WAPI Network Optimization and Simulation System Based on Digital Twin

Abstract

The present invention discloses a WAPI network optimization and simulation system based on digital twin, belonging to the technical field of wireless network optimization. Specifically, it includes: collecting physical entity data of the WAPI network by using sensors and monitoring devices and transmitting it to the data processing center; the data processing center constructs a digital twin model accordingly and compares it with the actual data; based on the digital twin model, an optimization algorithm is used to evaluate and analyze the network, find problems and formulate network optimization strategies; simulating the network optimization strategies on the digital twin model and adjusting the network optimization strategies according to the simulation results; applying the adjusted network optimization strategies to the actual network, continuously monitoring through the model, comparing the data before and after optimization, and further optimizing the model and network optimization strategies based on the feedback information; this application solves the problems of low efficiency and poor accuracy in traditional WAPI network optimization, realizes efficient and accurate optimization through digital twin, and improves network performance and reliability.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless network optimization, specifically a WAPI network optimization and simulation system based on digital twin. Background Art

[0002] The Wireless LAN Authentication and Privacy Infrastructure (WAPI) network has unique advantages in terms of security and stability. However, traditional network optimization methods often rely on on-site testing and empirical judgment, suffering from problems such as low efficiency, high cost, and difficulty in comprehensively and accurately grasping the network operation status. Digital twin technology can achieve real-time monitoring, analysis, and optimization of the network by constructing a virtual digital model corresponding to the physical network. However, the current digital twin optimization and simulation system for WAPI networks is not yet perfect.

[0003] For example, the Chinese patent application with the publication number CN116801429A discloses an optimal networking method for a photovoltaic power generation field supporting WAPI, including: dividing the photovoltaic power generation field into N groups, where N>1, and each group of photovoltaic power generation fields contains n photovoltaic panels, where n>1; configuring a fiber optic network in each group of photovoltaic power generation fields, and installing an Internet of Things sub-station device on each photovoltaic panel; establishing wireless network communication using the self-organizing node adjacent search algorithm, and forming an automatic networking link for photovoltaic panels through the connection of antennae between Internet of Things sub-station devices; using the Dijkstra algorithm to optimize the optimal path of the network transmission distance between photovoltaic panels, and then using the WAPI encryption technology and protocol to perform security authentication on this technical solution through the devices of the Internet of Things sub-stations installed on the photovoltaic panels of the photovoltaic power generation field, and realizing the optimal networking method for a photovoltaic power generation field supporting WAPI through the mutual connection between Internet of Things sub-station devices, forming an automatic networking.

[0004] The above existing technologies all have the following problems: lack of accurate network model construction and calibration; lack of a systematic network performance optimization and evaluation mechanism; only disclose the establishment of wireless network communication using the self-organizing node adjacent search algorithm and the optimization of the transmission distance using the Dijkstra algorithm, but do not elaborate on error handling, retransmission mechanisms, etc. during the data transmission process. Summary of the Invention

[0005] In view of the deficiencies of the prior art, the present invention proposes a WAPI network optimization and simulation system based on digital twin, which uses sensors and monitoring devices to collect physical entity data of the WAPI network and transmits it to the data processing center; the data processing center constructs a digital twin model accordingly and compares it with the actual data; based on the digital twin model, an optimization algorithm is used to evaluate and analyze the network, find problems and formulate network optimization strategies; the network optimization strategies are simulated on the digital twin model, and the network optimization strategies are adjusted according to the simulation results; the adjusted network optimization strategies are applied to the actual network, and through continuous monitoring of the model, the data before and after optimization are compared, and the model and network optimization strategies are further optimized based on the feedback information; this application solves the problems of low efficiency and poor accuracy in traditional WAPI network optimization, and realizes efficient and accurate optimization through digital twin, improving network performance and reliability.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A WAPI network optimization and simulation system based on digital twin, comprising: a data acquisition module, a digital twin model construction module, a model calibration and verification module, a network optimization strategy formulation module, a simulation and effect evaluation module, an implementation and feedback module;

[0008] The data acquisition module includes a sensor unit and a data transmission unit. The sensor unit is configured with an intelligent sensing strategy for sensing physical entity data of the network; the data transmission unit is configured with an automatic repeat request hybrid dynamic adjustment of retransmission times strategy for transmitting the physical entity data collected by the sensor unit to the data processing center;

[0009] The digital twin model construction module includes a topology construction unit. The topology construction unit is configured with an improved depth-first search algorithm, which combines the network layering idea to identify the connection relationships in the WAPI network and construct a network topology structure;

[0010] The model calibration and verification module includes a data comparison unit. The data comparison unit is configured with a difference detection algorithm for comparing the network performance indicators output by the digital twin model with the actual network operation data at each time point, calculating the difference value and trend change between the two;

[0011] The network optimization strategy formulation module includes a problem diagnosis unit and a strategy generation unit. The strategy generation unit is used to generate network optimization strategies by synthesizing multiple performance indicators.

