Communication network performance monitoring and optimization method and system

Abstract

This invention relates to the field of communication network optimization technology, and discloses a method and system for monitoring and optimizing communication network performance. The method collects performance data from nodes affecting transmission quality in the communication network to construct a network transmission quality model. It configures a network monitoring probe model and performs performance tests to obtain its latency detection capability and packet loss rate detection accuracy, then integrates these into the network transmission quality model. The total data traffic load within a preset time window is calculated, and the theoretical monitoring load is obtained by combining the probe's latency detection capability. Based on the theoretical monitoring load and packet loss rate detection accuracy, a data sampling frequency control command is generated, dynamically adjusting the operating frequency of the probe data acquisition module accordingly. The optimization effect of this method on network transmission quality is then evaluated. This method, by dynamically adjusting the sampling frequency, achieves precise adaptation to network conditions, improves the real-time performance and resource utilization of monitoring, and integrates multi-dimensional performance indicators, providing comprehensive support for network transmission quality optimization.

Description

Technical Field

[0001] This invention relates to the field of communication network optimization technology, specifically to a method and system for monitoring and optimizing communication network performance. Background Technology

[0002] With the rapid development of communication technology and the continuous expansion of various network application scenarios, from traditional voice communication to high-definition video transmission, IoT data interaction, and cloud computing services, the data traffic carried by communication networks is growing exponentially, and the network structure is becoming increasingly complex. Against this backdrop, the stability and reliability of network transmission quality have become key factors affecting user experience and business continuity. However, existing communication network performance monitoring methods often have many limitations when facing dynamically changing network environments.

[0003] Currently, most network monitoring systems rely on fixed sampling frequencies for data collection, making it difficult to adapt to drastic fluctuations in network traffic. When the network is under high load, a fixed sampling frequency may lead to redundant monitoring data, consuming excessive network resources and hindering the transmission of normal services. Conversely, during low-load periods, insufficient sampling may cause key performance anomalies to be missed, resulting in delayed fault detection. Furthermore, traditional monitoring models often treat performance indicators such as latency and packet loss rate in isolation, lacking a comprehensive consideration of the relationships between various performance nodes. This makes it difficult to construct a comprehensive network transmission quality assessment system, leaving the formulation of optimization strategies without accurate data support.

[0004] Existing monitoring probes suffer from insufficient integration of performance testing with real-world network environments. The probes' latency detection capabilities and packet loss rate detection accuracy are not dynamically correlated with real-time network load, significantly reducing the effectiveness and relevance of monitoring data. These issues collectively result in communication network performance monitoring failing to meet the demands of complex network environments in terms of real-time performance, accuracy, and resource utilization. Therefore, a monitoring optimization method that can dynamically adapt to network conditions and integrate multi-dimensional performance indicators is needed. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for monitoring and optimizing the performance of communication networks, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides a method for monitoring and optimizing the performance of a communication network, the method comprising:

[0007] Collect performance nodes that affect transmission quality in the communication network, and construct a network transmission quality model based on the performance nodes;

[0008] Configure the network monitoring probe model, perform performance testing on the network monitoring probe model, and obtain the latency detection capability and packet loss rate detection accuracy of the network monitoring probe model.

[0009] The network monitoring probe model is integrated into the network transmission quality model;

[0010] Calculate the total data traffic load of the communication network within a preset time window, and combine the latency detection capability of the network monitoring probe model to calculate the theoretical monitoring load of the network monitoring probe model within the preset time window;

[0011] Based on the theoretical monitoring load and packet loss rate detection accuracy, a data sampling frequency control command for the network monitoring probe model is generated.

[0012] According to the data sampling frequency control command, the operating frequency of the data acquisition module of the network monitoring probe model is dynamically adjusted;

[0013] The effectiveness of the network monitoring probe model in optimizing the transmission quality of the communication network was evaluated.

[0014] Preferably, the collection of performance nodes affecting transmission quality in the communication network, and the construction of a network transmission quality model based on the performance nodes, includes:

[0015] Extract the topology information of the communication network and divide the communication network into multiple logical subnets;

[0016] Identify the key performance nodes in each logical subnet and generate a set of performance nodes for the communication network.

[0017] Measure the bandwidth carrying capacity of each performance node and analyze the external interference factors acting on the performance node;

[0018] Based on the bandwidth capacity and external interference factors, a data transmission balance equation for the communication network is established.

[0019] Preferably, the step of establishing the data transmission balance equation of the communication network based on the bandwidth carrying capacity and external interference factors includes:

[0020] Analyze the data transmission coupling effect between performance nodes and quantify the impact intensity of the external interference factors on each performance node;

[0021] The data transmission balance equation is constructed by combining the bandwidth carrying capacity of performance nodes, data transmission coupling effect, and the influence intensity of external interference factors.

[0022] Preferably, configuring the network monitoring probe model and performing performance testing operations on the network monitoring probe model includes:

[0023] The latency response characteristics and packet loss detection accuracy of the network monitoring probe model were tested under high load, low load and fluctuating load conditions to obtain the latency detection capability and packet loss detection accuracy.

[0024] Preferably, integrating the network monitoring probe model into the network transmission quality model includes:

[0025] Analyze the influence weight of each performance node in the network transmission quality model and locate the target performance node with the largest influence weight.

[0026] The network monitoring probe model is deployed in the logical subnet corresponding to the target performance node.

[0027] Preferably, the total data traffic load of the computing communication network within a preset time window includes:

[0028] The total data traffic load is obtained by aggregating the data traffic generation of each performance node within a preset time window.

[0029] The theoretical monitoring load of the computational network monitoring probe model includes:

[0030] Obtain the basic monitoring overhead of the network monitoring probe model within a preset time window, and calculate the theoretical monitoring load by combining it with the total data traffic load;

[0031] The generated data sampling frequency control command includes:

[0032] Based on the theoretical monitoring load and packet loss rate detection accuracy, the optimal sampling frequency of the network monitoring probe model within a preset time window is determined.

[0033] Preferably, the data sampling frequency control instruction includes:

[0034] The data throughput of the network monitoring probe model at a unit sampling frequency is tested, and a frequency switching time plan for the data acquisition module is generated based on the optimal sampling frequency.

[0035] Preferably, the dynamic adjustment of the operating frequency of the data acquisition module of the network monitoring probe model includes:

[0036] The preset time window is divided into discrete time periods, and the frequency switching time plan is evenly distributed to each discrete time period.

