Wireless mouse strong interference source avoiding method based on wireless frequency hopping model

Abstract

The invention relates to the technical field of computer-aided communication, in particular to a wireless mouse strong interference source avoiding method based on a wireless frequency hopping model, which comprises the following steps: acquiring multi-band signal characteristics by monitoring electromagnetic spectrum data of an environment where a wireless mouse is located in real time; constructing an interference source analysis model by using an improved weighted fuzzy clustering algorithm, and accurately identifying the type and intensity distribution of an interference source; according to an analysis result, dynamically generating a frequency hopping strategy by using a channel state prediction model based on reinforcement learning, and determining priority sequences of a main channel and a standby channel; when the interference on the main channel exceeds a preset threshold value, performing real-time evaluation by means of a composite quality index and switching to an optimal standby channel; meanwhile, through a closed-loop feedback mechanism, according to the difference between actual channel quality and expected channel quality, a stochastic gradient descent method is adopted to optimize frequency hopping strategy parameters, and dynamic self-adaptive adjustment of interference avoidance is achieved. Complex electromagnetic environment interference can be effectively dealt with, and the communication stability of the wireless mouse and the use experience of a user are remarkably improved.

Description

Technical Field

[0001] The invention relates to the technical field of computer-aided communication, in particular to a method for avoiding strong interference sources of a wireless mouse based on a wireless frequency hopping model. Background Art

[0002] With the popularization of electronic devices and the widespread application of wireless communication technology, wireless mouse has become a common input device for computer users. In various work and life scenarios, people have increasing requirements for the user experience and stability of wireless mouse. However, the current wireless mouse faces severe electromagnetic interference challenges in actual use, including:

[0003] The electromagnetic environment is becoming increasingly complex: Modern office spaces are usually filled with a large number of wireless devices, such as wireless routers, Bluetooth speakers, wireless keyboards, etc., which work together within a limited electromagnetic spectrum. At the same time, there are more powerful and complex electromagnetic interference sources in industrial environments, such as motors, welding machines, high-frequency induction equipment, etc. The electromagnetic signals generated by these interference sources will be intertwined in space, causing the electromagnetic environment of the wireless mouse to become extremely complex. The frequency, strength and modulation mode of the interference signals are different, which seriously affects the signal transmission quality of the wireless mouse.

[0004] Traditional frequency hopping technology has limitations: To deal with interference, traditional wireless mice mostly use frequency hopping technology, that is, switching between multiple channels to find a relatively clean communication channel; but traditional frequency hopping technology is often based on a fixed frequency hopping sequence or a simple channel detection mechanism; on the one hand, the fixed frequency hopping sequence lacks adaptability to the real-time interference environment, and it is easy to switch repeatedly in the frequency band with severe interference, and it is impossible to effectively avoid strong interference sources; on the other hand, the simple channel detection mechanism is difficult to accurately identify the type and intensity distribution of complex interference sources, which may result in the selection of the interfered channel when selecting the backup channel, thus failing to ensure a stable communication connection;

[0005] Interference has a significant impact on the performance of wireless mice: when a wireless mouse is subject to strong interference, problems such as pointer drift, slow click response, or even complete loss of connection may occur; this not only seriously affects the user's work efficiency, such as in design drawing, data analysis, and other work that requires high mouse operation accuracy and real-time performance, the mouse anomalies caused by interference will cause deviations in work results or require repeated operations; in entertainment scenarios, such as during gaming, mouse delays and misoperations caused by interference will greatly reduce the user's gaming experience.

[0006] Therefore, in order to solve the above problems, a method for avoiding strong interference sources of wireless mouse based on wireless frequency hopping model is proposed. Summary of the invention

[0007] The purpose of the present invention is to provide a method for avoiding strong interference sources of a wireless mouse based on a wireless frequency hopping model to solve the problems raised in the above background technology.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A method for avoiding strong interference sources of a wireless mouse based on a wireless frequency hopping model comprises the following steps:

[0010] S1. Real-time monitoring of the electromagnetic spectrum data of the environment where the wireless mouse is located to obtain multi-band signal characteristics;

[0011] S2. Build an interference source analysis model based on the improved machine learning algorithm to analyze signal characteristics and identify the type and intensity distribution of interference sources;

[0012] S3. Dynamically generate a frequency hopping strategy based on the interference source analysis results and determine the priority sequence of the main channel and the backup channel;

[0013] S4. When the interference on the main channel exceeds a preset threshold, switch to the optimal backup channel based on the real-time channel quality evaluation result;

[0014] S5. Optimize the frequency hopping strategy parameters through a closed-loop feedback mechanism to achieve dynamic adaptive adjustment of interference avoidance.

