Network performance evaluation method and device based on dynamic evidence reasoning rule

Abstract

The invention discloses a network performance evaluation method based on a dynamic evidence reasoning rule, and the method comprises the steps: dynamically adjusting a reference value of a test index based on the change of a network state; the adjusted reference value is used for converting the monitoring data into a confidence distribution form, and the test index is converted into a corresponding evidence form according to an IF-Then mapping relation; fusing the evidences corresponding to the plurality of indexes based on an evidence reasoning rule to obtain an evaluation result; and by setting constraint conditions and optimization targets of the parameters, optimizing the model parameters by adopting an optimization algorithm. The invention further discloses a network performance evaluation device based on the dynamic evidence reasoning rule. According to the network performance evaluation method based on the dynamic evidence reasoning rule, the actual performance of the network in different scenes can be reflected more accurately according to the change of the network state, and the evaluation requirements of different network environments and service types are met.

Description

Technical Field

[0001] The present application relates to the field of network intelligent operation and maintenance technology, and more specifically, to a network performance evaluation method and device based on dynamic evidence reasoning rules. Background Art

[0002] With the rapid development of technologies such as 5G, edge computing, and artificial intelligence, network architecture is evolving towards cloud-network integration and computing-network integration, which places higher demands on network operation and maintenance. As an important part of intelligent operation and maintenance, the accuracy of network performance evaluation directly affects the judgment of network status and the formulation of optimization decisions. However, existing network performance evaluation methods have some limitations. Traditional evaluation methods are mainly divided into quantitative data-based methods, qualitative knowledge-based methods, and mixed information-based methods. Quantitative data-based methods are prone to overfitting, and the evaluation results are difficult to interpret; qualitative knowledge-based methods rely on expert experience and knowledge, and the evaluation accuracy is relatively low; mixed information-based methods combine quantitative data and qualitative knowledge, and evaluate by integrating multi-source information, which has great application potential.

[0003] As a hybrid information evaluation method, the Evidential Reasoning (ER) rule can transform various test indicators into evidence forms and fuse them through orthogonal sum operations, providing a new idea for comprehensively grasping network performance. However, the evidence reference values ​​in the ER rules are usually static and cannot adapt to the dynamic changes of network status. Therefore, static reference values ​​are difficult to accurately reflect network performance in the face of complex and changing network environments, resulting in insufficient evaluation accuracy. This limits the effectiveness of ER rules in practical applications, and an evaluation method that can dynamically adjust reference values ​​is urgently needed to improve the accuracy and reliability of the evaluation. Summary of the invention

[0004] In response to at least one defect or improvement need in the prior art, the present invention provides a network performance evaluation method and device based on dynamic evidence reasoning rules, which can solve at least one of the problems existing in the above-mentioned background technology.

[0005] To achieve the above object, according to a first aspect of the present invention, a network performance evaluation method based on dynamic evidence reasoning rules is provided, the method comprising:

[0006] Dynamically adjust the reference value of the test indicator based on changes in network status;

[0007] The adjusted reference values ​​are used to convert the monitoring data into a confidence distribution form, and the test indicators are converted into corresponding evidence forms according to the IF-Then mapping relationship;

[0008] Based on the evidence reasoning rules, the evidence corresponding to multiple indicators is integrated to obtain the evaluation results;

[0009] By setting the constraint conditions and optimization objectives of the parameters, the optimization algorithm is used to optimize the model parameters.

[0010] Furthermore, in the above-mentioned network performance evaluation method based on dynamic evidence reasoning rules, the network state includes the network being in a stable state, the network being in a slowly changing state, and the network being in a step-changing state.

[0011] Furthermore, the above-mentioned network performance evaluation method based on dynamic evidence reasoning rules, dynamically adjusting the reference value of the test indicator based on the change of the network state, specifically includes:

[0012] When the network is in a stable state, there are a total of T 1 observation time, and the monitoring data corresponding to each observation time is expressed as Its mean is expressed as

[0013]

[0014] Calculate the distance between each monitoring data and the mean have

[0015]

[0016] Based on the calculated maximum distance and minimum distance For the N reference levels G of this indicator n ,n=1,2,…,N, the step amount Δd i Expressed as

[0017]

[0018] Arrange the reference values ​​in ascending order. When N is an odd number, the index reference value is

[0019]

[0020] When N is an even number, the reference value of the index is

[0021]

[0022] Furthermore, the above-mentioned network performance evaluation method based on dynamic evidence reasoning rules, dynamically adjusting the reference value of the test indicator based on the change of the network state, specifically includes:

[0023] When the network is in a slowly changing state, there is a total of T 2 observation time, and the monitoring data corresponding to each observation time is expressed as Adjacent moments The amount of change Expressed as

[0024]

[0025] Will As a new observation sequence, the indicator x is calculated using the same calculation method as when the network is in a stable state. i The reference value of the indicator under this new observation sequence Will The indicator reference value restored to the original sequence

[0026] Furthermore, the above-mentioned network performance evaluation method based on dynamic evidence reasoning rules, dynamically adjusting the reference value of the test indicator based on the change of the network state, specifically includes:

[0027] When the network is in a step change state, the step change state is decomposed into a stable state and a slowly changing state according to the change of the state, and the dynamic reference value of the indicator is calculated according to the stable state and the slowly changing state respectively.

