calculates the first and second optimization coefficients based on the network jitter, traffic peak, and device performance of the marked nodes, constructs a support vector machine model to determine the optimization execution time, and feeds it back to the network status collection module; Combined with the network jitter frequency, traffic peak distribution, and terminal device performance parameters within the optimization execution time, a random forest algorithm is used to generate the optimization strategy coefficient, which is passed to the user feedback module; The user feedback module receives the optimization strategy coefficient, generates a recommended optimization strategy for the user side to select and adjust the network configuration of the marked nodes.

2. A 5G video call quality management system based on network analysis according to claim 1, characterized in that: Taking the current sampling moment as a reference, a fixed time period traced back forward is used as the acquisition time window; Calculating the ratio of the bandwidth used by a certain network node within the acquisition time window to the available bandwidth of the node as the bandwidth utilization rate; Calculating the arrival delay difference between consecutive transmitted data packets to obtain the delay fluctuation; Counting the proportion of lost data packets during the data transmission process to obtain the packet loss rate.

3. A 5G video call quality management system based on network analysis according to claim 2, characterized in that: Performing standardization processing on the bandwidth utilization rate, delay fluctuation, and packet loss rate to obtain a standardized feature vector; Defining a node load status variable and constructing a sample set; Counting the frequencies of each load category in the sample set and calculating the prior probability of the node load status; Using a one-dimensional Gaussian distribution model to calculate the conditional probability distributions of the bandwidth utilization rate, delay fluctuation, and packet loss rate under different load states; Combined with the feature vector, calculating the posterior probabilities of the node belonging to the high-load and low-load states; If the posterior probability of the node belonging to the high-load state is greater than the preset determination threshold, then the node is determined to be a high-load node and used as a marked node.

4. A 5G video call quality management system based on network analysis according to claim 3, characterized in that: When the node is determined to be a high-load node and used as a marked node, a preset marked time period is traced back forward based on the current moment; Collecting the network jitter frequency and traffic peak distribution of the marked node within the marked time period; Collecting the CPU utilization rate, memory occupancy rate, and image encoding and compression frame rate as the terminal device performance parameters.

5. A 5G video call quality management system based on network analysis according to claim 4, characterized in that: Performing weighted averaging on the network jitter frequency and traffic peak distribution to obtain the first optimization coefficient; Construct a linear logistic regression for CPU utilization rate, memory occupancy rate, and image encoding and compression frame rate to obtain the second optimization coefficient.

6. A 5G video call quality management system based on network analysis according to claim 5, characterized in that: Combine the first optimization coefficient and the second optimization coefficient to construct a support vector machine model to obtain the optimized execution time; Merge the first and second optimization coefficients into a coefficient feature vector; Use the radial basis kernel function for the coefficient feature vector, and calculate the optimal hyperplane that maximizes the classification boundary by minimizing the Lagrangian objective function; Obtain the support vector set and corresponding coefficients through training; Calculate the classification scores at each time point using the support vector set and corresponding coefficients, and select the time point with the highest score as the optimized execution time; Collect the network jitter frequency, traffic peak distribution, and terminal device performance parameters of the marked node at the optimized execution time.

7. A 5G video call quality management system based on network analysis according to claim 6, characterized in that: After normalizing the network jitter frequency, traffic peak distribution, and terminal device performance parameters of the marked node, construct an optimized feature vector; Use the feature vector as input and calculate the optimized policy coefficient using a random forest model; Use the bootstrap sampling method to extract multiple groups of training samples from the optimized feature vector, and build a decision tree for each group; Each decision tree recursively selects the feature with the largest Gini gain to divide the data to form a complete decision tree; Input the new node features into all decision trees simultaneously, and collect the prediction scores of each tree; Take the average of all prediction scores to obtain the optimized policy coefficient.

8. A 5G video call quality management system based on network analysis according to claim 7, characterized in that: Compare the optimized policy coefficient of each marked node with the user experience threshold to obtain the optimized recommendation level; If the optimized policy coefficient is greater than or equal to the user experience threshold, the optimized recommendation level is strong recommendation for optimization; If the optimized policy coefficient is less than the user experience threshold, the optimized recommendation level is strong recommendation for optimization.

9. A 5G video call quality management system based on network analysis according to claim 8, characterized in that: Extract the corresponding network configuration parameter adjustment suggestions according to the optimized recommendation level, and generate a recommended optimization strategy; Send the recommended optimization strategy to the user terminal system to adjust the access configuration and scheduling strategy of the network environment where the current node is located.

Images