[0015] As a preferred solution, the interference source analysis model in step S2 adopts an improved weighted fuzzy clustering algorithm IW-FCM, which specifically includes:

[0016] Construct a multi-dimensional feature vector set based on signal strength, spectrum occupancy, and modulation characteristics;

[0017] The cluster centers and membership matrices are optimized using the following objective function: ,in, is the weighted fuzzy clustering objective function; For the The dynamic weight factor of the interference source is positively correlated with the historical interference frequency; For sample For Class membership; For the Cluster centers of similar interference sources; is the fuzzy index, and its value range is ; is the total number of interference source categories; is the total number of samples; For the samples; represents the square of the Euclidean distance;

[0018] Output interference source classification results and confidence scores.

[0019] As a preferred solution, the dynamic weight factor The calculation method is: ,in, is the adjustment coefficient, and its value range is ; For the The frequency of occurrence of similar interference sources in historical data; is the maximum frequency of interference sources in historical data, used for normalization; is the average signal-to-noise ratio of the channel corresponding to this type of interference source; It is the maximum value of the average signal-to-noise ratio of the channel in the historical data, which is used for normalization.

[0020] As a preferred solution, the generation of the dynamic frequency hopping strategy in step S3 includes:

[0021] Construct a channel state prediction model based on reinforcement learning, Defining the state space ;

[0022] The channel selection strategy is iteratively optimized by the following Bellman equation: ,in, For the status Next select action The long-term benefit value is used to evaluate the pros and cons of taking a certain action in a certain state. is the learning rate; For immediate rewards; is the discount factor; To take action The next state to which the back channel transfers; For the next state possible actions to be taken; For the next state The maximum long-term benefit value that can be obtained from taking all possible actions;

[0023] Output the frequency hopping sequence to ensure that the interval between the backup channel and the frequency band of the current interference source is ≥15MHz.

[0024] As a preferred solution, the channel quality assessment in step S4 is implemented by the following formula: ,in, is the weighting coefficient, satisfying and , used to adjust the relative importance of each indicator in the composite quality indicator; is the signal-to-noise ratio of the channel; is the bit error rate of the channel; is the bit error rate threshold; is the signal fluctuation variance; is the signal fluctuation variance threshold.

[0025] As a preferred solution, the closed-loop feedback mechanism in step S5 includes:

[0026] Define the strategy optimization function ,in, Optimize the function for the strategy; is the frequency hopping strategy parameter vector; is the number of time steps, representing the total time length of the entire optimization process; is the current time step, ranging from 1 to ; is the attenuation factor; is the expected channel quality; is the actual channel quality;

[0027] Update the parameters using stochastic gradient descent: ,in, An updated frequency hopping strategy parameter vector; is the frequency hopping strategy parameter vector before updating; is the learning rate; It is the gradient of the loss function with respect to the parameters, indicating the direction of parameter update.

[0028] As a preferred solution, the method further includes the step of spatially locating the interference source:

[0029] Acquire signal arrival angle AoA information through dual antenna array;

[0030] Combined with the signal strength fingerprint library, the interference source coordinates are calculated using the following maximum likelihood estimation formula: ,in, is the estimated value of the interference source coordinates; It means to find the parameter value that makes the following expression reach the maximum value; is the number of sampling points; The number of the sampling point, ranging from 1 to ; For the The received signal strength of each sampling point; Predicted values ​​for the preset electromagnetic propagation model; is the standard deviation of measurement error.

[0031] It can be seen from the technical solution provided by the present invention that the method for avoiding strong interference sources of a wireless mouse based on a wireless frequency hopping model provided by the present invention has the following beneficial effects:

[0032] Efficient interference identification and response:

[0033] Accurate interference source analysis: Through improved machine learning algorithms, such as the improved weighted fuzzy clustering algorithm (IW-FCM), a multi-dimensional feature vector set is constructed by integrating signal strength, spectrum occupancy, modulation characteristics, etc., and dynamic weight factors are used to fully consider factors such as historical interference frequency and signal-to-noise ratio. The type and intensity distribution of interference sources can be accurately identified. Compared with traditional methods, this is more accurate in identifying interference sources in complex electromagnetic environments, greatly improving the reliability of interference source analysis.

[0034] Quick response to interference: Real-time monitoring of electromagnetic spectrum data. Once the interference on the main channel exceeds the preset threshold, it can quickly switch to the optimal backup channel based on the real-time channel quality assessment results. The whole process is fast and accurate, effectively reducing the impact of interference on the normal operation of the wireless mouse and ensuring the continuity and smoothness of operation.