[0037] After a single discrete time period ends, verify whether the actual data throughput of the data acquisition module meets expectations;

[0038] Based on the verification results, the frequency switching time plan for subsequent discrete time periods is compensated and corrected until the adjustment operation of the entire preset time window is completed.

[0039] Preferably, the evaluation of the network monitoring probe model's effect on optimizing the transmission quality of the communication network includes:

[0040] Obtain the changes in latency and packet loss rate of the communication network before and after optimization;

[0041] Based on the changes in delay and packet loss rate, the actual monitoring load of the communication network is calculated.

[0042] By comparing the deviation between the actual monitoring load and the theoretical monitoring load, a frequency control authority index for the data acquisition module is generated.

[0043] The network transmission quality optimization effect is evaluated based on the frequency control authority index.

[0044] Preferably, the present invention further includes a communication network performance monitoring and optimization system for implementing the communication network performance monitoring and optimization method described above, the system comprising:

[0045] The transmission quality modeling module is used to collect performance nodes that affect transmission quality in the communication network and construct a network transmission quality model based on the performance nodes.

[0046] The probe performance testing module is used to configure the network monitoring probe model, execute the performance testing operation of the network monitoring probe model, and obtain the latency detection capability and packet loss rate detection accuracy.

[0047] The model fusion module is used to integrate the network monitoring probe model into the network transmission quality model;

[0048] The monitoring load calculation module is used to calculate the total data traffic load of the communication network within a preset time window, and to calculate the theoretical monitoring load in combination with the delay detection capability.

[0049] The sampling control module is used to generate data sampling frequency control instructions for the network monitoring probe model based on the theoretical monitoring load and packet loss rate detection accuracy.

[0050] A dynamic adjustment module is used to dynamically adjust the operating frequency of the data acquisition module according to the data sampling frequency control command;

[0051] An optimization evaluation module is used to evaluate the effect of the network monitoring probe model on the transmission quality optimization of the communication network.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] By collecting data on performance nodes affecting transmission quality and constructing a network transmission quality model, a systematic analysis and integration of key network components was achieved. This breaks through the limitations of isolated performance indicator processing in traditional monitoring, and can more comprehensively reflect the overall network operating status. This multi-node-based model construction method effectively uncovers the correlations between various performance parameters, providing a more realistic analytical foundation for subsequent monitoring and optimization.

[0054] Configuring and performing performance tests on network monitoring probe models to obtain their latency detection capabilities and packet loss rate detection accuracy helps clarify the probe's performance boundaries and applicable scenarios, providing a precise basis for integrating the probe with the network transmission quality model. After integrating the probe model into the transmission quality model, the two form a collaborative mechanism. The probe's monitoring data can directly serve the model's assessment of network status, avoiding data silos and improving the overall integrity and consistency of the monitoring system.

[0055] By calculating the total data traffic load within a preset time window and combining it with the probe's latency detection capability, the theoretical monitoring load is derived. This ensures that the determination of the monitoring load is no longer dependent on empirical values ​​but is closely related to the actual network load. Based on this theoretical load and packet loss rate detection accuracy, data sampling frequency control commands are generated, enabling dynamic adjustment of the sampling frequency. This reduces unnecessary sampling to save resources under high load and increases sampling to capture more details under low load, thereby improving network resource utilization efficiency while ensuring monitoring effectiveness.

[0056] By dynamically adjusting the operating frequency of the data acquisition module, the monitoring system can adapt to changes in network traffic in real time, enhancing the flexibility and adaptability of monitoring and avoiding the monitoring redundancy or insufficiency that may occur with a fixed sampling frequency. Furthermore, the evaluation of the network transmission quality optimization effect verifies the effectiveness of the entire monitoring and optimization process, providing feedback for continuous improvement of the method and forming a closed loop of "monitoring-optimization-evaluation," thereby driving the continuous improvement of communication network performance monitoring levels. Attached Figure Description

[0057] Figure 1 This is a schematic diagram illustrating the working principle of the communication network performance monitoring and optimization method described in this invention.

[0058] Figure 2 A flowchart for performance node acquisition and network transmission quality model construction;

[0059] Figure 3 A flowchart for constructing the data transmission balance equation;

[0060] Figure 4 Flowchart for network monitoring probe model integration;

[0061] Figure 5 A flowchart for generating data sampling frequency control commands. Detailed Implementation

[0062] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0063] Please see Figure 1 This invention provides a method for monitoring and optimizing the performance of a communication network, the method comprising:

[0064] This process involves collecting data on performance nodes affecting transmission quality in the communication network and constructing a network transmission quality model based on these nodes. A network monitoring probe model is configured, and its performance is tested to obtain latency detection capabilities and packet loss rate detection accuracy. The network monitoring probe model is then integrated into the network transmission quality model. The total data traffic load of the communication network within a preset time window is calculated, and combined with latency detection capabilities, the theoretical monitoring load of the network monitoring probe model within the preset time window is calculated. Based on the theoretical monitoring load and packet loss rate detection accuracy, data sampling frequency control instructions for the network monitoring probe model are generated. The operating frequency of the data acquisition module of the network monitoring probe model is dynamically adjusted according to the data sampling frequency control instructions. Finally, the effectiveness of the network monitoring probe model in optimizing the transmission quality of the communication network is evaluated.

[0065] Example 1: See Figure 2 The topology information of the communication network is extracted, and the network is divided into multiple logical subnets by analyzing the connection relationships and routing paths of network devices. The logical subnets are divided based on the physical location of devices, service functions, or traffic boundaries, forming independent data transmission units. Key performance nodes in each logical subnet are identified, and a set of performance nodes is generated based on parameters such as node data forwarding volume, number of connected devices, and historical failure rate. This set includes key network devices such as router core nodes, gateway aggregation nodes, and backbone switch nodes. The bandwidth capacity of each performance node is measured, and actual data transmission tests are performed, recording the maximum throughput rate of the node under saturation. External interference factors affecting performance nodes are analyzed, including dynamic factors such as electromagnetic environmental noise, adjacent channel crosstalk, and sudden high-priority service preemption. A spectrum analyzer is used to collect environmental interference signal characteristics, and the trigger frequency of abnormal traffic samples is statistically analyzed.