[0028] Furthermore, the above-mentioned network performance evaluation method based on dynamic evidence reasoning rules, wherein the adjusted reference value is used to convert the monitoring data into a confidence distribution form, and the test index is converted into a corresponding evidence form according to the IF-Then mapping relationship, specifically includes:

[0029] The indicator and the corresponding dynamic reference value are described as the confidence distribution form as follows

[0030]

[0031] Among them, δ n,i (t), n=1,2,…,N represents the index x i At time t relative to the reference level G n Confidence level;

[0032] Index x i The corresponding evidence n,i Expressed as

[0033] e n,i :Ifx i (t)ish i (G n ), then y(t)is{(H 1 ,β 1,j,i ),(H 2 ,β 2,j,i )…,(H K ,β K,j,i )}

[0034] Among them, y(t) represents the system output, H k ,k=1,2,…,K represents different output states, β k,n,i ,k=1,2,…,K represents the number of cells assigned to H k Confidence level;

[0035] Evidence n,i The weight w n,i Expressed as

[0036]

[0037] in, Indicates the index x i own weight.

[0038] Furthermore, the above-mentioned network performance evaluation method based on dynamic evidence reasoning rules, wherein the evidence corresponding to multiple indicators is integrated based on the evidence reasoning rules to obtain the evaluation result, specifically includes:

[0039] Arrange the evidence corresponding to all indicators in order, denoted as e = {e l ; l=1,2,…,L}, the weight of evidence, reliability and confidence are expressed as w={w l ; l=1,2,…,L}, r={r l ; l=1,2,…,L} and β={β k,l ; l=1,2,…,L}, the total number of evidence is L, and the evaluation result is expressed as

[0040]

[0041] Among them, η(t) is an intermediate variable, expressed as

[0042]

[0043] Among them, p k,e(L) (t) represents the confidence of the fusion result at time t relative to level k, p k,e(L) The utility of (t) is u k,e(L) The numerical form of the evaluation result is expressed as

[0044]

[0045] Furthermore, in the above-mentioned network performance evaluation method based on dynamic evidence reasoning rules, the constraint conditions and optimization objectives of setting parameters specifically include:

[0046] The constraints on the parameters are:

[0047] For the reliability of evidence l ,have

[0048] 0 <r l <1

[0049] For the indicator weight w i ,have

[0050]

[0051] in, and Respectively represent the preset w i The upper and lower bounds of

[0052] For the confidence level β k,l ,have

[0053]

[0054] The optimization objectives of the parameters are:

[0055] For real output of network performance Its mean square error MSE is expressed as

[0056]

[0057] Establish optimization goals

[0058]

[0059] Furthermore, in the above-mentioned network performance evaluation method based on dynamic evidence reasoning rules, the optimization algorithm is used to optimize the model parameters, including based on the parameter constraints and optimization objectives, and the projection covariance matrix adaptive evolution algorithm is used as an optimization tool to optimize the model parameters.

[0060] According to a second aspect of the present invention, there is also provided a network performance evaluation device based on dynamic evidence reasoning rules, comprising:

[0061] An adjustment module is used to dynamically adjust the reference value of the test indicator based on changes in network status;

[0062] The conversion module is used to convert the monitoring data into a confidence distribution form using the adjusted reference value, and convert the test indicators into corresponding evidence forms according to the IF-Then mapping relationship;

[0063] The fusion module is used to fuse the evidence corresponding to multiple indicators based on the evidence reasoning rules to obtain the evaluation results;

[0064] The optimization module is used to optimize the model parameters by setting the constraint conditions and optimization objectives of the parameters and using the optimization algorithm.

[0065] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0066] The network performance evaluation method based on dynamic evidence reasoning rules provided by the present invention dynamically adjusts the reference value of the test index, integrates the evidence corresponding to multiple indicators based on the evidence reasoning rules, comprehensively considers the weight, reliability and confidence of each indicator, converts the monitoring data into a confidence distribution form and further into an evidence form, provides a clear quantitative basis for the evaluation results, and makes the evaluation process and results have strong interpretability. In the parameter optimization process, by setting reasonable constraints, the optimal balance between evaluation accuracy and interpretability of evaluation results is achieved. The method can more accurately reflect the actual performance of the network in different scenarios according to the changes in the network state, ensure that the evaluation results are comprehensive and accurate, have strong adaptability and flexibility, and can flexibly set and adjust the evaluation parameters according to actual needs to meet the evaluation needs of different network environments and business types. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0068] Figure 1 A flowchart of a network performance evaluation method based on dynamic evidence reasoning rules provided in an embodiment of the present application;