[0012] Specifically, the sensor unit adopts an intelligent sensing strategy. The specific steps of the intelligent sensing strategy include:

[0013] A1: Start the sensor and record the network traffic value at the current moment as the initial traffic value ;

[0014] A2: Set the initial sampling interval , and perform data collection according to the initial sampling interval . After each data collection is completed, obtain the network traffic value of the current network , and calculate the network traffic change rate based on the network traffic value of the current network , where i represents the collection times index, represents the i -th network traffic change rate when collecting data for the -th time, i represents the network traffic value recorded during the

[0015] -1-th collection.

[0016] Specifically, the specific steps of the intelligent perception strategy further include: A3: According to the calculated network traffic change rate , calculate the sampling interval for the next data collection , where i represents the sampling interval when collecting data for the k -th time,

[0017] represents a constant; ;

[0018] If , then set the sampling interval to ;

[0019] If , then set the sampling interval to ;

[0020] If , then set the sampling interval to ;

[0021] A5: The sensor waits for the next data collection moment to arrive according to the updated sampling interval, and repeats A2 to A4 after the moment arrives to complete data collection.

[0022] Specifically, the data transmission unit adopts an automatic repeat request hybrid dynamic adjustment retransmission times strategy, and the specific steps include:

[0023] B1: When the data transmission unit starts, turn on the channel quality monitoring function, and obtain the initial channel error rate through communication protocol interaction with the data processing center , and at the same time, set a timer , for periodically updating the measured value of the channel bit error rate;

[0024] B2: Obtain the physical entity data collected by the sensor unit, classify the physical entity data using a predefined method, and assign an initial data importance coefficient to each type of data ;

[0025] B3: After the sensor unit collects the physical entity data, the data transmission unit packs the collected physical entity data into a transmission data packet format and adds a data type identifier as the header information.

[0026] Specifically, the specific steps of the automatic repeat request hybrid dynamic adjustment of the retransmission times strategy further include:

[0027] B4: According to the current channel bit error rate and the corresponding data importance coefficient , calculate the retransmission times for each transmission , and send the data packet to the data processing center. At the same time, start a timer and wait for the confirmation feedback signal from the data processing center; where, represents the retransmission times of the physical entity data transmission for the i th collection, represents rounding up for represents the importance coefficient of the physical entity data transmission for the i th collection, represents the channel bit error rate during the physical entity data transmission for the i th collection, represents the logarithmic function;

[0028] B5: Obtain the feedback signal confirmation time of the data processing center ;

[0029] If , it means that the data transmission is successful, stop the timer, and record the transmission time and retransmission times of this transmission;

[0030] If , enter the retransmission operation.

[0031] Specifically, the specific steps of the automatic repeat request hybrid dynamic adjustment of the retransmission times strategy further include:

[0032] B6: According to the calculated retransmission times , perform the retransmission operation of the data packet. Each time a retransmission is performed, restart the timer and wait for the confirmation feedback signal;

[0033] If the confirmation feedback signal is received, stop the retransmission;

[0034] If the acknowledgement feedback signal is not received after reaching the upper limit of the retransmission count, record the failure of this transmission;

[0035] B7: According to the set timer , re-measure the channel bit error rate, update it to , and dynamically adjust the data importance coefficient.

[0036] Specifically, the topology construction unit adopts an improved depth-first search algorithm. The specific steps of the improved depth-first search algorithm include:

[0037] C1: Mark all access points and terminal device network nodes as unvisited, and create an empty data structure to store network topology structure information. The created empty data structure is an adjacency list;

[0038] C2: Consider the access point as the first layer, and the terminal devices directly connected to the access point as the second layer. Set an initial layer label for each node;

[0039] C3: Select an access point as the starting node for depth-first search, visit the starting access point, mark it as visited, and record the starting access point layer information;

[0040] C4: Find the neighbor nodes connected to the starting access point, and determine the connection weights of the neighbor nodes according to where b and c represent weight coefficients, s represents signal strength, w , and represents the bandwidth.