[0066] Based on bandwidth capacity and external interference factors, a data transmission balance equation for the communication network is established. The interaction relationships among the parameters in the data transmission balance equation are analyzed, and a mathematical model describing the relationship between node resource allocation and interference cancellation is constructed. Data transmission status is represented using matrix equations, where matrix rows correspond to logical subnets, and matrix columns correspond to performance node attributes. Filled values ​​include measured bandwidth, interference attenuation coefficients, and node buffer capacity. After deployment of the logical subnets, actual data transmission path parameters are collected using packet tagging technology, covering hop count, path switching frequency, and cross-subnet latency. When verifying the matrix equations, the values ​​of each variable are initialized synchronously, and the logical subnet topology constraints are input.

[0067] Obtain the changes in latency and packet loss rate of the communication network before and after optimization. Perform network status snapshots of equal duration before and after probe deployment, compare the end-to-end latency average difference for the same service flow, and calculate the absolute change value. Statistically analyze the change in the number of lost data packets from the same source and destination within a preset packet loss detection period to generate the relative change in packet loss rate. Calculate the actual monitoring load of the communication network, which is determined by the total network resources occupied by the probe model during operation, including monitoring data replication bandwidth, analysis and processing overhead, and result reporting traffic. Compare the deviation between the actual monitoring load and the theoretical monitoring load, quantify the dispersion using the standard deviation algorithm, and match a preset correction coefficient table based on the deviation range.

[0068] The frequency control permission index for the data acquisition module is generated. The permission index is calculated by weighting three parameters: deviation value, network service type weight, and current load urgency, outputting a normalized value between 0 and 1. The index generation process includes initializing weight coefficients, loading the subnet service priority configuration table, and performing multi-source data fusion operations. The optimization effect is evaluated based on the frequency control permission index; the permission index is positively correlated with the transmission quality optimization effect, and a higher index value indicates a more significant improvement in network performance by the probe model. This process involves defining a coefficient update mechanism for the data transmission balance equation, automatically triggering equation parameter recalibration when the external interference intensity changes beyond a threshold. The dynamic weight adjustment of external interference factors uses a sliding window algorithm, updating the weight allocation ratio based on historical interference impact records.

[0069] When constructing the network transmission quality model, the correlation between nodes in subnets is considered, logical subnet boundary nodes with data dependencies are marked, and cross-subnet transmission compensation parameters are added to the data transmission balance equation. The model covers all performance node sets, and an independent status monitoring unit is established for each node to upload node traffic load and bit error rate data in real time. When evaluating the optimization effect, resource allocation priority is adjusted according to the frequency control authority index, with high authority index areas receiving priority probe resource allocation. The stability changes of network transmission are quantified, and indicators such as transmission path jitter rate, retransmission request frequency, and bandwidth utilization fluctuation amplitude are extracted to establish a baseline for stability changes. During the model integration phase, bandwidth carrying capacity measurements are calibrated synchronously, and bandwidth stress tests are periodically performed to refresh the node's maximum throughput record. The calibration operation is correlated with the analysis results of external interference factors, and a bandwidth retest mechanism is triggered when a new type of interference is added.

[0070] After generating the frequency control permission index, deviation threshold detection is performed, and two levels of deviation alarm thresholds are set. The primary threshold triggers model parameter fine-tuning instructions, while the advanced threshold activates the topology refactoring process. Model parameter re-optimization requirements are managed through configuration version numbers, marking the set of nodes to be optimized and pushing it to the network management platform. The solution results of the data transmission balance equation are compared with the actual network state; if three consecutive abnormal deviations occur, the model output is frozen, and a manual intervention verification process is initiated. After the logical subnetting results are implemented, the traffic balance between subnets is continuously monitored; when the cross-subnet traffic ratio exceeds the design threshold, subnet boundary refactoring is triggered. Dynamic maintenance of the performance node set includes new node admission detection and automatic removal of failed nodes; node status changes are synchronized to the data transmission balance equation variable set in real time.

[0071] Example 2: See Figure 3 This process identifies data transmission coupling effects between performance nodes. It analyzes network routing protocols and traffic engineering strategies to map data flow dependencies between nodes. Coupling effect analysis includes three correlation modes: hard coupling from direct physical connections, soft coupling from shared transmission links, and logical coupling caused by cross-device business logic dependencies. Actual forwarded data packet sequences between nodes are collected, and source-destination node pairing frequency and path overlap parameters are extracted. Each coupling mode is quantified: hard coupling records physical port rate matching, soft coupling measures shared buffer contention ratio, and logical coupling analyzes application layer session binding strength. A correlation strength matrix is ​​output, with rows and columns representing performance node indices, and matrix element values ​​reflecting the frequency and depth of interaction between nodes.

[0072] The impact of external interference factors on each performance node is quantified. First, a classification system for external interference factors is established, including four main categories: environmental electromagnetic interference, equipment heat dissipation fluctuations, concurrent protocol conflicts, and management command interruptions. For each type of interference source, an intensity measurement scheme is designed: environmental electromagnetic interference uses a field strength probe to capture radiation values ​​in a specific frequency band; heat dissipation fluctuations are recorded by recording the correlation curve between temperature sensor data and fan speed; protocol conflicts are detected by detecting the number of ARP broadcast storm triggers; and management command interruptions are analyzed by analyzing the CPU cycles occupied by SNMP polling. The raw measurements are normalized to an impact intensity coefficient in the range of 0-1, where 0 represents no impact and 1 represents the maximum interference state. A time decay factor is introduced into the intensity coefficient calculation, giving higher weight to recent interference events.

[0073] A data transmission balance equation is constructed by combining the bandwidth capacity of performance nodes, data transmission coupling effects, and the intensity of external interference factors. Bandwidth capacity serves as the basic input vector, with each element corresponding to the node's measured maximum throughput. Before construction, the execution vector is standardized to eliminate dimensional differences between nodes. The coupling effect matrix is ​​multiplied by the bandwidth vector to generate the node load transfer result, reflecting the additional pressure each node bears from forwarding by other nodes. The intensity of external interference factors exists in the form of a diagonal matrix, with the main diagonal elements representing the independent interference coefficients of each node. These coefficients are multiplied by the load transfer result to produce an interference-corrected load distribution.