[0069] Figure 2 A schematic diagram of a stable network state provided in an embodiment of the present application;

[0070] Figure 3 A schematic diagram of a slowly changing state of a network provided in an embodiment of the present application;

[0071] Figure 4 A schematic diagram of a network step state provided in an embodiment of the present application;

[0072] Figure 5 A schematic diagram of a certain campus network indicator observation data provided in an embodiment of the present application;

[0073] Figure 6 A schematic diagram of the evaluation results of a certain campus network after optimization provided in an embodiment of the present application. DETAILED DESCRIPTION

[0074] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0075] The terms "first", "second", "third", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0076] Figure 1 A flow chart of a network performance evaluation method based on dynamic evidence reasoning rules provided in an embodiment of the present application is as follows: Figure 1 As shown, a network performance evaluation method based on dynamic evidence reasoning rules provided in an embodiment of the present application includes:

[0077] Dynamically adjust the reference value of the test indicator based on changes in network status;

[0078] The adjusted reference values ​​are used to convert the monitoring data into a confidence distribution form, and the test indicators are converted into corresponding evidence forms according to the IF-Then mapping relationship;

[0079] Based on the evidence reasoning rules, the evidence corresponding to multiple indicators is integrated to obtain the evaluation results;

[0080] By setting the constraint conditions and optimization objectives of the parameters, the optimization algorithm is used to optimize the model parameters.

[0081] Specifically, the network traffic, bandwidth utilization and other key indicators are monitored in real time to obtain a series of monitoring data. According to the changes in the network status, these data are divided into different time periods for analysis. For example, during working hours, the traffic is large and the network is in a slowly changing state; while at night, the traffic is small and the network tends to be stable. For data in a stable state, its mean and variance can be calculated to determine the reference point and fluctuation range of the reference value. For data in a slowly changing state, its change trend can be analyzed and a dynamic reference value adjustment strategy can be set, such as gradually increasing the reference value according to the traffic growth rate. By dynamically adjusting the reference value, the performance of the network in different states can be more accurately reflected, which helps to promptly discover potential performance problems, such as abnormal throughput decline caused by equipment failure.

[0082] After the dynamic reference value is determined, the monitoring data is compared with the reference value to calculate its confidence relative to different reference levels. Taking delay (TD) as an example, assuming that its reference value is divided into three levels: good, medium, and poor, the delay monitored at a certain moment is calculated according to its closeness to the reference value. Its confidence relative to the good level, confidence relative to the medium level, and confidence relative to the poor level. The TD indicator and its confidence are expressed as evidence, where good, medium, and poor are different states of the system output, and the confidence is assigned to each state. The monitoring data is converted into a confidence distribution form and further converted into an evidence form, which provides a basis for subsequent evidence fusion. In this way, the confidence level of the indicator at different performance levels can be quantified, making the evaluation process more scientific and accurate.

[0083] Arrange the evidence of all indicators in order, such as TD, packet loss rate (PLR), TP and other indicators, and fuse them using evidence reasoning rules according to the weight, reliability and confidence of each indicator. By calculating the product of the weight and confidence of each evidence, and handling conflicts between different evidences, the confidence of the fusion result relative to the three performance levels of high, medium and low at a certain moment is obtained. Through evidence fusion, the impact of multiple indicators on network performance can be comprehensively considered to obtain a more comprehensive and accurate evaluation result. For example, at a certain moment, although the delay indicator performs well, the packet loss rate is high. Through fusion, the contribution of each indicator can be balanced to avoid the excessive impact of a single indicator on the evaluation result, ensuring that the evaluation result can truly reflect the overall performance status of the network.

[0084] The optimization goal is set to minimize the mean square error (MSE), that is, the error between the evaluation result and the actual state of network performance. Preset the weights for the evidence reliability, indicator weight, and confidence respectively. Using optimization algorithms, such as genetic algorithms or particle swarm optimization algorithms, to adjust the weights, reliability, and confidence of each indicator under the premise of meeting the constraints, so as to minimize the MSE, can better reflect the changes in network performance, thereby improving the evaluation accuracy. Through parameter optimization, the accuracy and stability of the evaluation model can be further improved. The optimized model parameters are more in line with the operating rules of the actual network, making the evaluation results more reliable.