[0041] Specifically, the specific steps of the improved depth-first search algorithm further include:

[0042] C5: Determine the layer of each neighbor node according to its type;

[0043] If the neighbor node is a terminal device, the layer of the neighbor node is the current access point layer plus 1;

[0044] If the neighbor node is an access point other than the currently visited access point, the layer of the neighbor node is the current access point layer;

[0045] C6: For each unvisited neighbor node, during the recursive process, continuously update the node's access status, layer information, and connection relationship, and record them into the topology structure data structure;

[0046] C7: After completing the depth-first search of all nodes, convert the adjacency list data structure into a topology graph form, where nodes represent access points and terminal devices, edges represent connection relationships, and the weights of the edges represent connection weights, to obtain the complete network topology structure. ​

[0047] Specifically, the data comparison unit adopts a difference detection algorithm. The specific steps of the difference detection algorithm are as follows:

[0048] D1: Collect the network performance index data in the digital twin model and the actual network operation data in the same time interval, and assign a unique performance index identifier to each index;

[0049] D2: Starting from the first time point, compare each performance index in the network performance index data output by the digital twin model and the actual network operation data in turn;

[0050] D3: For each performance index, use the formula to calculate the difference value, where represents the difference value of the performance index t at the time point j , represents the value of the performance index t output by the digital twin model at the time point j , represents the value of the performance index t in the operation data of the actual network at the time point j ;

[0051] D4: Store the difference value, the corresponding timestamp and the performance index identifier at each time point into a pre-constructed difference data list;

[0052] D5: Use the difference method to calculate the trend change, and determine whether the trend is rising, falling or stable according to the trend calculation result;

[0053] D6: Organize the analysis results of the difference value and the trend change into a report form.

[0054] Specifically, the network optimization strategy includes: adjusting the access point position, optimizing the channel allocation, enhancing network security, performing load balancing, and adjusting the transmission power.

[0055] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0056] 1. The present invention proposes a WAPI network optimization and simulation system based on digital twin, and has optimized improvements in the architecture, operation steps and processes. The system has the advantages of simple process, low investment and operation costs, and low production work costs.

[0057] 2. The present invention proposes a WAPI network optimization and simulation system based on digital twin. In the data acquisition link, the intelligent perception strategy of the sensor unit can flexibly perceive the data of network physical entities, and can automatically adjust the perception frequency and accuracy according to the network environment, avoiding energy waste and data redundancy while ensuring the acquisition of key and accurate data; while the automatic repeat request hybrid dynamic adjustment of the retransmission times strategy configured in the data transmission unit can reasonably determine the retransmission times with reference to the channel quality and data importance, improving the reliability and efficiency of data transmission.

[0058] 3. The present invention proposes a WAPI network optimization and simulation system based on digital twin. The digital twin model construction module in the system uses an improved depth-first search algorithm combined with the network layering idea to accurately identify the network connection relationship and construct the topological structure, providing a practical network model for subsequent analysis; the model calibration and verification module uses the difference detection algorithm to realize the comparison of the network performance index data output by the digital twin model and the actual network operation data at each time point, accurately grasping the difference value and trend change between the two, and ensuring the accuracy of the model; the network optimization strategy formulation module generates optimization strategies by comprehensively considering multiple performance indicators, can effectively diagnose network problems and give targeted solutions, realize the all-round optimization of the WAPI network, and overall improve the performance and operation quality of the network. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is the architecture diagram of the WAPI network optimization and simulation system based on digital twin of the present invention;

[0060] Figure 2 is the implementation flowchart of the sensor unit of the WAPI network optimization and simulation system based on digital twin of the present invention;

[0061] Figure 3 is the implementation flowchart of the data transmission unit of the WAPI network optimization and simulation system based on digital twin of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0062] Example 1

[0063] Please refer to Figure 1 , an example provided by the present invention: a WAPI network optimization and simulation system based on digital twin, including:

[0064] a data acquisition module, a digital twin model construction module, a model calibration and verification module, a network optimization strategy formulation module, a simulation and effect evaluation module, and an implementation and feedback module;

[0065] The data acquisition module, multiple sensors and network monitoring devices distributed in the WAPI network, are used to collect the physical entity data of the network in real time and transmit the data to the data processing center through wired or wireless communication methods;

[0066] The digital twin model construction module, located in the data processing center, is responsible for constructing the digital twin model of the WAPI network based on the collected data, and using advanced modeling software and algorithms to achieve the precise construction and real-time update of the model;

[0067] The model calibration and verification module is used to calibrate and verify the digital twin model by comparing it with the actual network operation data, ensuring the accuracy and reliability of the model, and adopting automated verification tools and algorithms to improve the verification efficiency;

[0068] The network optimization strategy formulation module is used to generate network optimization strategies according to the analysis results of the digital twin model, using optimization algorithms and considering various factors and constraints to ensure the feasibility and effectiveness of the network optimization strategies;

[0069] The simulation and effect evaluation module is used to conduct simulation experiments on the network optimization strategies on the digital twin model, evaluate the optimized network performance indicators, and display the simulation results through a visual interface to facilitate user analysis and decision-making;

[0070] The implementation and feedback module is used to apply the network optimization strategies to the actual network and feedback the actual network operation data to the digital twin model to achieve continuous monitoring and improvement of the network optimization effect, forming a closed-loop optimization process.