[0074] The equation construction process forms a three-layer superimposed structure: the first layer is the bandwidth base capacity layer, the second layer is the coupling diffusion layer, and the third layer is the interference attenuation layer. The structure output vector is the predicted value of the data transmission balance state, and the vector dimension is consistent with the number of performance nodes. In the numerical initialization phase, the measured baseline value of the bandwidth is loaded, the initial cells of the coupling matrix are set with equal weight coefficients, and the interference coefficient is preset based on the historical average value. Data processing is performed in steps. First, the product of the bandwidth vector and the coupling matrix is ​​calculated to generate an intermediate variable. This intermediate variable is then multiplied element-wise with the interference diagonal matrix to output the final balance state prediction value. After each calculation, the output vector boundary is checked to limit the value range to within the physical capabilities of the nodes.

[0075] The transmission balance equation construction includes defining boundary constraints. A minimum available bandwidth threshold is set for each performance node, dynamically adjusted based on the critical service traffic demands of the node. A maximum interference tolerance limit is set; interference exceeding this limit triggers an alarm mechanism. Boundary conditions are treated as hard constraints during equation solving, enforced using the Lagrange multiplier method. During construction, actual node traffic data is continuously collected and compared with the equation predictions. When deviations exceed allowable limits, a parameter optimization mechanism is activated: coupling matrix weights are dynamically refreshed based on actual traffic path changes, interference coefficients are recalibrated based on real-time monitoring data, and the bandwidth baseline value is updated according to a preset periodic stress test result.

[0076] During iterative equation solving, the coupling effect weights are adjusted based on the actual data flow efficiency between nodes. When the forwarding delay from a node to other nodes continues to increase, its corresponding coupling effect coefficient is reduced. When a node has path redundancy backups, its coupling relationship is split into multiple alternative paths. The adaptive adjustment magnitude of the weights is proportional to the rate of traffic change, while a maximum adjustment step size limit is set to prevent oscillations. The quantification results of external interference factors are directly connected to the real-time alarm stream of the network management system. When the interference intensity suddenly exceeds the preset gradient threshold, the equation solving process is immediately frozen, and calculation is resumed after the interference event analysis report is received.

[0077] After the equations are generated, simulated traffic stress tests are run on an independent verification platform. A virtual topology mirror environment is constructed, and multiple sets of mixed service flow data packets are injected. The status data of each node in the simulated environment is recorded, including buffer queue length, transmission delay distribution, and packet loss event locations. The recorded data is compared with the equation prediction output using trend matching analysis to verify the accuracy of the equation's predictions. For nodes with high prediction deviations, the process of identifying their coupling relationships and interference quantification data is traced back, and location and repair operations are performed. The validated equation parameters are solidified into a configuration file and synchronized to all associated network management components. In subsequent operating cycles, the configuration version number is bound to the network change record to support historical status retrospective analysis. The solution results of the data transmission balance equation serve as the decision input for the deployment location of the network monitoring probe model, and also provide load distribution benchmark parameters for sampling frequency control.

[0078] Example 3: See Figure 4 The performance testing of the network monitoring probe model is performed by simulating network environments under different load conditions. Load conditions are divided into three types: high load, low load, and fluctuating load, each with corresponding traffic characteristics. High load simulates network traffic approaching or reaching physical link capacity, low load reflects idle or lightly loaded network conditions, and fluctuating load exhibits drastic, periodic or random traffic changes. Before testing, a traffic generator is configured, and baseline parameters are set, including packet size distribution, sending interval, and protocol type ratio. The traffic generator is connected to the monitoring interface of the probe model under test, forming a closed-loop test environment.

[0079] In high-load testing, the traffic generator sends 90% to 95% of the maximum allowed data volume at a constant rate for a sufficient duration to allow the network to stabilize. The timestamp sequence of data packets captured by the probe model is recorded, and the statistical distribution of end-to-end latency is calculated. Latency detection capability is characterized by the following metrics: average latency reflects overall response speed, latency standard deviation reflects detection stability, and the 99th percentile latency indicates the performance boundary under extreme conditions. Packet loss detection accuracy testing employs an active tagging method, adding detection tags to a specific proportion of data packets at the sending end, and statistically analyzing the success rate and false positive rate of tag matching at the receiving end.

[0080] The low-load test reduces the transmission rate of the traffic generator to below 10% of the link capacity, maintaining this level for at least the same duration as the high-load test cycle. During this test, the basic processing latency of the probe model and its resource consumption in idle states are primarily observed. The minimum detection latency of the probe model under contention-free conditions is recorded as a performance benchmark. The packet loss detection accuracy test injects a controllable number of abnormal packet loss events under low-load conditions to verify the probe's sensitivity to sparse packet loss.

[0081] Fluctuating load testing employs a composite traffic pattern, with base traffic maintained at 30% to 50% of link capacity, superimposed with periodic burst traffic pulses. Pulse amplitude varies between 20% and 80% of link capacity, with pulse widths ranging from milliseconds to seconds. Pulse parameters are dynamically adjusted during testing to cover different combinations of frequency and amplitude. Latency response characteristics are described using latency variation curves within the pulse period, focusing on capturing response lag at the rising and falling edges. Packet loss detection accuracy testing injects packet loss events during pulse peaks to analyze the probe's detection consistency under different load change rates.

[0082] The influence weights of each performance node in the network transmission quality model are determined through multi-dimensional evaluation. Evaluation parameters include the proportion of traffic handled by the node, the service criticality of connected devices, historical fault records, and the importance of its topological location. A judgment matrix is ​​constructed using the analytic hierarchy process (AHP), and weight values ​​are calculated from eigenvectors. A consistency check is introduced during the weight calculation process; when the consistency ratio of the judgment matrix exceeds a threshold, the parameter scaling is readjusted.

[0083] The target performance node with the greatest impact is located using an iterative screening method. The first round of screening retains nodes whose weight values ​​are more than twice the standard deviation of the average weight. The second round compares the traffic processing capacity and fault impact range of these nodes to ultimately determine the node with the highest weight. The logical subnets corresponding to the target nodes are automatically associated based on the network topology, and the subnet boundaries are jointly defined by routing policies and VLAN partitioning.