[0035] Optimize frequency hopping strategy:

[0036] Dynamic frequency hopping generation: Build a channel state prediction model based on reinforcement learning, iteratively optimize the channel selection strategy through the Bellman equation, and output the frequency hopping sequence to ensure that the interval between the backup channel and the current interference source frequency band is ≥15MHz; this method of dynamically generating frequency hopping strategies can be flexibly adjusted according to real-time interference conditions, significantly improving the scientificity and effectiveness of channel selection and reducing the probability of interference;

[0037] Continuous strategy optimization: Using a closed-loop feedback mechanism, the frequency hopping strategy parameters are continuously optimized using the stochastic gradient descent method based on the difference between the actual channel quality and the expected channel quality. This allows the frequency hopping strategy to continuously improve as the environment changes, always maintaining the optimal or near-optimal state, greatly enhancing the adaptability of the wireless mouse in different interference environments.

[0038] Improve communication quality:

[0039] Composite quality assessment: The composite quality index Q is used to comprehensively consider factors such as signal-to-noise ratio, bit error rate, and signal fluctuation variance to evaluate channel quality, ensuring that the quality of the channel can be comprehensively and accurately judged when selecting a channel. Switching channels based on this can maximize the stability and reliability of wireless mouse communication, reduce data transmission errors and interruptions, and improve user experience.

[0040] Interference spatial positioning assistance: It has the function of spatial positioning of interference sources. It obtains the signal arrival angle (AoA) information through the dual antenna array, and calculates the coordinates of the interference source by combining the signal strength fingerprint library and the maximum likelihood estimation formula. This function helps to gain a deeper understanding of the distribution of interference sources, provides strong support for further optimizing frequency hopping strategies and interference avoidance measures, and fundamentally improves communication quality.

[0041] Wide applicability:

[0042] Multi-scenario application: This method is applicable to various complex electromagnetic environments. Whether it is the interference generated by numerous electronic devices in office places or the strong electromagnetic interference in industrial environments, wireless mice can work stably with this intelligent avoidance method, and have broad application prospects.

[0043] Device compatibility: The present invention focuses on the interference avoidance method of the wireless mouse, which makes little change to the hardware of the wireless mouse itself, is easily compatible with various existing wireless mouse products, is easy to promote and apply, and can significantly improve the performance of the wireless mouse without increasing too much cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 The present invention is a schematic diagram of the steps of a method for avoiding strong interference sources of a wireless mouse based on a wireless frequency hopping model. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0046] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0047] like Figure 1 As shown, an embodiment of the present invention provides a method for avoiding strong interference sources of a wireless mouse based on a wireless frequency hopping model, comprising the following steps:

[0048] S1. Real-time monitoring of the electromagnetic spectrum data of the environment where the wireless mouse is located to obtain multi-band signal characteristics;

[0049] S2. Build an interference source analysis model based on the improved machine learning algorithm to analyze signal characteristics and identify the type and intensity distribution of interference sources;

[0050] S3. Dynamically generate a frequency hopping strategy based on the interference source analysis results and determine the priority sequence of the main channel and the backup channel;

[0051] S4. When the interference on the main channel exceeds a preset threshold, switch to the optimal backup channel based on the real-time channel quality evaluation result;

[0052] S5. Optimize the frequency hopping strategy parameters through a closed-loop feedback mechanism to achieve dynamic adaptive adjustment of interference avoidance.

[0053] In this embodiment, the specific operation steps of step S1 include:

[0054] Step S1-1: Initialize monitoring equipment:

[0055] Turn on the electromagnetic spectrum monitoring device that comes with the wireless mouse to ensure that the device is working properly. Its operating frequency range should cover all frequency bands that the wireless mouse may use.

[0056] Calibrate the monitoring equipment to eliminate the impact of the equipment's own errors on the measurement results to ensure the accuracy of subsequent measurement data;

[0057] Step S1-2: Setting monitoring parameters:

[0058] Determine the monitoring time interval, for example, collect spectrum data every 100 milliseconds to achieve real-time monitoring;

[0059] Set the frequency range to be monitored, covering the main frequency band of the wireless mouse and adjacent frequency bands where interference may occur;

[0060] Step S1-3: Collecting electromagnetic spectrum data:

[0061] According to the set time interval and frequency band range, the monitoring equipment is used to collect data on the electromagnetic spectrum of the environment where the wireless mouse is located;

[0062] Record the basic parameters of each frequency band, such as signal strength, frequency, phase, etc. These parameters will be used for subsequent signal feature analysis;