[0071] The data acquisition module includes: a sensor unit and a data transmission unit;

[0072] The sensor unit is used to sense the physical entity data of the network using intelligent sensing strategies;

[0073] The data transmission unit is used to accurately and timely transmit various types of network physical entity data collected by the sensor unit to the data processing center through a wireless communication protocol, using an automatic repeat request hybrid dynamic adjustment of the retransmission times strategy, ensuring that the data can smoothly enter the subsequent processing process and avoiding the impact of data loss or transmission delay on the system.

[0074] The digital twin model construction module includes: a topology construction unit, a parameter configuration unit, and a model update unit;

[0075] The topology construction unit is used to identify the access points, terminal devices, and their connection relationships in the WAPI network using an improved depth-first search algorithm combined with the network layering idea, and construct the network topology structure;

[0076] A parameter configuration unit, which is used to determine corresponding parameters according to the collected physical entity data, such as the transmission power parameter of an access point, the frequency band parameter of a channel, the communication protocol related parameters of a device, etc., and assign these parameters to the corresponding elements in the digital twin model;

[0077] A model update unit, which is used to monitor the changes of newly incoming data in real time. When the collected data reflects changes in the network device status, network topology changes, or other factors affecting the network, it timely updates the constructed digital twin model accordingly to ensure that the model can always accurately reflect the real-time state of the physical network and ensure the synchronization between the model and the actual network.

[0078] The model calibration and verification module includes: a data comparison unit and a parameter adjustment unit;

[0079] The data comparison unit is used to use a difference detection algorithm based on time series analysis to compare and analyze the network performance metrics output by the digital twin model, such as throughput and latency, with the actual network operation data at each time point, and calculate the difference value and trend change between the two;

[0080] The parameter adjustment unit is used to optimize and adjust the model parameters using a genetic algorithm based on the above difference detection results. The key parameters in the model, such as the attenuation coefficient in the signal propagation model and the device transmission power, are encoded as chromosomes for the adjustment step. Through genetic operations such as selection, crossover, and mutation, it searches for the parameter combination that minimizes the difference between the model and the actual data. Among them, the genetic algorithm is the prior art content in this field and is not the creative solution of this application, so it will not be elaborated here.

[0081] The network optimization strategy formulation module includes: a problem diagnosis unit and a strategy generation unit;

[0082] The problem diagnosis unit is used to deeply analyze the network performance data in the digital twin model using a method combining an expert system and data analysis, and quickly and accurately diagnose problems existing in the network, such as signal coverage blind spots and severely interfered channel areas;

[0083] The strategy generation unit is used to generate network optimization strategies by using a non-dominated sorting genetic algorithm and integrating multiple performance metrics. Among them, the non-dominated sorting genetic algorithm is the prior art content in this field and is not the creative solution of this application, so it will not be elaborated here.

[0084] In summary, the overall implementation process of the WAPI network optimization and simulation system based on digital twins includes:

[0085] S1: Collect the physical entity data of the WAPI network using sensors and network monitoring devices, including information such as the signal strength of access points, channel utilization rate, number of user connections, device operating status, etc., and transmit this data to the data processing center;

[0086] S2: The data processing center constructs a digital twin model of the WAPI network based on the collected physical entity data, compares the output results of the digital twin model with the actual network operation data, and at the same time, uses error analysis to verify the comparison results;

[0087] S3: Based on the digital twin model, use optimization algorithms to evaluate and analyze the performance of the WAPI network, identify bottlenecks and potential problems in the network, such as insufficient signal coverage and severe channel interference, and according to the analysis results, formulate corresponding network optimization strategies, including access point location adjustment, channel allocation optimization, transmit power adjustment, etc.;

[0088] S4: Conduct simulation experiments on the network optimization strategy on the digital twin model, simulate the operating state of the optimized network, evaluate the changes in various performance indicators, such as throughput, latency, packet loss rate, etc., and according to the simulation results, adjust and improve the network optimization strategy to ensure that it can achieve the expected optimization effect in the actual network;

[0089] S5: Apply the verified network optimization strategy to the actual WAPI network, continuously monitor the operating state of the network through the digital twin model, and at the same time, collect the actual data after network optimization, compare it with the data before optimization, analyze the optimization effect, and use the feedback information to further optimize the digital twin model and network optimization strategy.