[0084] Deploying the network monitoring probe model in the logical subnet corresponding to the target performance node involves two levels: physical connection and logical configuration. Physical connection ensures that the probe's monitoring port is connected to the core switch mirror port or splitter of the target subnet to acquire complete data traffic. Logical configuration includes setting filtering rules to focus on traffic related to the target node, adjusting the sampling rate to match subnet traffic characteristics, and configuring the reporting path to avoid critical forwarding links of the target node. After deployment, baseline testing is performed to verify that the probe's monitoring range covers all critical interfaces of the target node. The latency detection capability of the probe model is quantified using the following formula:

[0085] in, This represents the mean square error of the delay detection, where N is the total number of test samples. It is the delay value of the i-th data packet recorded by the probe. This represents the actual time delay value measured by the benchmark system. This formula is specifically used for time delay detection accuracy evaluation in this embodiment and has no sign overlap with the mathematical expressions in other embodiments. The calculation result is used to calibrate the internal clock compensation parameters of the probe, reducing systematic time delay measurement bias.

[0086] The accuracy of packet loss rate detection is evaluated using a receiver-side verification method. The sending end records complete packet sequence numbers, and the receiving end calculates the proportion of missing sequence numbers as the true packet loss rate, comparing it periodically with the packet loss rate reported by the probe. The accuracy metrics include two dimensions: false negative rate and false positive rate. The false negative rate reflects the proportion of true packet losses not detected by the probe, while the false positive rate is the proportion of normal packets that the probe mistakenly identifies as lost. During testing, the network packet loss rate is dynamically adjusted to cover different intensity ranges from 0.1% to 10%.

[0087] Monitoring data from target performance nodes is transmitted to the analysis system via an independent channel, avoiding additional load on the existing network. Data transmission employs compression and differential coding techniques to reduce monitoring overhead. The analysis system integrates latency and packet loss data reported by probes, combines it with network topology information to generate a quality heatmap, visually displaying the performance status of target nodes and their associated paths. The heatmap update frequency is synchronized with the probe sampling rate, and real-time alarms are triggered when changes in important parameters exceed thresholds.

[0088] During long-term operation of the target subnet, the probe model continuously adapts to changes in network traffic patterns. Adaptive mechanisms include dynamically adjusting sampling strategies, optimizing detection algorithm parameters, and learning a normal latency baseline. When a network upgrade or configuration change is detected, a calibration process is automatically triggered to re-establish the performance baseline. The deployment effectiveness of the probe is indirectly evaluated by comparing the fault recovery time and performance fluctuation of the target node before and after deployment, while simultaneously monitoring whether the probe's own resource consumption remains within the design range.

[0089] Traffic characteristic analysis of the logical subnet provides the basis for probe model configuration. Deep packet inspection (DPI) technology is used to identify the main application protocols and their timing characteristics within the subnet, setting differentiated detection sensitivities for different protocol types. Fine-grained latency monitoring is employed for real-time voice and video traffic with high requirements, while packet loss rate statistics are emphasized for batch data transmission traffic. Protocol analysis results are also used to optimize probe filtering rules, reducing the processing overhead of irrelevant traffic. Periodic re-analysis of subnet traffic characteristics ensures that probe configuration is updated synchronously with actual network usage patterns.

[0090] The target node's performance data continuously interacts with the network transmission quality model. Real-time metrics collected by the probe are input into the model for state prediction, and the model's output guides the probe to adjust its monitoring focus. When the model predicts a performance degradation on a certain path, the probe increases the monitoring density for that path; when the model detects an abnormal pattern, the probe initiates detailed diagnostic data collection. This two-way interaction forms a closed-loop optimization system, concentrating monitoring resources on the areas of the network that require the most attention. The frequency of data exchange is strictly limited during the interaction process to prevent the monitoring activity itself from becoming a source of network load.

[0091] Example 4: See Figure 5 Within a preset time window, the data traffic generation of each performance node in the communication network is collected through a distributed counter. Each node deploys a lightweight traffic statistics agent to record the number and bytes of inbound and outbound data packets at a fixed time granularity. The time window is divided using a sliding window mechanism, with the window length dynamically adjusted according to the network size, typically ranging from 5 minutes to 1 hour. At the end of the window, the traffic statistics agent aggregates the data and transmits it to the aggregation node via a compression protocol. The aggregation node performs deduplication verification to eliminate duplicate reporting caused by network latency and merges the traffic data from all performance nodes to generate the total data traffic load. This load value includes two dimensions: the total number of data packets and the total number of bytes, used for different types of monitoring load calculations.

[0092] The basic monitoring overhead of the network monitoring probe model was determined through offline calibration experiments. The calibration process was conducted in an isolated test environment to eliminate the impact of network fluctuations. The CPU utilization, memory consumption, and network throughput of the probe model were measured under zero load; these inherent overheads constitute the basic monitoring overhead. The basic monitoring overhead in actual operation also needs to be adjusted for network environment factors, including dynamic parameters such as protocol parsing complexity and encrypted traffic processing overhead. The overhead data is stored in a configuration file for direct access in different scenarios.

[0093] The theoretical monitoring load is calculated by combining the total data traffic load and latency detection capability. The latency detection capability parameter reflects the efficiency of the probe model in processing unit traffic; this parameter was obtained in the performance test of Example 3. The calculation process decomposes the total data traffic load according to protocol type, matching different types with corresponding processing coefficients. For example, TCP traffic is weighted by the number of connections, and UDP traffic is weighted by the packet rate. The weighted traffic value is added to the basic monitoring overhead and then multiplied by the latency detection capability factor to output the theoretical monitoring load value. The load value is continuously updated according to a time window, forming a load change curve.

[0094] Determining the optimal sampling frequency requires balancing monitoring accuracy and resource consumption. Using the theoretical monitoring load as input, a matching interval is searched in a pre-defined load-frequency mapping table. This mapping table, predefined according to the probe model, contains discrete load intervals and corresponding recommended sampling frequencies. When the theoretical monitoring load falls at the boundary between two intervals, a linear interpolation method is used to calculate the intermediate value. After determining the initial sampling frequency, a packet loss rate detection accuracy constraint is superimposed. The higher the accuracy requirement, the greater the upward adjustment of the sampling frequency, but it cannot exceed the maximum sampling rate supported by the probe hardware. The final optimal sampling frequency is the maximum value that satisfies all constraints.

[0095] The data sampling frequency control command is generated using a structured data format. The command includes fields such as time window number, target sampling frequency, effective timestamp, and checksum. The command queue is arranged in time window order, with each window corresponding to an independent command. When switching windows, the new control command is sent to the probe model through a secure channel. The probe's configuration management module parses and executes the frequency switch. The command transmission process employs a transaction mechanism, automatically retransmitting if no confirmation response is received, ensuring reliable completion of configuration changes.

