Self-adaptive adjustment method and system based on portable wifi channel

Abstract

The invention discloses a self-adaptive adjustment method and system based on a portable wifi channel, relates to the technical field of communication, and solves the problem that a channel performance index, a network performance index and an environmental influence index are difficult to analyze so as to generate a channel quality comprehensive score; time sequence analysis and a deep learning model are difficult to predict a flow demand peak value; the optimal channel is difficult to dynamically select by using a reinforcement learning algorithm; the reinforcement learning technology and a deep learning model are difficult to be combined to identify whether congestion occurs or not and dynamically adjust a channel; and the channel adjustment effect is difficult to monitor in real time and the grade is difficult to evaluate. According to the method, the communication quality and the network utilization rate of the portable WiFi equipment are improved by monitoring the channel state in real time, comprehensively evaluating the channel quality, predicting the flow demand, dynamically selecting the optimal channel, identifying and adjusting the channel congestion and evaluating the adjusting effect, and the method adapts to a complex and changeable communication environment.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular relates to a method and system for adaptively adjusting a portable Wi-Fi channel. Background Art

[0002] With the rapid development of mobile internet, portable Wi-Fi devices, due to their portability and flexibility, have become an essential tool for users to maintain network connectivity in various scenarios. However, when using portable Wi-Fi devices, channels are susceptible to interference from the surrounding environment, leading to channel congestion and impacting communication quality. Therefore, in complex and changing wireless environments, portable Wi-Fi devices require real-time channel monitoring to ensure stable network performance and user experience. By monitoring the portable Wi-Fi channel in real time, adaptive channel adjustments are implemented using a combination of technologies such as time series analysis, deep learning models, and reinforcement learning algorithms.

[0003] The existing adaptive adjustment methods for portable Wi-Fi channels have the following problems: it is difficult to analyze the channel performance index, network performance index, and environmental impact index to generate a comprehensive channel quality score; it is difficult to combine time series analysis and deep learning models to predict traffic demand peaks; it is difficult to use reinforcement learning algorithms to dynamically select the optimal channel; it is difficult to combine reinforcement learning technology with deep learning models to identify congestion and dynamically adjust the channel; and it is difficult to monitor the channel adjustment effect in real time and evaluate the level. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a method and system for adaptively adjusting a portable Wi-Fi channel to solve the above-mentioned problem. To this end, a first aspect of the present invention provides a method for adaptively adjusting a portable Wi-Fi channel, comprising the following steps: S1: Real-time monitoring and collection of channel monitoring data and environmental impact data of the portable Wi-Fi channel; S2: Preprocess the collected channel monitoring data and environmental impact data by cleaning and standardizing them respectively; S3: Analyze and calculate the channel performance index, network performance index, and environmental impact index based on preprocessed channel monitoring data and environmental impact data. Perform a real-time assessment of channel quality based on the channel performance index, network performance index, and environmental impact index to generate a comprehensive score for the portable Wi-Fi channel quality. Use time series analysis combined with deep learning models to predict user data traffic demands and calculate peak traffic demand. S4: Utilizes a reinforcement learning algorithm to dynamically select the optimal channel based on channel quality scores and predicted peak traffic demand. Based on the selected optimal channel, reinforcement learning technology combined with a deep learning model is used to identify congestion and dynamically adjust the channel. S5: Monitor the adjusted optimal channel in real time, calculate the effect score and evaluate the effect level of the optimal channel.

[0005] Furthermore, the channel monitoring data in step S1 includes: the number of devices on the channel, device movement speed, signal strength, channel occupancy, device battery power, device temperature and network traffic data; the environmental impact data includes: ambient temperature, ambient humidity, ambient noise value, air quality index and light intensity; among them, the network traffic data includes: bandwidth utilization, data packet loss rate, network throughput and delay, and transmission efficiency.

[0006] Furthermore, the step S3 of analyzing and calculating the channel performance index, the network performance index and the environmental impact index based on the pre-processed channel monitoring data and the environmental impact data includes the following steps: Calculate the channel performance index formula: Get the channel performance index Y, where is the signal strength of the i-th device; and Respectively represent the maximum and minimum values ​​of signal strength; is the total number of devices; is the moving speed of the i-th device; is the device temperature of the i-th device; is the maximum value of the device temperature; Z is the channel occupancy rate; Indicates the battery power usage of the i-th device; is the maximum value of battery power consumption; is the battery life of the i-th device; Calculate the network performance index formula by: The network performance index W is obtained, where k is the bandwidth utilization; d is the packet loss rate; L is the network throughput; c is the delay; x is the transmission efficiency; By calculating the environmental impact index formula: The environmental impact index H is obtained, where is the ambient temperature; is the ambient humidity; β is the ambient noise value; I is the air quality index, and Q is the light intensity.