[0063] Step S1-4: Data preprocessing:

[0064] Filter the collected raw spectrum data to remove noise and interference signals and improve the data quality;

[0065] Normalize the filtered data to unify the data of different frequency bands to the same scale to facilitate subsequent feature extraction and analysis;

[0066] Step S1-5: Extract multi-band signal features:

[0067] Extracting features of multi-band signals from preprocessed data, including but not limited to signal strength, spectrum occupancy, modulation features, etc.;

[0068] For signal strength, calculate the average signal strength and maximum signal strength of each frequency band; for spectrum occupancy, count the proportion of time each frequency band is occupied by the signal; for modulation characteristics, analyze the signal modulation mode (such as ASK, FSK, PSK, etc.) and modulation parameters;

[0069] Step S1-6: Storing and updating signal feature data:

[0070] The extracted multi-band signal feature data is stored in the database for subsequent interference source analysis and frequency hopping strategy generation;

[0071] As time goes by, the signal characteristic data is continuously updated to reflect the real-time changes in the environment where the wireless mouse is located.

[0072] In this embodiment, the specific operation steps of step S2 include:

[0073] Step S2-1: Data preparation:

[0074] Feature vector set construction:

[0075] From the multi-band signal features obtained in step S1, a multi-dimensional feature vector set is constructed according to the signal strength, spectrum occupancy, and modulation features; for example, for each signal sample, the signal strength is quantified into a numerical value, the spectrum occupancy is expressed as a ratio between 0 and 1, and the modulation feature is expressed by a specific code; assuming that there are n signal samples, each sample has m features, then an n×m feature matrix can be obtained, and each row of the matrix represents a feature vector of a signal sample;

[0076] Normalize the feature vector set to eliminate the impact of different dimensions between features. Common normalization methods include minimum-maximum normalization and Z-score normalization. Taking minimum-maximum normalization as an example, for each feature in the feature vector , we can use the formula Normalize, where and are the minimum and maximum values ​​of the feature respectively;

[0077] Historical data collection:

[0078] Collect past interference source data, including signal characteristics corresponding to different types of interference sources and their occurrence frequencies; these historical data can be used to calculate subsequent dynamic weight factors;

[0079] Step S2-2: Determine the parameters of the improved weighted fuzzy clustering algorithm (IW-FCM):

[0080] Fuzzy index m selection:

[0081] In the range of 1.5≤m≤3.0, select the appropriate fuzzy index m according to the specific application scenario and data characteristics. Generally, the larger the m value, the more fuzzy the clustering result; the smaller the m value, the closer the clustering result is to hard clustering. Multiple experiments can be conducted to compare the clustering effects under different m values ​​and then select the optimal m value.

[0082] The adjustment coefficient λ is determined by:

[0083] The adjustment coefficient λ is determined within the range of 0.6≤λ≤0.9; λ is used to adjust the weights of the historical interference frequency and the signal-to-noise ratio in the calculation of the dynamic weight factor; the appropriate λ value can be determined in combination with actual conditions, such as the stability of the interference source and the change in channel quality;

[0084] Step S3-3: Calculate dynamic weight factor :

[0085] Historical frequency statistics:

[0086] Based on historical data, the The frequency of occurrence of similar interference sources in historical data For example, in the past 100 monitorings, If the interference source appears 20 times, then ;

[0087] Find the maximum frequency of all interference sources in historical data ;

[0088] Average signal-to-noise ratio calculation:

[0089] Calculate the Average signal-to-noise ratio of the channel corresponding to the interference source ; It can be obtained by measuring the signal-to-noise ratio of the corresponding channel when this type of interference source appears multiple times and calculating the average value;

[0090] Find the maximum value of the average signal-to-noise ratio of the channel corresponding to all interference sources in the historical data ;

[0091] Dynamic weight factor calculation:

[0092] Using the formula Calculate the Dynamic weight factor of similar interference source ,in, is the adjustment coefficient, and its value range is ; For the The frequency of occurrence of similar interference sources in historical data; is the maximum frequency of interference sources in historical data, used for normalization; is the average signal-to-noise ratio of the channel corresponding to this type of interference source; is the maximum value of the channel average signal-to-noise ratio in historical data, used for normalization;

[0093] Step S2-4: Initialize cluster centers and membership matrices:

[0094] Initialization of cluster centers:

[0095] Randomly select c samples as the initial cluster centers ( ), where c is the number of pre-set interference source categories;

[0096] Initialization of membership matrix:

[0097] For each sample ( ), randomly initialize its Class membership , while satisfying and ;