[0096] Table 1: Aggregation of performance node traffic data within a typical time window.

[0097] The table data comes from anonymized samples of a real network operating environment, showing traffic statistics for three performance nodes within a 15-minute time window. The number of inbound and outbound packets reflects the workload processed by the nodes, the number of bytes is used to calculate bandwidth usage, and the protocol type distribution affects probe processing strategies. The summary row displays the total data traffic load for this window, serving as an input benchmark for subsequent calculations.

[0098] The probe model's dynamic frequency adjustment mechanism enables adaptive resource allocation. When the theoretical monitoring load indicates that a higher sampling frequency is needed for a certain time window, the probe automatically allocates more computing resources to the data acquisition module, potentially temporarily reducing resource allocation for non-critical functions. The frequency adjustment process is smooth, avoiding data discontinuity caused by abrupt changes in sampling intervals. A snapshot of the current sampling state is saved during frequency switching to ensure that business monitoring is not affected by configuration changes. Under extremely high load conditions, the probe initiates a degradation mode, prioritizing the integrity of core indicator collection.

[0099] The time window segmentation strategy affects the timeliness of monitoring and resource consumption. Shorter windows can reflect network status changes faster but increase system overhead; longer windows save resources but may miss important events. Actual deployment employs a multi-level time window system: a fixed-length window at the basic level to ensure routine monitoring, and a variable-length window at the event-triggered level to handle emergencies. Window switching is aligned with the statistical periods of network devices to avoid data inconsistencies caused by time deviations.

[0100] Protocol type distribution data guides sampling strategy optimization. Sampling of TCP traffic focuses on tracking connection establishment and closure processes, retransmission events, and window size changes; UDP traffic monitoring emphasizes packet interval jitter and loss pattern identification. These differentiated sampling requirements are achieved through additional parameters in the control commands, enabling the probe to employ the most effective monitoring methods for different protocols. When protocol distribution changes exceed a threshold, the sampling strategy is recalculated to maintain the match between the monitoring method and the current traffic characteristics.

[0101] Analyzing the discrepancy between theoretical and actual operating loads continuously optimizes parameter settings. A larger load margin is allowed in the initial deployment phase, gradually narrowing the error range of load forecasting as operational data accumulates. The load calculation model is periodically retrained with the latest network data to adapt to network architecture changes and evolving business models. A load pattern library developed over a long period supports predictive resource allocation, pre-adjusting probe configurations before anticipated high-load time windows.

[0102] The effectiveness of sampling frequency control commands is evaluated through a closed-loop feedback mechanism. After each frequency switch, the probe reports the actual resource utilization and data acquisition quality indicators. The control center compares the deviation between the expected results and the actual results and adjusts the generation parameters of subsequent commands. This adaptive adjustment allows the system to gradually approach its optimal operating state, reducing the need for manual intervention. The feedback data is also used to construct probe performance profiles and identify the best operating mode under specific load conditions.

[0103] The dynamic balance between network status changes and sampling frequency is maintained through an event-driven mechanism. When network topology changes, traffic patterns change abruptly, or equipment failures occur, the event handler immediately adjusts the remaining duration of the current time window and recalculates the theoretical monitoring load. Critical events directly trigger emergency sampling mode, breaking through conventional frequency limits to obtain detailed diagnostic data. Event processing priorities are linked to business criticality, ensuring that the monitoring continuity of important services is not affected by regular scheduling.

[0104] The probe model implements layered monitoring of resource usage. The basic resource layer tracks hardware metrics such as CPU and memory to ensure that frequency adjustment operations do not cause system overload; the service metrics layer monitors the completeness and timeliness of data collection to verify whether the sampling frequency meets monitoring requirements; and the network impact layer assesses the impact of probe behavior on the network under test to prevent monitored traffic from becoming a new source of interference. These three layers of monitoring data jointly guide the fine-tuning of the sampling frequency.

[0105] Metadata management for time windows supports monitoring, backtracking, and analysis. Each window records complete configuration parameters, network status, and execution results, forming a traceable operational log. Historical data supports multi-dimensional queries by time range, node grouping, or business type, helping to analyze long-term network performance evolution trends. Log storage adopts a circular buffer structure to balance storage overhead and historical depth requirements. Key window data is permanently archived for use in-depth root cause analysis.

[0106] Versioned management of the sampling frequency control strategy enables smooth upgrades. When a new frequency algorithm or optimized parameters are introduced, the old and new versions of the strategy are run in parallel to compare their effects. Once the new strategy is confirmed to be consistently superior to the old version, its application scope is gradually expanded. A version rollback mechanism ensures rapid recovery to a known good working state in case of anomalies. Strategy version information is embedded in control instructions to ensure that the probe model always uses matching processing logic.

[0107] Spatiotemporal correlation analysis of node traffic data enhances the accuracy of load forecasting. It considers not only the absolute traffic value within a single window but also analyzes correlation characteristics such as traffic transfer patterns between adjacent nodes and historical traffic patterns over the same time period. These analytical results correct the calculated baseline load values, more accurately reflecting actual monitoring needs. The spatiotemporal analysis algorithm employs a sliding window technique to balance computational complexity and forecast timeliness.

[0108] Distributed collaboration in the probe model improves the efficiency of large-scale network monitoring. In multi-probe deployment scenarios, load balancing among probes is considered when generating control commands to avoid overloading some nodes while others are idle. Probes share network status awareness data and collaboratively adjust sampling focus areas. The collaboration mechanism adopts a hierarchical control architecture, combining local autonomous decision-making with global coordination and optimization.

[0109] Example 5: The data throughput of the network monitoring probe model at a unit sampling frequency was obtained through a standardized testing procedure. The test environment was configured as a dedicated isolated network to avoid external traffic interference. The test used a standardized set of data packet samples, including typical data packets of different sizes and protocol types. During the test, the sampling frequency was fixed as the baseline value, and the number of data packets successfully processed by the probe model per unit time was recorded, while the computational resource consumption during processing was monitored. The throughput test was executed continuously for multiple rounds, with the data packet distribution ratio varying in each round, and the average throughput and resource consumption range were output. The test results formed a baseline parameter table, linking the sampling frequency to the processing capacity.