[0085] The network performance evaluation method based on dynamic evidence reasoning rules provided in the embodiment of the present application dynamically adjusts the reference value of the test index, integrates the evidence corresponding to multiple indicators based on the evidence reasoning rules, comprehensively considers the weight, reliability and confidence of each indicator, converts the monitoring data into a confidence distribution form and further into an evidence form, and provides a clear quantitative basis for the evaluation results, so that the evaluation process and results have strong interpretability. In the parameter optimization process, by setting reasonable constraints, the optimal balance between evaluation accuracy and interpretability of evaluation results is achieved. This method can more accurately reflect the actual performance of the network in different scenarios according to the changes in network status, ensure that the evaluation results are comprehensive and accurate, have strong adaptability and flexibility, and can flexibly set and adjust the evaluation parameters according to actual needs to meet the evaluation needs of different network environments and business types.

[0086] Optionally, in the network performance evaluation method based on dynamic evidence reasoning rules provided in an embodiment of the present application, the network state includes the network being in a stable state, the network being in a slowly changing state, and the network being in a step-change state.

[0087] Specifically, in an actual network environment, the network status will change due to a variety of factors. For example, in a data center network, when the number of users is small and the business traffic is stable, the network is in a stable state; when a large number of users access the network at the same time and the business traffic gradually increases, the network is in a slowly changing state; when a large-scale event causes a sudden surge in the number of users and business traffic, the network is in a step-change state. By monitoring network traffic, bandwidth utilization and other indicators in real time, it is possible to determine which state the network is in and adopt corresponding evaluation strategies based on different states.

[0088] Optionally, the network performance evaluation method based on dynamic evidence reasoning rules provided in the embodiment of the present application, wherein the reference value of the test indicator is dynamically adjusted based on the change of the network state, specifically includes:

[0089] When the network is in a stable state, there are a total of T 1 observation time, and the monitoring data corresponding to each observation time is expressed as Its mean is expressed as

[0090]

[0091] Calculate the distance between each monitoring data and the mean have

[0092]

[0093] Based on the calculated maximum distance and minimum distance For the N reference levels G of this indicator n ,n=1,2,…,N, the step amount Δd i Expressed as

[0094]

[0095] Arrange the reference values ​​in ascending order. When N is an odd number, the index reference value is

[0096]

[0097] When N is an even number, the reference value of the indicator is

[0098]

[0099] Specifically, when the network is in a stable state, the monitoring data x i (t) A trend around a constant, such as Figure 2 As shown, at this time, the reference value of the indicator should be adjusted based on this constant.

[0100] Assume that there are M test indicators in the network, and the monitoring data of each indicator at time t is represented by x i (t),i=1,2,…,M,t=1,2,…,T, where T represents the number of all observation moments.

[0101] When the network is in a stable state, continuous data collection is performed on a key indicator, and these data are sorted to obtain a data sequence. The mean of the collected data sequence is calculated as the benchmark reference value of the indicator in a stable state. The distance between each observed data and the mean is calculated, that is, the difference between each data point and the mean. From all the calculated distances, find the maximum distance and the minimum distance. The maximum distance indicates the maximum degree of deviation of the data point from the mean, and the minimum distance indicates the minimum degree of deviation of the data point from the mean. The step amount is calculated according to the number of reference levels of the indicator. The step amount is determined based on the maximum distance and the minimum distance and is used to divide the intervals between different reference levels. Arrange the reference values ​​in ascending order, and calculate the reference values ​​of different reference levels based on the step amount and the mean. When N is an odd number, the indicator reference value is centered on the mean and expands to both sides; when N is an even number, the reference value is calculated in a slightly different way, but it is also determined based on the mean and the step amount. According to the above process, h i (G n ) can fully reflect the indicator x i Changing trends at different stages.

[0102] By calculating the mean as the benchmark reference value, the evaluation results can better represent the overall performance of the network in this state, avoiding the significant impact of fluctuations in individual data points on the evaluation results. By calculating the distance, determining the maximum and minimum distances, and calculating the step amount, the boundaries between different performance levels can be reasonably divided. It can reflect the subtle changes in network performance more carefully and provide clear quantitative standards for different levels of performance status. Set different numbers of reference levels according to actual needs, and adjust the calculation method of step amounts and reference values ​​accordingly. This flexibility enables the evaluation method to adapt to different evaluation accuracy requirements and application scenarios, improving the applicability and practicality of the evaluation method.

[0103] Optionally, the network performance evaluation method based on dynamic evidence reasoning rules provided in the embodiment of the present application, wherein the reference value of the test indicator is dynamically adjusted based on the change of the network state, specifically includes:

[0104] When the network is in a slowly changing state, there is a total of T 2 observation time, and the monitoring data corresponding to each observation time is expressed as Adjacent moments The amount of change Expressed as

[0105]

[0106] Will As a new observation sequence, the indicator x is calculated using the same calculation method as when the network is in a stable state. iThe reference value of the indicator under this new observation sequence Will The indicator reference value restored to the original sequence

[0107] Specifically, when the network is in a slowly changing state, the monitoring data x i (t) Shows a regular increasing or decreasing trend, such as Figure 3 At this point, the data can be preprocessed and x i The change in (t).