[0090] Exemplarily, in a WAPI network of an office area, first, through the data acquisition module, sensors are deployed at each access point and key locations to collect data such as signal strength, channel quality, and user connection status. After these data are transmitted to the data processing center, the digital twin model construction module uses these data to construct a digital twin model of the office area WAPI network, including the location distribution of access points and the signal propagation range. Then, the model calibration and verification module compares the network performance indicators output by the digital twin model with the actual network data, and by adjusting parameters in the signal propagation model and other means, makes the model accurately reflect the actual network situation. For example, through multiple comparisons, it is found that there is a deviation between the signal strength predicted by the model in a certain area and the actual measured value. After analyzing factors such as the obstacle characteristics in this area, the model parameters are adjusted to make the deviation within an acceptable range. Secondly, based on the calibrated model, the network optimization strategy formulation module finds that the channel utilization rate is too high and network congestion occurs due to a large number of users near several access points. So, network optimization strategies are formulated, including adjusting the channels of some access points, appropriately increasing the transmission power of some access points to expand the coverage range, and reducing the load of access points in user-dense areas. At the same time, the simulation and effect evaluation module simulates these network optimization strategies on the digital twin model, simulates the operating state of the optimized network, and finds that the throughput is significantly improved, the latency is reduced, and the packet loss rate is decreased, indicating that the network optimization strategy is effective. Finally, the network optimization strategy implementation and feedback module applies these network optimization strategies to the actual network, and continuously monitors the network operating state through the digital twin model, collects actual data and compares it with that before optimization to further optimize the model and strategy. For example, it is found that there is still a problem of weak signal in a certain area after optimization, and the location and parameters of the access points in this area are further adjusted, thereby continuously improving the performance of the WAPI network.

[0091] Embodiment 2

[0092] Please refer to Figure 2 , in this embodiment, the sensor unit adopts an intelligent sensing strategy, and the specific steps of the intelligent sensing strategy include:

[0093] A1: Start the sensor and record the current network traffic value as the initial traffic value ;

[0094] A2: Set the initial sampling interval , and perform data collection according to the initial sampling interval . After each data collection is completed, obtain the current network traffic value , and calculate the network traffic change rate based on the current network traffic value , where i represents the collection times index, represents the iThe network traffic change rate during the first data collection represents the network traffic value recorded during the i -1st data collection;

[0095] A3: Based on the calculated network traffic change rate , calculate the sampling interval for the next data collection , where represents the sampling interval during the i th data collection, k represents a constant;

[0096] A4: Set the threshold fluctuation range of the sampling interval to ;

[0097] If , then set the sampling interval to ;

[0098] If , then set the sampling interval to ;

[0099] If , then set the sampling interval to ;

[0100] A5: The sensor waits for the next data collection moment according to the updated sampling interval, and repeats A2 to A4 after the arrival moment to complete the data collection.

[0101] Embodiment 3

[0102] Please refer to Figure 3 , in this embodiment, the data transmission unit adopts an automatic repeat request hybrid dynamic adjustment retransmission times strategy, and the specific steps include:

[0103] B1: When the data transmission unit is started, turn on the channel quality monitoring function, and obtain the initial channel error rate through the communication protocol interaction with the data processing center. At the same time, set a timer for regularly updating the measured value of the channel error rate;

[0104] B2: Obtain the physical entity data collected by the sensor unit, classify the physical entity data using a predefined method, and assign an initial data importance coefficient to each type of data;

[0105] In the present invention, the k-means clustering algorithm is adopted during classification, and the k-means clustering algorithm is the prior art content in this field and is not the creative solution of this application, so it will not be elaborated here.

[0106] Exemplarily, data such as key performance indicators of the access point, such as signal strength and channel utilization, may be assigned a higher coefficient, such as 5; while some auxiliary device status information, such as device temperature, may be assigned a lower coefficient, such as 1; these coefficients can be stored in a lookup table for quick query during data transmission.

[0107] B3: After the sensor unit collects physical entity data, the data transmission unit packs the collected physical entity data into a packet format suitable for transmission and adds a data type identifier as the header information;

[0108] B4: According to the current channel bit error rate and the corresponding data importance coefficient , calculate the number of retransmissions for each transmission , and send the data packet to the data processing center. At the same time, start a timer and wait for the confirmation feedback signal from the data processing center, where represents the number of retransmissions for the transmission of the physical entity data collected for the i th time, represents rounding up , represents the importance coefficient for the transmission of the physical entity data collected for the i th time, represents the channel bit error rate during the transmission of the physical entity data collected for the i th time, represents the logarithmic function;

[0109] B5: Obtain the feedback signal confirmation time of the data processing center ;

[0110] If , it means the data transmission is successful, stop the timer, and record the transmission time and the number of retransmissions for this transmission;

[0111] If , enter the retransmission operation;

[0112] B6: According to the calculated number of retransmissions , perform the retransmission operation of the data packet. Each time a retransmission occurs, restart the timer and wait for the confirmation feedback signal;

[0113] If the confirmation feedback signal is received, stop the retransmission;

[0114] If the confirmation feedback signal is not received after reaching the retransmission count limit, record that this transmission fails;

[0115] B7: According to the set timer , re-measure the channel bit error rate and update it to and dynamically adjust the data importance coefficient.

[0116] Exemplarily, if a certain type of data is found to play a key role in network fault diagnosis within a certain time period, then its data importance coefficient is correspondingly increased; then, according to the new and , calculate the retransmission times for the next data transmission , and continue with the data transmission process.