[0110] The frequency switching schedule for the data acquisition module is generated based on the optimal sampling frequency and throughput test data. The schedule includes a list of frequency sequence change points within the complete time window, with each change point specifying a target frequency value and a switching timestamp. The schedule generation algorithm first calculates the difference between the optimal sampling frequency and the current frequency, and then, combined with the time cost required for a unit frequency change in the throughput test, derives a safe switching time range. Following the principle of uniform switching point distribution, the entire preset time window is divided into discrete time periods of equal length, with switching operations scheduled at the boundary points between adjacent time periods. The number of time periods is set considering the stability requirements of frequency conversion; too many time periods increase system overhead, while too few decrease adjustment accuracy.

[0111] The process of dividing preset time windows into discrete time segments is strictly aligned with the system clock cycle. The start time of each time segment is calibrated according to the network device's time synchronization protocol to ensure that all nodes use the same time slice counting benchmark. The length of a single time segment is determined by dividing the total window duration by the planned number of switches. The lower limit of the length is limited by the duration for which the probe hardware status is saved, while the upper limit is constrained by the rate of change of network status. Each time segment is assigned a unique number, which is associated with a corresponding set of frequency control parameters. The time window segmentation results are broadcast to relevant network components to maintain global consistency in the monitoring timing.

[0112] After a single discrete time period, the actual data throughput of the data acquisition module is verified. The verification operation reads the total number of processed data packets recorded by the probe's hardware counter and compares it to the theoretically expected throughput for that time period. The expected value is calculated based on the sampling frequency and duration effective for that time period. The deviation statistics between the actual throughput and the expected value include both absolute value differences and relative difference ratios. The verification process simultaneously monitors the probe's internal resource status, including cache queue depth, peak memory usage, and processor interrupt frequency. A verification report is automatically generated, indicating the severity level of any deviation exceeding limits.

[0113] Based on the verification results, compensation adjustments are made to the frequency switching time plan. If the deviation is below the tolerance threshold, the original plan for subsequent time periods is maintained; if the deviation exceeds the threshold, the compensation algorithm is activated. The compensation algorithm calculates the impact of the current deviation on the overall monitoring target and derives the required frequency compensation coefficient. This coefficient is applied to the planned frequency value for subsequent time periods; positive deviations correct the frequency downwards, and negative deviations correct the frequency upwards. The correction range is limited by the maximum and minimum frequency boundaries supported by the probe; for continuous deviations across multiple time periods, a cumulative compensation strategy is used. The correction operation updates the list of boundary points for subsequent time periods, generating a new switching time series.

[0114] The frequency switching schedule dynamically evolves within the entire preset time window. At the end of each time slice, the schedule parameters for the remaining time slots are iteratively adjusted based on the latest validation results. The adjustment process maintains the total planned load constant, only redistributing the frequency values ​​for each time slot. The revised schedule version number is incremented, and change records are stored in a bound manner with the time window number. Historical data on schedule evolution is used to analyze the accuracy of the network state prediction model and support subsequent algorithm optimization.

[0115] The data acquisition module's operating frequency updates strictly adhere to the switching schedule. Upon reaching the planned switching point, the probe executes a pre-defined state transition process: pausing the current sampling thread → saving the sampling context → loading the new frequency parameters → initializing hardware registers → restarting the sampling thread. The switching action is completed within milliseconds, and any data packets lost during the process are recorded in quality control metrics. In high-frequency switching scenarios, an incremental configuration update strategy is employed, changing only the differing parameters to reduce switching time.

[0116] Verification of actual data throughput incorporates multi-dimensional auxiliary metrics. In addition to core throughput figures, analysis includes packet processing latency distribution, the proportion of abnormal packets, and protocol parsing error rates. These metrics help differentiate the sources of throughput deviations: insufficient system processing capacity leading to overall performance degradation, or specific protocol processing defects causing localized performance deterioration. Multi-dimensional metrics assist in pinpointing the root cause of problems and guide the development of targeted compensation strategies. Throughput verification data is integrated with the probe log system, supporting backtracking analysis for any time period.

[0117] The time slice end trigger mechanism employs a dual-condition system. The primary condition is the expiration of the preset duration, implemented through a high-precision timer; the secondary condition is a sudden network event signal, such as a traffic surge alarm or equipment failure notification. The secondary condition takes priority, immediately interrupting the current time slice and entering the verification process. This design ensures rapid response during significant network changes, avoiding the mechanical waiting for the time slice to end.

[0118] The compensation and correction strategy is differentiated based on the type of deviation. Systematic deviations are compensated using linear compensation coefficients, random deviations are handled with a smoothing filter algorithm, and persistent deviations trigger a planned refactoring process. Correction parameters are normalized to eliminate the impact of network size differences. The final determination of the compensation value references historical correction effect data to avoid over-correction that could cause oscillations. The decision-making logic of the correction strategy incorporates a built-in foolproof mechanism to prohibit adjustments to frequency values ​​outside the safe range.

[0119] Discrete time-slice management is implemented using a state machine. The time-slice state comprises four phases: preparation, execution, verification, and transition. The preparation phase loads configuration parameters and initializes the hardware; the execution phase performs routine data acquisition; the verification phase compares throughput; and the transition phase completes compensation calculations and plan updates. State transitions are coordinated by a central scheduler, and the time consumed in each phase is recorded for performance optimization. The state machine implementation ensures that even during complex compensation and correction operations, the system maintains a deterministic behavior pattern.

[0120] The closed-loop control of the frequency switching time schedule relies on the adjustment of the feedback loop strength. The feedback coefficient is dynamically adjusted based on the stability of the network environment: in a stable environment, the feedback strength is reduced to prevent sensitive oscillations; in a turbulent environment, the feedback strength is increased to accelerate the response. The feedback strength parameter is correlated with the standard deviation of the verification results of the time slices, and the coefficient value is automatically reduced when the deviation is low for multiple consecutive time slices. The feedback adjustment range has upper and lower limits to prevent the system from entering a positive feedback runaway state.

[0121] The entire dynamic adjustment process generates operational audit logs. These logs include the switching time points for each time slice, set frequency values, actual throughput, deviation values, and details of compensation and corrections. Audit data is stored in a tamper-proof format and supports third-party verification. Long-term audit data forms a frequency optimization knowledge base, extracting the best control models for typical network scenarios. After the audit logs are generated, an integrity check is triggered; any data anomalies trigger an alarm and freeze the adjustment operation.