[0007] Furthermore, in step S3, the channel quality is comprehensively evaluated in real time based on the channel performance index, the network performance index, and the environmental impact index to generate a comprehensive score of the portable Wi-Fi channel quality, including the following steps: Formula for calculating comprehensive channel quality score: Get the comprehensive channel quality score , where Y, W and H represent the channel performance index, network performance index and environmental impact index respectively; 、 、 Represent the weight coefficients of channel performance index, network performance index and environmental impact index respectively; is the channel occupancy; is the adjustment coefficient of channel occupancy, which is 0.6, and t is the time.

[0008] Furthermore, in step S3, the user's data traffic demand is predicted by combining time series analysis with a deep learning model, and the traffic demand peak is calculated, including the following steps: Collect historical time series data sets, including: historical traffic consumption time series, device movement GPS coordinate time series, environmental noise intensity, temperature time series data, humidity time series data, number of online devices, and channel occupancy rate; preprocess the collected historical time series data, including data cleaning and standardization, and aligning by timestamp to form a unified time step sequence; Combining time series analysis with deep learning models, we constructed a spatiotemporal attention network model to predict user data traffic needs. In the temporal dimension, we used a long short-term memory (LSTM) network to process traffic sequences and extract long-term dependency features. In the spatial dimension, we converted the device's GPS coordinate time series into a trajectory graph by connecting consecutive GPS points as edges, with edge weights calculated as the inverse of the Euclidean distance. We then used a graph neural network to model device movement trajectories and capture spatial correlations. The preprocessed time series data set is input into the spatiotemporal attention network model for training. Based on the trained spatiotemporal attention network model, the time series data collected and preprocessed through real-time monitoring is input into the trained spatiotemporal attention network model, and the output prediction result is the data traffic demand prediction value and timestamp. The prediction results are smoothed using the moving average method or the exponential smoothing method. Based on the smoothed prediction results, the peak point of traffic demand, that is, the time point when the traffic demand reaches its maximum value and the corresponding traffic value, are identified, and the peak traffic demand value is finally calculated.

[0009] Furthermore, in step S4, a reinforcement learning algorithm is used to dynamically select the optimal channel based on the comprehensive channel quality score and the peak traffic demand, including the following steps: Based on the traffic demand peak predicted and calculated by the spatiotemporal attention network model, channels with bandwidth capacity lower than the traffic demand peak are excluded from the portable Wi-Fi channels. According to the peak traffic demand, the weight of the comprehensive channel quality score is dynamically adjusted; during the peak traffic demand period, 0.6, 0.3, is 0.1; during non-peak traffic demand periods, 0.4, 0.4, is 0.2; Based on the assigned weights, the comprehensive channel quality scores for different time periods are calculated and ranked in descending order as candidate channels. Among the candidate channels, the channel with the highest comprehensive channel quality score is selected as the optimal channel. The Q-learning algorithm, a reinforcement learning algorithm, is used to switch to the optimal channel. Each portable Wi-Fi device acts as an intelligent agent. In the Q-learning algorithm, the action-value function Q(s, a) is learned for each state and action pair. The agent then chooses action a to switch to the optimal channel based on the current channel state s. The agent performs the selected action and switches the portable WiFi device to the optimal channel; after switching the portable WiFi device to the selected optimal channel, communication begins; During the switching process, communication on the current channel is maintained and unsent data packets are cached. After the switch, the optimal channel is monitored in real time, and the immediate reward and long-term reward are calculated based on the reward function of the Q-learning algorithm. Based on the immediate reward and long-term reward, the corresponding Q value in the Q-value table is updated. By evaluating the convergence of the Q-value table, training is stopped when convergence is achieved, otherwise iterative learning continues.