[0098] Step S2-5: Iteratively optimize cluster centers and membership matrices:

[0099] Objective function calculation:

[0100] Using the formula Calculate the weighted fuzzy clustering objective function The value of is the weighted fuzzy clustering objective function; For the The dynamic weight factor of the interference source is positively correlated with the historical interference frequency; For sample For Class membership; For the Cluster centers of similar interference sources; is the fuzzy index, and its value range is ; is the total number of interference source categories; is the total number of samples; For the samples; represents the square of the Euclidean distance;

[0101] Membership matrix update:

[0102] Update the membership matrix according to the following formula: ;

[0103] Cluster center update:

[0104] The cluster center is updated using the following formula: ;

[0105] Convergence judgment:

[0106] Repeat the steps of objective function calculation, membership matrix update and cluster center update until the objective function The change is less than a preset threshold (e.g. 10 -6 ), or the maximum number of iterations is reached, at which point the clustering process is considered to have converged;

[0107] Step S2-6: Output interference source classification results and confidence scores:

[0108] Interference source classification:

[0109] For each sample , classify it into the category with the largest membership, thus obtaining the classification result of the interference source;

[0110] Confidence score calculation:

[0111] The confidence score is calculated based on the sample's membership in the class to which it belongs; for example, if the sample For The class membership is , then you can As the confidence score of the sample classification;

[0112] The confidence scores of all samples of each type of interference source are averaged to obtain the overall confidence score of the classification of this type of interference source.

[0113] In this embodiment, the specific operation steps of step S3 are as follows:

[0114] Step S3-1: Data preparation:

[0115] Obtain interference source analysis results: Obtain information such as the type, intensity distribution, classification results, and confidence score of the interference source from step S2; this information will serve as an important basis for generating a frequency hopping strategy;

[0116] Define channel set: define all channels that the wireless mouse can use to form a channel set ; This set covers the primary channel and possible backup channels;

[0117] Step S3-2: Construct a channel state prediction model based on reinforcement learning:

[0118] Defining the state space :

[0119] Setting the state space ; For each channel, its current state is determined according to the interference source analysis result; for example, if the interference intensity of the channel is lower than a preset low threshold, it is determined to be in an idle state; if the interference intensity is between the low threshold and the high threshold, it is in a light interference state; if it is higher than the high threshold, it is in a heavy interference state;

[0120] Define the action space:

[0121] Action Space Channel Set , that is, each action Indicates selecting a channel;

[0122] initialization surface:

[0123] Create one of surface , used to store in state Next select action The long-term return value; initially, All values ​​in the table are set to 0;

[0124] Step S3-3: Setting reinforcement learning parameters:

[0125] Learning Rate choose:

[0126] exist In the range of ; The learning rate controls each update When the step size is , a larger learning rate can make the algorithm converge faster, but may cause unstable convergence; a smaller learning rate converges slower but is more stable;

[0127] Discount Factor Sure:

[0128] exist Determine the discount factor within the range ; The discount factor is used to weigh the importance of immediate rewards and future rewards. Indicates that future rewards are more important, smaller They focus more on immediate rewards;

[0129] Step S3-4: Calculate instant reward :

[0130] Defining Channel Quality Index

[0131] Define a channel quality index based on the strength and type of the interference source. ; For example, we can consider signal-to-noise ratio, bit error rate and other factors to calculate ;

[0132] Calculate instant rewards :

[0133] According to the channel quality index Calculate instant rewards ; For example, if If the value is higher than a certain excellent threshold, a higher positive reward is given; if If the result is below a certain threshold, a negative reward is given; if the result is between the two, a smaller positive reward or zero reward is given.

[0134] Step S3-5: Iteratively optimize the channel selection strategy:

[0135] Environment Interaction:

[0136] From the current state Start, according to Table Select an Action (i.e. select a channel); you can use - Greedy strategy, with a certain probability Randomly select an action to The probability of choosing The action with the largest value;

[0137] Execute an action After that, the environment will feedback the next state and instant rewards ;

[0138] renew surface:

[0139] Using the Bellman equation renew Table, in which For the status Next select action The long-term benefit value is used to evaluate the pros and cons of taking a certain action in a certain state. is the learning rate; For immediate rewards; is the discount factor; To take action The next state to which the back channel transfers; For the next state possible actions to be taken; For the next state The maximum long-term benefit value that can be obtained from taking all possible actions;

[0140] Convergence judgment:

[0141] Repeated environmental interaction and Table update steps until The table converges, that is The change in value is less than the preset threshold, or the maximum number of iterations is reached;