[0117] The topology construction unit adopts an improved depth-first search algorithm. The specific steps of the improved depth-first search algorithm include:

[0118] C1: Mark all access points and terminal device network nodes as unvisited, and create an empty data structure to store network topology structure information. The created empty data structure is an adjacency list;

[0119] C2: Consider the access point as the first layer, and the terminal devices directly connected to the access point as the second layer. Set an initial layer label for each node;

[0120] C3: Select an access point as the starting node for the depth-first search, then start accessing the starting access point, mark it as visited, and record the starting access point layer information;

[0121] C4: Find the neighbor nodes connected to the starting access point, and determine the connection weights of the neighbor nodes according to , where b and c represent weight coefficients, s represents signal strength, w , and represents the bandwidth;

[0122] C5: Determine its layer according to the type of neighbor node;

[0123] If the neighbor node is a terminal device, then the layer of the neighbor node is the current access point layer plus 1;

[0124] If the neighbor node is an access point other than the currently accessed access point, then the layer of the neighbor node is the current access point layer;

[0125] C6: For each unvisited neighbor node, during the recursive process, continuously update the node's access status, layer information, and connection relationship, and record them in the topology structure data structure;

[0126] C7: After completing the depth-first search of all reachable nodes, convert the adjacency list data structure into an intuitive topological graph representation form, where the nodes represent access points and terminal devices, the edges represent the connection relationships between them, and the weights of the edges represent the connection weights, to obtain the complete network topology structure.

[0127] The data comparison unit adopts a difference detection algorithm. The specific steps of the difference detection algorithm include:

[0128] D1: Collect the network performance index data in the digital twin model and the actual network operation data in the same time interval, and assign a unique performance index identifier to each network performance index. Among them, the network performance indexes include throughput, latency, and packet loss rate. At the same time, the addition and assignment of the unique identifier are the prior art content in this field and not the creative solution of this application, so it will not be elaborated here;

[0129] D2: Starting from the first time point, compare each performance index in the network performance index data output by the digital twin model and the actual network operation data in turn;

[0130] D3: For each performance index, use the formula to calculate the difference value, where represents the difference value of the performance index t at the time point j , represents the value of the performance index t output by the digital twin model at the time point j , represents the value of the performance index t in the operation data of the actual network at the time point j ;

[0131] D4: Store the difference value, the corresponding timestamp, and the performance index identifier at each time point into a pre-constructed difference data list;

[0132] D5: Use the difference method to calculate the trend change , and determine whether the trend is rising, falling, or stable according to the trend calculation result. Among them, represents the difference value of the performance index t at the time point j -1;

[0133] Among them, if is greater than 0 for multiple consecutive time points, it is considered that the difference value of this performance index shows an upward trend, indicating that the deviation between the digital twin model and the actual network data is gradually increasing.

[0134] D6: Organize the analysis results of the difference value and the trend change into a report form. Among them, the report includes: the average difference value, the maximum difference value, and the trend change chart of each performance index, such as a line chart showing the change trend of the difference value over time, etc.

[0135] Network optimization strategies include: adjusting access point locations, optimizing channel allocation, enhancing network security, performing load balancing, and adjusting transmission power.

[0136] Furthermore, the access point location adjustment strategy includes:

[0137] Macro layout adjustment: Within the WAPI network coverage area, the location of the access points is replanned based on the coverage blind spots or weak signal areas analyzed by the digital twin model. For example, for a WAPI network in a large shopping mall, if the digital twin model shows that the signal coverage in the corner area of a certain floor is insufficient, the strategy generation unit may generate a strategy to appropriately move nearby access points to that area to improve coverage.

[0138] Micro-position fine-tuning: In addition to macro-position movement, it can also include adjustments to micro-position parameters such as the height and angle of the access point. For example, by adjusting the antenna angle of the access point, its signal can better cover specific user-dense areas, reduce signal waste in open areas, and improve signal utilization.

[0139] Channel allocation optimization strategies include:

[0140] Frequency band selection and allocation: Consider the frequency band resources available in the WAPI network and allocate the optimal frequency band to the access point based on the characteristics of different frequency bands, such as interference conditions and transmission rates. For example, in an environment with many interference sources, frequency bands with strong anti-interference capabilities are allocated to access points that are most affected, avoiding channel interference between adjacent access points and improving channel multiplexing efficiency.

[0141] Dynamic channel adjustment: Based on the dynamic changes in network traffic and the real-time usage of channels, a strategy for dynamically adjusting channels is formulated. For example, when the channel utilization of a certain access point is too high and the channels of other access points are relatively idle, the strategy generation unit generates a strategy to transfer some users or traffic to idle channels to balance the channel load and improve the overall network performance.