[0108] When the network is monitored to be in a slowly changing state, continuous data collection is performed on key indicators. Since the data shows a slowly changing trend, the data needs to be preprocessed to highlight its changing characteristics. The change amount of the data at adjacent observation times is calculated, that is, the difference between each data point and its previous data point. This sequence reflects the rate of change of the delay in a slowly changing state. The change amount sequence is used as the new observation data, and the reference value of the indicator under the sequence is calculated using the same calculation method as the stable state. First, the mean of the change amount sequence is calculated, and then the distance between each change amount and the mean is calculated to determine the maximum and minimum distances. Then the step amount is calculated, and the reference values ​​of different reference levels are determined based on the step amount and the mean. The calculated reference value is restored to the original sequence to obtain the dynamic reference value of the delay indicator in a slowly changing state.

[0109] By calculating the amount of change in data at adjacent moments, the performance change trend of the network in a slowly changing state can be accurately captured. The sequence of changes is treated as new observation data, so that the evaluation method can adapt to the characteristics of slow changes in the network. In a slowly changing state, the data changes in amplitude are small, but the duration is long. By analyzing the amount of change, the stability of network performance can be more accurately evaluated. By restoring the reference value to the original sequence, the reference value of the delay indicator is dynamically adjusted. This dynamic adjustment mechanism enables the reference value to change with the change of the network state, avoiding the evaluation deviation caused by the fixed reference value.

[0110] Optionally, the network performance evaluation method based on dynamic evidence reasoning rules provided in the embodiment of the present application, wherein the reference value of the test indicator is dynamically adjusted based on the change of the network state, specifically includes:

[0111] When the network is in a step change state, the step change state is decomposed into a stable state and a slowly changing state according to the change of the state, and the dynamic reference value of the indicator is calculated according to the stable state and the slowly changing state respectively.

[0112] Specifically, when the network is in a step change state, the monitoring data x i (t) shows a drastic change trend, such as Figure 4 This state can be considered as a combination of the above two states. For example, x i (t) A sudden change from a stable state to a slowly changing state. In this state, the calculation method of the dynamic reference value of the indicator can refer to the stable state and the slowly changing state.

[0113] When monitoring network performance, data of key indicators are collected in real time, and the network is identified to be in a step change state. For example, within a certain period of time, the data of the TP indicator suddenly changes from a stable state to a slowly changing state. This mutation is a step change. The step change state is decomposed into two stages: a stable state and a slowly changing state. For the data in the stable state part, the reference value calculation method of the stable state is used to calculate the reference value in the stable state. For the data in the slowly changing state part, the change amount of the data at adjacent moments is calculated to obtain a change amount sequence, and then the reference value in the slowly changing state is calculated using the same method as the stable state. The reference values ​​of the stable state and the slowly changing state are combined to obtain the dynamic reference value of the TP indicator in the step change state, indicating that during the step change process, the reference value of the TP indicator gradually transitions from the level of the stable state to the level of the slowly changing state.

[0114] Optionally, the network performance evaluation method based on dynamic evidence reasoning rules provided in the embodiment of the present application, wherein the adjusted reference value is used to convert the monitoring data into a confidence distribution form, and the test indicator is converted into a corresponding evidence form according to the IF-Then mapping relationship, specifically including:

[0115] The indicator and the corresponding dynamic reference value are described as the confidence distribution form as follows

[0116]

[0117] Among them, δ n,i (t), n=1,2,…,N represents the index x i At time t relative to the reference level G n Confidence level;

[0118] Index x i The corresponding evidence n,i Expressed as

[0119] e n,i :Ifx i (t)ish i (G n ), then y(t)is{(H 1 ,β 1,j,i ),(H 2 ,β 2,j,i )…,(H K ,β K,j,i)}

[0120] Among them, y(t) represents the system output, H k ,k=1,2,…,K represents different output states, β k,n,i ,k=1,2,…,K represents the number of cells assigned to H k confidence level;

[0121] Evidence n,i The weight w n,i Expressed as

[0122]

[0123] in, Indicates the index x i own weight.

[0124] Specifically, based on the above three states, the indicator x i Dynamic reference value h i (G n ) is fully represented. According to the rule-based information transformation method, the input information can be described as a lower confidence distribution. The evidence corresponding to the key indicators is obtained by calculation.

[0125] By calculating the confidence level, the comparison results between the monitoring data and the dynamic reference value are quantified into specific values, so that the performance of the indicator can be presented in a standardized form. The introduction of dynamic reference values ​​enables the evaluation method to be flexibly adjusted according to changes in network status and adapt to dynamic changes in network performance. By converting the monitoring data into evidence, the confidence level of the indicator at different performance levels is comprehensively considered, making the evaluation results more comprehensive and accurate.