[0142] Step S3-6: Generate frequency hopping sequence:

[0143] Determine the primary channel:

[0144] Based on the final converged Q table, the channel corresponding to the action with the largest Q value in the current state is selected as the main channel;

[0145] Filter alternate channels:

[0146] From channel collection Filter out channels with a frequency interval of ≥15MHz from the current interference source as candidate sets of backup channels;

[0147] Determine the alternate channel priority sequence:

[0148] For each channel in the candidate set of backup channels, according to its Sort the values, The higher the value, the higher the priority of the channel, thus determining the priority sequence of the backup channels;

[0149] Step S3-7: Storing and updating frequency hopping strategy:

[0150] Storage frequency hopping strategy:

[0151] The priority sequence of the primary channel and the backup channel is stored in a database or a configuration file for subsequent use;

[0152] Update frequency hopping strategy in real time:

[0153] As time goes by and the environment changes, steps S3-2 to S3-6 are continuously repeated to update the frequency hopping strategy in real time to adapt to new interference situations.

[0154] In this embodiment, the specific operation steps of step S4 include:

[0155] Step S4-1: Continuously monitor the interference situation of the main channel:

[0156] To set interference monitoring parameters:

[0157] Determine the specific parameters used to monitor the interference of the main channel, such as signal strength, signal-to-noise ratio (SNR), bit error rate (BER), etc. These parameters can directly reflect the degree of interference to the main channel;

[0158] Set the time interval for interference monitoring, for example, collect interference data every 200 milliseconds to ensure real-time understanding of the interference situation of the main channel;

[0159] Collect main channel interference data:

[0160] Using the monitoring equipment that comes with the wireless mouse, collect interference data on the main channel according to the set time intervals and monitoring parameters;

[0161] Record the interference parameter values ​​collected each time to provide data support for subsequent judgment of whether they exceed the preset threshold;

[0162] Step S4-2: Determine whether the main channel interference exceeds a preset threshold:

[0163] Determine the preset threshold:

[0164] According to the working requirements and actual application scenarios of the wireless mouse, set corresponding preset thresholds for each monitoring parameter; for example, set the preset threshold of the signal-to-noise ratio to 20dB, and the preset threshold of the bit error rate to ;

[0165] Comparative judgment:

[0166] Compare the main channel interference parameter values ​​collected in real time with the preset thresholds; if any monitoring parameter value exceeds its corresponding preset threshold, it is determined that the main channel is interfered with beyond the preset threshold and channel switching is required;

[0167] Step S4-3: Real-time evaluation of backup channel quality:

[0168] Determine composite quality indicators Parameters:

[0169] Clarify composite quality indicators (in, is the weighting coefficient, satisfying and , used to adjust the relative importance of each indicator in the composite quality indicator; is the signal-to-noise ratio of the channel; is the bit error rate of the channel; is the bit error rate threshold; is the signal fluctuation variance; is the value of each parameter in the signal fluctuation variance threshold); Defaults to , Defaults to , and determine the weighting coefficient according to the actual situation , must meet and ;

[0170] Collect backup channel data:

[0171] For each backup channel in the backup channel candidate set determined in step S3, collect its signal-to-noise ratio (SNR), bit error rate (BER) and signal fluctuation variance And other data;

[0172] Calculate the composite quality index of the backup channel :

[0173] Substitute the collected data of each spare channel into the composite quality index formula to calculate the quality of each spare channel. value;

[0174] Step S4-4: Select the best backup channel:

[0175] Compare alternate channel quality:

[0176] Composite quality index for all spare channels The values ​​are compared;

[0177] Determine the optimal backup channel:

[0178] choose The backup channel with the largest value is used as the optimal backup channel, which has the best communication quality under the current situation;

[0179] Step S4-5: Switch to the optimal backup channel:

[0180] Send the switch command:

[0181] The control module of the wireless mouse sends a channel switching instruction to the communication module, instructing it to switch from the main channel to the optimal backup channel;

[0182] Complete channel switching:

[0183] After receiving the switching instruction, the communication module adjusts the operating frequency, establishes a communication connection with the optimal backup channel, and completes the channel switching process;

[0184] Step S4-6: Record channel switching information:

[0185] Record switching details:

[0186] Record the channel switching related information, such as the switching time, the original main channel information, the optimal backup channel information switched to, etc., into a log file or database;

[0187] Information analysis and feedback:

[0188] Analyze the recorded channel switching information to provide data reference for subsequent frequency hopping strategy optimization and interference avoidance, and further improve the system's anti-interference capability.