[0142] Transmit power adjustment strategies include:

[0143] Power boost strategy: For areas with insufficient signal coverage, the strategy generation unit may generate a strategy to increase the transmit power of the access point. However, it should be noted that the increase in transmit power must be within the prescribed range to avoid interference with other wireless devices. For example, in the edge area of an industrial park, in order to ensure that logistics vehicles outside the factory can also access the network stably, the transmit power of the access point in the area can be appropriately increased.

[0144] Power reduction strategy: In areas with good signal coverage and relatively sparse user distribution, reduce the transmission power of access points to reduce unnecessary energy consumption and potential interference to other frequency bands. For example, during non-working hours at night in an office building, when most users have left and the network load is low, the transmission power of access points can be reduced.

[0145] The load balancing strategy includes:

[0146] User access balance: By adjusting the parameters of access points (such as signal strength threshold, association priority, etc.), guide user devices to access each access point evenly. For example, in a hotel's WAPI network, avoid a large number of users concentrating on accessing an access point near the front desk. Instead, according to the load situation of each access point, guide newly connected users to access points with lighter loads to prevent local network congestion.

[0147] Traffic distribution balance: For the data traffic in the network, formulate strategies to reasonably distribute the traffic to different access points and links. For example, in a multi-access point campus network, according to the server resources and bandwidth conditions connected to each access point, evenly distribute large data traffic services such as video streams and file downloads to different access points to avoid performance degradation of a certain access point due to excessive traffic.

[0148] The network security enhancement strategy includes:

[0149] Optimization of authentication strategy: The WAPI network has a unique authentication mechanism. The policy generation unit can optimize the authentication strategy. For example, adjust the parameters of the authentication server to improve the efficiency and security of authentication, ensuring that only legitimate user devices can access the network. This may include measures such as shortening the authentication response time and strengthening the management of authentication keys;

[0150] Adjustment of encryption strategy: According to the security requirements and performance requirements of the network, adjust the encryption algorithm and encryption strength. On the premise of ensuring data privacy, minimize the impact of encryption on network performance. For example, for the WAPI network access part of the enterprise's finance department with extremely high requirements for data security, adopt a higher-level encryption algorithm and a longer key length; while for the network access in some public areas that are sensitive to performance but have relatively low security requirements, adopt appropriate encryption methods to balance security and performance.

[0151] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make changes, modifications, substitutions, and variations to the above embodiments without departing from the purpose and scope of the present invention. These all fall within the protection scope of the present invention.

Claims

1. A WAPI network optimization and simulation system based on digital twin, characterized in that, Including: A data acquisition module, a digital twin model construction module, a model calibration and verification module, a network optimization strategy formulation module, a simulation and effect evaluation module, and an implementation and feedback module; The data acquisition module includes a sensor unit and a data transmission unit. An intelligent sensing strategy is configured in the sensor unit, and the intelligent sensing strategy is used to sense the physical entity data of the network. An automatic repeat request hybrid dynamic retransmission count adjustment strategy is configured in the data transmission unit, and is used to transmit the physical entity data collected by the sensor unit to the data processing center; The digital twin model construction module includes a topology construction unit. An improved depth-first search algorithm is configured in the topology construction unit, and the improved depth-first search algorithm, in combination with the network layering idea, identifies the connection relationships in the WAPI network and constructs a network topology structure; The model calibration and verification module includes a data comparison unit. A difference detection algorithm is configured in the data comparison unit, and the difference detection algorithm is used to compare the network performance indicators output by the digital twin model with the actual network operation data point by point in time, and calculate the difference value and trend change between the two; The network optimization strategy formulation module includes a problem diagnosis unit and a strategy generation unit. The strategy generation unit is used to generate a network optimization strategy by synthesizing multiple performance indicators; The topology construction unit adopts an improved depth-first search algorithm. The specific steps of the improved depth-first search algorithm include: C1: Mark all access points and terminal device network nodes as unvisited, and create an empty data structure to store network topology structure information. The created empty data structure is an adjacency list; C2: Consider the access points as the first layer, and the terminal devices directly connected to the access points as the second layer, and set an initial layer mark for each node; C3: Select an access point as the starting node for depth-first search, visit the starting access point, mark it as visited, and record the starting access point layer information; C4: Search for neighbor nodes connected to the starting access point and, according to determine the connection weight w of the neighbor nodes, where b and c represent weight coefficients, s represents signal strength, represents the bandwidth; The specific steps of the improved depth-first search algorithm further include: C5: Determine its layer according to the type of neighbor node; If the neighbor node is a terminal device, the layer of the neighbor node is the current access point layer plus 1; If the neighbor node is an access point other than the currently visited access point, the layer of the neighbor node is the current access point layer; C6: For each unvisited neighbor node, during the recursive process, continuously update the node's access status, layer information, and connection relationship, and record them into the topology structure data structure; C7: After completing the depth-first search of all nodes, convert the adjacency list data structure into a topology graph form, where the nodes represent access points and terminal devices, the edges represent connection relationships, and the weights of the edges represent connection weights, to obtain a complete network topology structure.