[0126] Optionally, the network performance evaluation method based on dynamic evidence reasoning rules provided in the embodiment of the present application, wherein the evidence corresponding to multiple indicators is integrated based on the evidence reasoning rules to obtain the evaluation result, specifically includes:

[0127] Arrange the evidence corresponding to all indicators in order, denoted as e = {e l ; l=1,2,…,L}, the weight of evidence, reliability and confidence are expressed as w={w l ; l=1,2,…,L}, r={r l ; l=1,2,…,L} and β={β k,l ; l=1,2,…,L}, the total number of evidence is L, and the evaluation result is expressed as

[0128]

[0129] Among them, η(t) is an intermediate variable, expressed as

[0130]

[0131] Among them, p k,e(L) (t) represents the confidence of the fusion result at time t relative to level k, p k,e(L) The utility of (t) is u k,e(L) The numerical form of the evaluation result is expressed as

[0132]

[0133] Specifically, collect evidence for each key indicator in the network, arrange the evidence according to the importance or order of the indicator, and assign weights and reliability to the evidence for each indicator. The weight reflects the importance of each indicator in the overall network performance evaluation, and the reliability indicates the credibility of the evidence. According to the evidence reasoning rules, the evidence of multiple indicators is fused. First, the weighted confidence of each evidence is calculated, and then the weighted confidence of all evidence is fused to obtain a comprehensive evaluation result. According to the fused evaluation results, the final state of the network performance is determined. For example, if the confidence of the "medium" state in the evaluation result is the highest, the network performance is judged to be in a medium state.

[0134] Through evidence fusion, the impact of multiple indicators on network performance can be comprehensively considered, avoiding the limitations of a single indicator on the evaluation results. The introduction of weights and reliability makes the evaluation process more refined. Weights are allocated according to the importance of indicators to ensure that key indicators occupy a larger proportion in the evaluation; reliability reflects the credibility of evidence, which helps to reduce the interference of unreliable evidence on the evaluation results, thereby improving the accuracy of the evaluation.

[0135] Optionally, in the network performance evaluation method based on dynamic evidence reasoning rules provided in the embodiment of the present application, the constraint conditions and optimization objectives of setting parameters specifically include:

[0136] The constraints on the parameters are:

[0137] For the reliability of evidence l ,have

[0138] 0 <r l <1

[0139] For the indicator weight w i ,have

[0140]

[0141] in, and Respectively represent the preset w i The upper and lower bounds of

[0142] For the confidence level β k,l ,have

[0143]

[0144] The optimization objectives of the parameters are:

[0145] For real output of network performance Its mean square error MSE is expressed as

[0146]

[0147] Establish optimization goals

[0148]

[0149] Specifically, in the above DER rule-based network performance evaluation model, the parameters to be optimized include the evidence reliability r l , indicator weight w i and the confidence level β k,l In order to make the evaluation results interpretable, the physical meaning of the parameters before and after optimization needs to be kept unchanged, so it is necessary to set reasonable constraints for the parameters in the evaluation model.

[0150] Reliability of evidence l It reflects the ability of the information source to provide evidence results. For the reliability of evidence, the constraint condition is 0 < reliability < 1, ensuring that the reliability is within a reasonable range and avoiding distortion of the evaluation results due to being too high or too low. i It reflects the relative importance of different indicators. For indicator weights, the constraint condition is 0<weight<1, and the sum of the weights of each indicator is 1, ensuring that the weight distribution is reasonable and reflects the relative importance of each indicator in the evaluation. In addition, some extreme weight values ​​need to be avoided. k,l It is a generalized probability. For the confidence level, the constraint is 0≤confidence≤1, ensuring the non-negativity and normalization of the confidence level.

[0151] The optimization goal is set to minimize the mean square error (MSE), that is, the error between the evaluation results and the actual state of network performance. The optimization goal is set to improve the accuracy of the evaluation model and make the evaluation results as close to the actual performance of the network as possible.

[0152] By optimizing the model parameters, the error between the evaluation results and the actual state of network performance is minimized, significantly improving the accuracy of the evaluation model. During the optimization process, the setting of constraints ensures that the physical meaning of the parameters before and after optimization remains unchanged, making the optimized evaluation results clearly interpretable. The introduction of the optimization algorithm enhances the robustness of the evaluation model, enabling it to maintain a high evaluation performance in the face of different network environments and data distributions, adapt to changes in network status and data fluctuations, and avoid evaluation errors caused by unreasonable parameters.

[0153] Optionally, the network performance evaluation method based on dynamic evidence reasoning rules provided in the embodiment of the present application uses an optimization algorithm to optimize the model parameters, including optimizing the model parameters based on parameter constraints and optimization goals, and using a projection covariance matrix adaptive evolution algorithm as an optimization tool.