[0189] In this embodiment, the specific steps of step S5 include:

[0190] Step S5-1: Define strategy optimization related parameters:

[0191] Determine the frequency hopping strategy parameter vector :

[0192] Identify the parameters that need to be optimized in the frequency hopping strategy and combine them into a vector ; These parameters may include but are not limited to the generation rules of the frequency hopping sequence, the selection criteria of the primary channel and the backup channel, the time interval of the frequency hopping, etc.;

[0193] Set expected channel quality :

[0194] Determine an expected channel quality indicator based on the usage scenario and performance requirements of the wireless mouse ; This indicator can be based on the composite quality indicator To set, for example, for a higher value to ensure the stability and reliability of communication;

[0195] Determine the attenuation factor :

[0196] exist In the range of ; Used to weight the errors at different time steps so that the impact of recent errors is greater, thereby responding to environmental changes more promptly;

[0197] Step S5-2: Real-time monitoring of actual channel quality :

[0198] Collect channel data:

[0199] Continuously collect relevant parameters of the currently used channel, including signal-to-noise ratio (SNR), bit error rate (BER) and signal fluctuation variance wait;

[0200] Calculate the actual channel quality :

[0201] Substitute the collected channel data into the composite quality index formula: , calculate the actual quality of the current channel ;

[0202] Step S5-3: Calculate the strategy optimization function :

[0203] Determine the number of time steps :

[0204] Determine an appropriate time step based on the system's operating time and data acquisition frequency For example, if the system collects channel data every 100 milliseconds, If it is set to 100, it means that the channel quality data within the past 10 seconds is considered;

[0205] Calculate the sum of squared errors:

[0206] For each time step ( ),calculate , and according to the attenuation factor Weighted, that is ;

[0207] The sum is obtained to get the policy optimization function value:

[0208] Add the weighted squared errors of all time steps to get the policy optimization function (in, Optimize the function for the strategy; is the frequency hopping strategy parameter vector; is the number of time steps, representing the total time length of the entire optimization process; is the current time step, ranging from 1 to ; is the attenuation factor; is the expected channel quality; is the actual channel quality);

[0209] Step S5-4: Calculate the gradient of the loss function with respect to the parameters :

[0210] Select the gradient calculation method:

[0211] Gradients can be calculated using numerical methods (such as finite difference methods) or analytical methods (if the loss function is differentiable) ; The finite difference method approximates the gradient by calculating the difference of the function when the parameter changes slightly; the analytical method directly differentiates the loss function to obtain the gradient expression;

[0212] Calculate the gradient value:

[0213] Calculate the loss function based on the selected gradient calculation method Frequency hopping strategy parameter vector Gradient ;

[0214] Step S5-5: Update frequency hopping strategy parameters :

[0215] Determining the learning rate :

[0216] exist Choose a suitable learning rate within the range ; The learning rate controls the step size of each parameter update. A larger learning rate can make the parameters update faster, but may cause unstable convergence; a smaller learning rate will converge more slowly but more stably;

[0217] Update the parameter vector:

[0218] Update formula using stochastic gradient descent (in, An updated frequency hopping strategy parameter vector; is the frequency hopping strategy parameter vector before updating; is the learning rate; is the gradient of the loss function to the parameter, indicating the direction of parameter update), for the frequency hopping strategy parameter vector Make updates;

[0219] Step S5-6: Apply the updated frequency hopping strategy:

[0220] Update the frequency hopping policy settings:

[0221] The updated frequency hopping strategy parameters Applied to the frequency hopping strategy of wireless mouse, adjusting the generation of frequency hopping sequence, channel selection and other related operations;

[0222] Continuous monitoring and optimization:

[0223] Repeat steps S5-2 to S5-6, continuously monitor the actual channel quality, calculate the strategy optimization function, update the frequency hopping strategy parameters, and implement dynamic adaptive adjustment of interference avoidance, so that the wireless mouse can maintain good communication performance in different interference environments.

[0224] In this example, the method further includes a step of spatially locating the interference source:

[0225] Acquire signal arrival angle AoA information through dual antenna array;

[0226] Combined with the signal strength fingerprint library, the interference source coordinates are calculated using the following maximum likelihood estimation formula: ,in, is the estimated value of the interference source coordinates; It means to find the parameter value that makes the following expression reach the maximum value; is the number of sampling points; The number of the sampling point, ranging from 1 to ; For the The received signal strength of each sampling point; Predicted values ​​for the preset electromagnetic propagation model;

[0227] is the standard deviation of measurement error.