2. The digital twin-based WAPI network optimization and simulation system according to claim 1, wherein The sensor unit adopts an intelligent sensing strategy. The specific steps of the intelligent sensing strategy include: A1: Activate the sensor and record the network traffic value at the current moment as the initial traffic value ; A2: Set the initial sampling interval , and perform data acquisition according to the initial sampling interval . After each acquisition is completed, obtain the traffic value of the current network , and calculate the network traffic change rate based on the traffic value of the current network , where i represents the acquisition times index represents the network traffic change rate at the i-th data acquisition represents the network traffic value recorded at the (i - 1)-th acquisition 3. The WAPI network optimization and simulation system based on digital twin as described in claim 2, wherein, The specific steps of the intelligent sensing strategy further include: A3: According to the calculated network traffic change rate , calculate the sampling interval for the next data collection , where represents the sampling interval at the i-th data collection, and k represents a constant; A4: Set the threshold fluctuation range of the sampling interval to be ; If , then set the sampling interval to ; If , the sampling interval is set to ; If , then set the sampling interval to ; A5: The sensor waits for the next data acquisition moment according to the updated sampling interval. After the moment arrives, repeat A2 to A4 to complete data acquisition.

4. The digital twin-based WAPI network optimization and simulation system according to claim 3, wherein The data transmission unit adopts a strategy of automatically retransmitting requests and dynamically adjusting the number of retransmissions. The specific steps include: B1: When the data transmission unit starts, enable the channel quality monitoring function, and obtain the initial channel bit error rate through the communication protocol interaction with the data processing center. Meanwhile, set a timer. It is used to update the measured value of the channel bit error rate regularly. B2: Obtain the physical entity data collected by the sensor unit, classify the physical entity data using a predefined method, and assign an initial data importance coefficient to each category of data ; B3: After the sensor unit collects the physical entity data, the data transmission unit packs the collected physical entity data into a transmission data packet format and adds a data type identifier as the header information.

5. The digital twin-based WAPI network optimization and simulation system according to claim 4, characterized in that, The specific steps of the strategy of automatically retransmitting requests and dynamically adjusting the number of retransmissions further include: B4: According to the current channel bit error rate and the corresponding data importance coefficient , calculate the number of retransmissions for each transmission , and send the data packet to the data processing center. Meanwhile, start a timer and wait for the confirmation feedback signal from the data processing center; where represents the number of retransmissions for the physical entity data transmission in the i-th acquisition represents rounding up for represents the importance coefficient for the physical entity data transmission in the i-th acquisition represents the channel bit error rate during the physical entity data transmission in the i-th acquisition represents the logarithmic function B5: Obtain the feedback signal confirmation time of the data processing center ; If , it indicates that the data transmission is successful, stops the timer, and records the transmission time and the number of retransmissions of this transmission; If , then enter the retransmission operation.

6. The digital twin-based WAPI network optimization and simulation system according to claim 5, characterized in that The specific steps of the strategy of automatically retransmitting requests and dynamically adjusting the number of retransmissions further include: B6: Based on the calculated number of retransmissions , perform the retransmission operation of the data packet. Each time a retransmission occurs, restart the timer and wait for the acknowledgment feedback signal; If an acknowledgement feedback signal is received, stop retransmitting; If the acknowledgement feedback signal has not been received after reaching the upper limit of the number of retransmissions, record the failure of this transmission; B7: According to the set timer , re-measure the channel bit error rate and update it to , and dynamically adjust the data importance coefficient.

7. The WAPI network optimization and simulation system based on digital twin according to claim 6, characterized in that The data comparison unit adopts a difference detection algorithm. The specific steps of the difference detection algorithm include: D1: Collect the network performance index data in the digital twin model and the actual network operation data in the same time interval, and assign a unique performance index identifier to each index; D2: Starting from the first time point, compare each performance index in the network performance index data and the actual network operation data output by the digital twin model in turn; D3: For each performance metric, use the formula to calculate the difference value, where represents the difference value of performance metric j at time point t, represents the value of performance metric j output by the digital twin model at time point t, represents the value of performance metric j in the operation data of the actual network at time point t; D4: Store the difference value, the corresponding timestamp and the performance index identifier at each time point into a pre-constructed difference data list; D5: Use the difference method to calculate the trend change, and determine whether the trend is rising, falling or stable according to the trend calculation result; D6: Organize the analysis results of the difference value and the trend change into a report form.

8. The digital-twin-based WAPI network optimization and simulation system according to claim 7, characterized in that The network optimization strategy includes: adjusting the access point location, optimizing the channel allocation, enhancing network security, performing load balancing, and adjusting the transmission power.

Images