[0154] Specifically, the projection covariance matrix adaptive evolution algorithm (PCMA-EA) is used as an optimization tool. PCMA-EA is a population-based optimization algorithm that continuously iterates and optimizes model parameters by simulating operations such as mutation, selection, and recombination in the process of biological evolution. The specific steps include parameter initialization, Gaussian sampling, parameter projection, population update, and termination operations.

[0155] Under the premise of meeting the constraints, PCMA-EA is used to optimize the model parameters. First, the population size, optimization step size, and optimization times are initialized; then, the initial population is generated through Gaussian sampling, and the parameters are projected according to the constraints to ensure that the parameters meet the constraints; then, according to the size of MSE and parameter constraints, the population is continuously updated to eliminate inferior individuals and retain high-quality individuals; finally, when the number of optimizations reaches the set threshold or the MSE meets the requirements, the optimization is terminated to obtain the optimized model parameters.

[0156] As an efficient optimization algorithm, PCMA-EA can quickly converge to the optimal solution and shorten the time for model parameter optimization. Compared with traditional manual adjustment or simple optimization methods, PCMA-EA can significantly improve optimization efficiency and accelerate the development and application of evaluation models.

[0157] In a specific embodiment, in order to verify the effectiveness of the proposed method, a case study is carried out using a campus network as an example, including the following steps:

[0158] 1. Establishment of performance evaluation model

[0159] In this embodiment, time delay (TD), packet loss rate (PLR) and throughput (TP) are selected as the performance evaluation indicators of the campus network. By checking the historical operation information of the network, 100 sets of data of the three indicators are obtained, such as Figure 5 shown.

[0160] Combining expert knowledge, we set three evaluation levels of {good, medium, and poor} for the park network and indicator observation data. Figure 5 It can be seen that the observed data of TD and PLR are in a slowly changing state, so the corresponding reference values ​​are static, as shown in Table 1.

[0161] Table 1 Static reference values ​​of indicators

[0162]

[0163]

[0164] In addition, by Figure 5 It can be seen that the observed data of TP is transitioning from a slowly changing state to a stable state, so TP needs to set a dynamic reference value. The setting of the dynamic reference value is shown in Table 2.

[0165] Table 2 Dynamic reference values ​​of indicators

[0166]

[0167] According to Table 2, the reference values ​​of TP are 4, 3, and 2 at the 1st to 60th time points, respectively. Since TP transitions to another state after the 60th time point, its reference value is dynamically updated to 2.8, 2.25, and 1.7. At the same time, it is assumed that the network performance state can be described by three levels: high (High, H), medium (Medium, M), and low (Low, L). A performance evaluation model based on DER is constructed. Since some model parameters cannot be accurately set, parameter optimization of the evaluation model is required.

[0168] 2. Optimization of performance evaluation model

[0169] The above performance evaluation model is optimized, with 50 sets of data used as training data and the rest as test data. The optimized model parameters are shown in Table 3. In addition, the optimized indicator weights are 1, 0.95, and 0.57, respectively. By using the parameters in Table 3, the output results of the model evaluated on the test set are shown in Figure 6 As shown, the corresponding MSE can be calculated as 0.0268. Figure 6 It can be seen that the optimization results basically conform to the actual performance status of the campus network.

[0170] Table 3 Optimized model parameters

[0171]

[0172]

[0173] The embodiment of the present application also provides a network performance evaluation device based on dynamic evidence reasoning rules, including:

[0174] An adjustment module is used to dynamically adjust the reference value of the test indicator based on changes in network status;

[0175] The conversion module is used to convert the monitoring data into a confidence distribution form using the adjusted reference value, and convert the test indicators into corresponding evidence forms according to the IF-Then mapping relationship;

[0176] The fusion module is used to fuse the evidence corresponding to multiple indicators based on the evidence reasoning rules to obtain the evaluation results;

[0177] The optimization module is used to optimize the model parameters by setting the constraint conditions and optimization objectives of the parameters and using the optimization algorithm.

[0178] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0179] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0180] In the several embodiments provided in the present application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the units, which is only a logical function division. There may be other division methods in actual implementation, such as 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 mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0181] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0182] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0183] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the specification and practicing the disclosure here, those skilled in the art will easily think of the implementation scheme of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field not recorded in the present disclosure. The description and examples are regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

[0184] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0185] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A network performance evaluation method based on dynamic evidence reasoning rules, characterized in that: include: Dynamically adjust the reference value of the test indicator based on changes in network status; The adjusted reference values ​​are used to convert the monitoring data into a confidence distribution form, and the test indicators are converted into corresponding evidence forms according to the IF-Then mapping relationship; Based on the evidence reasoning rules, the evidence corresponding to multiple indicators is integrated to obtain the evaluation results; By setting the constraint conditions and optimization objectives of the parameters, the optimization algorithm is used to optimize the model parameters.