[0228] Although 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 the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for avoiding strong interference sources of a wireless mouse based on a wireless frequency hopping model, characterized in that: The following steps are involved: S1. Real-time monitoring of the electromagnetic spectrum data of the environment where the wireless mouse is located to obtain multi-band signal characteristics; S2. Building an interference source analysis model based on an improved machine learning algorithm, analyzing the signal characteristics, and identifying the type and intensity distribution of interference sources; S3. Dynamically generate a frequency hopping strategy based on the interference source analysis results and determine the priority sequence of the main channel and the backup channel; S4. When the interference on the main channel exceeds a preset threshold, switch to the optimal backup channel based on the real-time channel quality evaluation result; S5. Optimize the frequency hopping strategy parameters through a closed-loop feedback mechanism to achieve dynamic adaptive adjustment of interference avoidance.

2. The method for avoiding strong interference sources of a wireless mouse based on a wireless frequency hopping model according to claim 1, characterized in that: The interference source analysis model in step S2 adopts an improved weighted fuzzy clustering algorithm IW-FCM, which specifically includes: Construct a multi-dimensional feature vector set based on signal strength, spectrum occupancy, and modulation characteristics; The cluster centers and membership matrices are optimized using the following objective function: ,in, is the weighted fuzzy clustering objective function; For the The dynamic weight factor of the interference source is positively correlated with the historical interference frequency; For sample For Class membership; For the Cluster centers of similar interference sources; is the fuzzy index, and its value range is ; is the total number of interference source categories; is the total number of samples; For the samples; represents the square of the Euclidean distance; Output interference source classification results and confidence scores.

3. The method for avoiding strong interference sources of a wireless mouse based on a wireless frequency hopping model according to claim 2 is characterized in that: The dynamic weight factor The calculation method is: ,in, is the adjustment coefficient, and its value range is ; For the The frequency of occurrence of similar interference sources in historical data; is the maximum frequency of interference sources in historical data, used for normalization; is the average signal-to-noise ratio of the channel corresponding to this type of interference source; It is the maximum value of the average signal-to-noise ratio of the channel in the historical data, which is used for normalization.

4. The method for avoiding strong interference sources of a wireless mouse based on a wireless frequency hopping model according to claim 1, characterized in that: The dynamic frequency hopping strategy generation in step S3 includes: Construct a channel state prediction model based on reinforcement learning, Defining the state space ; The channel selection strategy is iteratively optimized by the following Bellman equation: ,in, For the status Next select action The long-term benefit value is used to evaluate the pros and cons of taking a certain action in a certain state. is the learning rate; For immediate rewards; is the discount factor; To take action The next state to which the back channel transfers; For the next state possible actions to be taken; For the next state The maximum long-term benefit value that can be obtained from taking all possible actions; Output the frequency hopping sequence to ensure that the interval between the backup channel and the frequency band of the current interference source is ≥15MHz.

5. The method for avoiding strong interference sources of a wireless mouse based on a wireless frequency hopping model according to claim 1, characterized in that: The channel quality assessment in step S4 is implemented by the following formula: ,in, is the weighting coefficient, satisfying and , used to adjust the relative importance of each indicator in the composite quality indicator; is the signal-to-noise ratio of the channel; is the bit error rate of the channel; is the bit error rate threshold; is the signal fluctuation variance; is the signal fluctuation variance threshold.

6. The method for avoiding strong interference sources of a wireless mouse based on a wireless frequency hopping model according to claim 1, characterized in that: The closed-loop feedback mechanism in step S5 includes: Define the strategy optimization function ,in, Optimize the function for the strategy; is the frequency hopping strategy parameter vector; is the number of time steps, representing the total time length of the entire optimization process; is the current time step, ranging from 1 to ; is the attenuation factor; is the expected channel quality; is the actual channel quality; Update the parameters using stochastic gradient descent: ,in, An updated frequency hopping strategy parameter vector; is the frequency hopping strategy parameter vector before updating; is the learning rate; It is the gradient of the loss function with respect to the parameters, indicating the direction of parameter update.

7. The method for avoiding strong interference sources of a wireless mouse based on a wireless frequency hopping model according to claim 1, characterized in that: It also includes the steps of spatial positioning of interference sources: Acquire signal arrival angle AoA information through dual antenna array; Combined with the signal strength fingerprint library, the interference source coordinates are calculated using the following maximum likelihood estimation formula: ,in, is the estimated value of the interference source coordinates; It means to find the parameter value that makes the following expression reach the maximum value; is the number of sampling points; The number of the sampling point, ranging from 1 to ; For the The received signal strength of each sampling point; Predicted values ​​for the preset electromagnetic propagation model; is the standard deviation of measurement error.

Images