[0010] Furthermore, in step S4, based on the selected optimal channel, reinforcement learning technology is combined with a deep learning model to identify whether it is congested and dynamically adjust the channel, including the following steps: Based on the selected optimal channel, channel scanning technology is used to collect the optimal channel status information in real time, including: channel utilization, bandwidth utilization, channel capacity, switching delay, and data loss; the collected data is pre-processed, including data cleaning and normalization; Combining reinforcement learning technology with deep learning models, a congestion identification model based on deep reinforcement learning is constructed. A historical optimal channel state information set, labeled with whether channel congestion occurs, is input into the congestion identification model based on deep reinforcement learning for training. The optimal channel state information set, collected and preprocessed in real time, is input into the trained congestion identification model based on deep reinforcement learning to output the congestion identification results. If channel congestion is not detected, communication continues on the current channel. If channel congestion is detected, a binary exponential backoff algorithm is used. Based on the binary exponential backoff algorithm, an adaptive mechanism is introduced to dynamically adjust the backoff window size based on the real-time channel status information and congestion level. The congestion level is calculated by taking the weighted average of channel utilization, bandwidth utilization, channel capacity, switching delay, and data loss. Backoff window dynamic adjustment formula: in, is the size of the backoff window when the kth channel conflict or congestion occurs; is the level of congestion; When the congestion level is greater than the congestion level threshold, the dynamic adjustment mechanism of the backoff window size is triggered, and the backoff window size is adjusted to ; Otherwise, adjust the backoff window size to ;in, The congestion threshold is 0.65, is the maximum congestion level.

[0011] Furthermore, in step S5, real-time monitoring of the adjusted optimal channel, calculation of the effect score, and evaluation of the effect level of the optimal channel are performed, including the following steps: Monitor the channel monitoring data and environmental impact data of the adjusted optimal channel in real time, and substitute them into the channel quality comprehensive scoring formula to calculate the channel quality comprehensive score; set the dynamic evaluation cycle to trigger once every 5 seconds to evaluate the effect; calculate the effect score formula: Get performance rating ;in, Indicates the comprehensive score of channel quality before channel adjustment; Indicates the comprehensive score of channel quality after channel adjustment; is the degree of congestion; is the congestion level threshold; According to the effect score, it is divided into three levels: excellent, good and poor; When the grade is excellent; when When When the evaluation level is poor, return to step S4 and re-adjust the channel selection.

[0012] Compared with the prior art, the present invention has the following beneficial effects: This method monitors and collects real-time channel monitoring data and environmental impact data from portable Wi-Fi channels, enabling timely access to the latest channel status and providing accurate data support for subsequent channel quality assessment and optimization. By comprehensively analyzing the channel performance index, network performance index, and environmental impact index, it can comprehensively assess channel quality and generate a more accurate overall score.

[0013] This method uses time series analysis combined with deep learning models to predict user data traffic demands. This allows for early prediction of peak traffic demand, providing a basis for the rational allocation of channel resources. It also uses a reinforcement learning algorithm to dynamically select the optimal channel. Combining reinforcement learning with deep learning models to identify channel congestion allows for timely channel adjustments to avoid congestion, improving communication efficiency and user experience.

[0014] The present invention can evaluate the effectiveness of channel adjustment, optimize communication quality, reduce interference, improve network utilization, and adapt to the ever-changing communication environment by real-time monitoring of the optimal channel before and after adjustment and analyzing the adjustment effect score. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 is a flow chart of the method of the present invention; Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0017] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0018] See also Figure 1 As shown, the first embodiment of the present invention provides an adaptive adjustment method based on a portable Wi-Fi channel, comprising the following steps: S1: Real-time monitoring and collection of channel monitoring data and environmental impact data of the portable Wi-Fi channel; S2: Preprocess the collected channel monitoring data and environmental impact data by cleaning and standardizing them respectively; S3: Analyze and calculate the channel performance index, network performance index, and environmental impact index based on preprocessed channel monitoring data and environmental impact data. Perform a real-time assessment of channel quality based on the channel performance index, network performance index, and environmental impact index to generate a comprehensive score for the portable Wi-Fi channel quality. Use time series analysis combined with deep learning models to predict user data traffic demands and calculate peak traffic demand. S4: Utilizes a reinforcement learning algorithm to dynamically select the optimal channel based on channel quality scores and predicted peak traffic demand. Based on the selected optimal channel, reinforcement learning technology combined with a deep learning model is used to identify congestion and dynamically adjust the channel. S5: Monitor the adjusted optimal channel in real time, calculate the effect score and evaluate the effect level of the optimal channel.