2. The network performance evaluation method based on dynamic evidence reasoning rules as claimed in claim 1 is characterized in that: The network status includes the network being in a stable state, the network being in a slowly changing state, and the network being in a step-changing state.

3. The network performance evaluation method based on dynamic evidence reasoning rules as claimed in claim 2 is characterized in that: The dynamically adjusting the reference value of the test indicator based on the change of the network status specifically includes: When the network is in a stable state, there are T1 observation times in total, and the monitoring data corresponding to each observation time is expressed as Its mean is expressed as Calculate the distance between each monitoring data and the mean have Based on the calculated maximum distance and minimum distance For the N reference levels G of this indicator n ,n=1,2,…,N, the step amount Δd i Expressed as Arrange the reference values ​​in ascending order. When N is an odd number, the index reference value is When N is an even number, the reference value of the indicator is 4. The network performance evaluation method based on dynamic evidence reasoning rules as claimed in claim 3 is characterized in that: The dynamically adjusting the reference value of the test indicator based on the change of the network status specifically includes: When the network is in a slowly changing state, there are T2 observation times in total, and the monitoring data corresponding to each observation time is expressed as Adjacent moments The amount of change Expressed as Will As a new observation sequence, the indicator x is calculated using the same calculation method as when the network is in a stable state. i The reference value of the indicator under this new observation sequence Will The indicator reference value restored to the original sequence 5. The network performance evaluation method based on dynamic evidence reasoning rules as claimed in claim 4 is characterized in that: The dynamically adjusting the reference value of the test indicator based on the change of the network status specifically includes: When the network is in a step change state, the step change state is decomposed into a stable state and a slowly changing state according to the change of the state, and the dynamic reference value of the indicator is calculated according to the stable state and the slowly changing state respectively.

6. The network performance evaluation method based on dynamic evidence reasoning rules as claimed in claim 5 is characterized in that: The adjusted reference value is used to convert the monitoring data into a confidence distribution form, and the test indicators are converted into corresponding evidence forms according to the IF-Then mapping relationship, specifically including: The indicator and the corresponding dynamic reference value are described as the confidence distribution form as follows Among them, δ n,i (t), n=1,2,…,N represents the index x i At time t relative to the reference level G n Confidence level; Index x i The corresponding evidence n,i Represented as e n,i :Ifx i (t)ish i (G n ), then y(t)is{(H1,β 1,j,i ),(H2,β 2,j,i )…,(H K ,β K,j,i )} Among them, y(t) represents the system output, H k ,k=1,2,…,K represents different output states, β k,n,i ,k=1,2,…,K represents the number of cells assigned to H k Confidence level; Evidence n,i The weight w n,i Expressed as in, Indicates the index x i own weight.

7. The network performance evaluation method based on dynamic evidence reasoning rules as claimed in claim 5 is characterized in that: The evidence corresponding to multiple indicators is integrated based on the evidence reasoning rule to obtain the evaluation result, which specifically includes: Arrange the evidence corresponding to all indicators in order, denoted as e = {e l ; l=1,2,…,L}, the weight of evidence, reliability and confidence are expressed as w={w l ; l=1,2,…,L}, r={r l ; l=1,2,…,L} and β={β k,l ; l=1,2,…,L}, the total number of evidence is L, and the evaluation result is expressed as Among them, η(t) is an intermediate variable, expressed as Among them, p k,e(L) (t) represents the confidence of the fusion result at time t relative to level k, p k,e(L) The utility of (t) is u k,e(L) The numerical form of the evaluation result is expressed as 8. The network performance evaluation method based on dynamic evidence reasoning rules as claimed in claim 7 is characterized in that: The constraint conditions and optimization objectives of the setting parameters specifically include: The constraints on the parameters are: For the reliability of evidence l ,have 0<r l <1 For the indicator weight w i ,have in, and Respectively represent the preset w i The upper and lower bounds of For the confidence level β k,l ,have The optimization objectives of the parameters are: For real output of network performance Its mean square error MSE is expressed as Establish optimization goals 9. The network performance evaluation method based on dynamic evidence reasoning rules as claimed in claim 8, characterized in that: The optimization algorithm is used to optimize the model parameters, including optimizing the model parameters based on the constraint conditions and optimization objectives of the parameters and using the projection covariance matrix adaptive evolution algorithm as an optimization tool.

10. A network performance evaluation device based on dynamic evidence reasoning rules, characterized in that: include: An adjustment module is used to dynamically adjust the reference value of the test indicator based on changes in network status; The conversion module is used to convert the monitoring data into a confidence distribution form using the adjusted reference value, and convert the test indicators into corresponding evidence forms according to the IF-Then mapping relationship; The fusion module is used to fuse the evidence corresponding to multiple indicators based on the evidence reasoning rules to obtain the evaluation results; The optimization module is used to optimize the model parameters by setting the constraint conditions and optimization objectives of the parameters and using the optimization algorithm.

Images