[0122] After all discrete time intervals are adjusted within the time window, an overall execution report is output. The report summarizes key indicators such as average frequency, total throughput achievement rate, and number of compensation actions within the window. The report also includes a deviation analysis from theoretical expectations, marking the time intervals of major deviations and their root causes. The final report is uploaded to the network management center as the core input for evaluating the system's operational status. When a new time window starts, all parameters are initialized, inheriting the optimized configuration from the previous window as the default baseline.

[0123] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0124] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring and optimizing the performance of a communication network, characterized in that, include: The process involves collecting data on performance nodes affecting transmission quality in a communication network, and constructing a network transmission quality model based on these nodes. Specifically, this includes: extracting the topology information of the communication network and dividing it into multiple logical subnets; identifying key performance nodes in each logical subnet to generate a set of performance nodes; measuring the bandwidth capacity of each performance node and analyzing external interference factors affecting it; and establishing a data transmission balance equation for the communication network based on the bandwidth capacity and external interference factors. Specifically, this involves: analyzing the data transmission coupling effect between performance nodes and quantifying the impact intensity of external interference factors on each performance node; and constructing the data transmission balance equation by combining the bandwidth capacity of the performance nodes, the data transmission coupling effect, and the impact intensity of external interference factors. Configure the network monitoring probe model, perform performance testing on the network monitoring probe model, and obtain the latency detection capability and packet loss rate detection accuracy of the network monitoring probe model. The network monitoring probe model is integrated into the network transmission quality model, specifically by analyzing the influence weight of each performance node in the network transmission quality model and locating the target performance node with the largest influence weight. The network monitoring probe model is deployed in the logical subnet corresponding to the target performance node; Calculate the total data traffic load of the communication network within a preset time window, and combine the latency detection capability of the network monitoring probe model to calculate the theoretical monitoring load of the network monitoring probe model within the preset time window; Based on the theoretical monitoring load and packet loss rate detection accuracy, a data sampling frequency control command for the network monitoring probe model is generated. According to the data sampling frequency control command, the operating frequency of the data acquisition module of the network monitoring probe model is dynamically adjusted; The effectiveness of the network monitoring probe model in optimizing the transmission quality of the communication network was evaluated.

2. The communication network performance monitoring and optimization method as described in claim 1, characterized in that, The configuration of the network monitoring probe model and the execution of performance testing operations for the network monitoring probe model include: The latency response characteristics and packet loss detection accuracy of the network monitoring probe model were tested under high load, low load and fluctuating load conditions to obtain the latency detection capability and packet loss detection accuracy.

3. The communication network performance monitoring and optimization method as described in claim 1, characterized in that, The total data traffic load of the computing communication network within a preset time window includes: The total data traffic load is obtained by aggregating the data traffic generation of each performance node within a preset time window. The calculation of the theoretical monitoring load of the network monitoring probe model includes: Obtain the basic monitoring overhead of the network monitoring probe model within a preset time window, and calculate the theoretical monitoring load by combining it with the total data traffic load; The data sampling frequency control instructions include: Based on the theoretical monitoring load and packet loss rate detection accuracy, the optimal sampling frequency of the network monitoring probe model within a preset time window is determined.

4. The communication network performance monitoring and optimization method as described in claim 3, characterized in that, The generated data sampling frequency control command includes: The data throughput of the network monitoring probe model at a unit sampling frequency is tested, and a frequency switching time plan for the data acquisition module is generated based on the optimal sampling frequency.

5. The communication network performance monitoring and optimization method as described in claim 4, characterized in that, The dynamic adjustment of the operating frequency of the data acquisition module of the network monitoring probe model includes: The preset time window is divided into discrete time periods, and the frequency switching time plan is evenly distributed to each discrete time period. After a single discrete time period ends, verify whether the actual data throughput of the data acquisition module meets expectations; Based on the verification results, the frequency switching time plan for subsequent discrete time periods is compensated and corrected until the adjustment operation of the entire preset time window is completed.

6. The communication network performance monitoring and optimization method as described in claim 1, characterized in that, The evaluation of the network monitoring probe model's effect on optimizing the transmission quality of the communication network includes: Obtain the changes in latency and packet loss rate of the communication network before and after optimization; Based on the changes in delay and packet loss rate, the actual monitoring load of the communication network is calculated. By comparing the deviation between the actual monitoring load and the theoretical monitoring load, a frequency control authority index for the data acquisition module is generated. The network transmission quality optimization effect is evaluated based on the frequency control authority index.

7. A communication network performance monitoring and optimization system, used to implement the communication network performance monitoring and optimization method according to any one of claims 1-6, characterized in that, The system includes: The transmission quality modeling module is used to collect performance nodes affecting transmission quality in the communication network and construct a network transmission quality model based on these performance nodes. Specifically, it includes: extracting the topology information of the communication network and dividing it into multiple logical subnets; identifying key performance nodes in each logical subnet and generating a set of performance nodes for the communication network; measuring the bandwidth carrying capacity of each performance node and analyzing external interference factors affecting it; and establishing a data transmission balance equation for the communication network based on the bandwidth carrying capacity and external interference factors. Specifically, it involves: analyzing the data transmission coupling effect between performance nodes and quantifying the impact intensity of external interference factors on each performance node; and constructing the data transmission balance equation by combining the bandwidth carrying capacity of the performance nodes, the data transmission coupling effect, and the impact intensity of external interference factors. The probe performance testing module is used to configure the network monitoring probe model, execute the performance testing operation of the network monitoring probe model, and obtain the latency detection capability and packet loss rate detection accuracy. The model fusion module is used to integrate the network monitoring probe model into the network transmission quality model, specifically by analyzing the influence weight of each performance node in the network transmission quality model and locating the target performance node with the largest influence weight. The monitoring load calculation module is used to calculate the total data traffic load of the communication network within a preset time window, and to calculate the theoretical monitoring load in combination with the delay detection capability. The sampling control module is used to generate data sampling frequency control instructions for the network monitoring probe model based on the theoretical monitoring load and packet loss rate detection accuracy. A dynamic adjustment module is used to dynamically adjust the operating frequency of the data acquisition module according to the data sampling frequency control command; An optimization evaluation module is used to evaluate the effect of the network monitoring probe model on the transmission quality optimization of the communication network.

Images