[0019] Specifically, the system monitors and collects channel monitoring data and environmental impact data from portable Wi-Fi channels in real time. Environmental impact data is collected using various sensors, including temperature and humidity sensors. Data cleaning involves removing outliers such as sudden changes in signal strength and abnormally high bit error rates, supplementing missing values ​​using sliding averages or interpolation, and normalizing data of varying dimensions to the same dimension using Z-score or Min-Max normalization. The system analyzes and calculates the channel performance index, network performance index, and environmental impact index based on the preprocessed channel monitoring and environmental impact data, and uses this to comprehensively assess channel quality. Time series analysis combined with deep learning models is used to predict user data traffic demand and calculate peak traffic demand, providing a basis for subsequent channel adjustments. Using the reinforcement learning algorithm Q-learning, the system selects the optimal channel based on the channel quality score and peak traffic demand. Reinforcement learning technology is used to adjust channels in real time, identifying congested channels and switching to more optimal channels. Channel performance after adjustment is monitored in real time, and an effectiveness score is calculated. The channel adjustment results are then categorized as excellent, good, and poor based on the effectiveness score.

[0020] In this embodiment, the channel monitoring data in step S1 includes: the number of devices on the channel, device movement speed, signal strength, channel occupancy, device battery power, device temperature and network traffic data; the environmental impact data includes: ambient temperature, ambient humidity, ambient noise value, air quality index and light intensity; among them, the network traffic data includes: bandwidth utilization, data packet loss rate, network throughput and delay, and transmission efficiency.

[0021] Specifically, in channel monitoring data collection, the number of devices is calculated by real-time counting the number of devices on the current channel; device movement speed is estimated using GPS or Wi-Fi positioning technology; signal strength is measured by real-time receiving signal strength indicators collected through the antenna interface; channel occupancy is measured by the channel utilization counter to calculate the proportion of time the channel is occupied; device battery level is obtained through the device's battery management module; device temperature is collected using the device's built-in temperature sensor; in network traffic data, bandwidth utilization is measured by counting the bandwidth usage of the current channel through the network interface; packet loss rate is calculated by counting the proportion of packet loss using network protocol stack statistics; network throughput is calculated by data transmission rate; latency is measured using Ping or TCP handshake time; and transmission efficiency is calculated by the ratio of actual transmission rate to theoretical maximum rate. In environmental impact data collection, ambient temperature is collected using the temperature sensor built into the portable WiFi device; ambient humidity is collected using a humidity sensor; ambient noise levels are collected using an electromagnetic interference detector; air quality index is collected using an air quality sensor or environmental monitoring station; and light intensity is collected using a photosensor.

[0022] In this embodiment, the step S3 includes the following steps: Calculate the channel performance index formula: Get the channel performance index Y, where is the signal strength of the i-th device; and Respectively represent the maximum and minimum values ​​of signal strength; is the total number of devices; is the moving speed of the i-th device; is the device temperature of the i-th device; is the maximum value of the device temperature; Z is the channel occupancy rate; Indicates the battery power usage of the i-th device; is the maximum value of battery power consumption; is the battery life of the i-th device; Calculate the network performance index formula by: The network performance index W is obtained, where k is the bandwidth utilization; d is the packet loss rate; L is the network throughput; c is the delay; x is the transmission efficiency; By calculating the environmental impact index formula: The environmental impact index H is obtained, where is the ambient temperature; is the ambient humidity; β is the ambient noise value; I is the air quality index, and Q is the light intensity.

[0023] Specifically, the channel performance index, network performance index, and environmental impact index are calculated by substituting the channel performance index formula, network performance index formula, and environmental impact index formula into the real-time collected and preprocessed channel monitoring data and environmental impact data. The maximum and minimum signal strength, maximum device temperature, and maximum battery power usage are based on the maximum and minimum values ​​of multiple signal strengths, device temperatures, and battery power usage collected from q types of portable Wi-Fi devices, respectively. Battery life is determined based on actual manufacturer data or experimental testing of the i-th portable Wi-Fi device.

[0024] In this embodiment, step S3 comprehensively evaluates the channel quality in real time based on the channel performance index, network performance index, and environmental impact index to generate a comprehensive score for the portable Wi-Fi channel quality, including the following steps: Formula for calculating comprehensive channel quality score: Get the comprehensive channel quality score , where Y, W and H represent the channel performance index, network performance index and environmental impact index respectively; 、 、 Represent the weight coefficients of channel performance index, network performance index and environmental impact index respectively; is the channel occupancy; is the adjustment coefficient of channel occupancy, which is 0.6, and t is the time.

[0025] Specifically, the calculated channel performance index, network performance index, and environmental impact index are substituted into the channel quality comprehensive scoring formula to obtain a comprehensive channel quality score. In this implementation, the weighting coefficients for the channel performance index, network performance index, and environmental impact index are 0.5, 0.3, and 0.2, respectively. In actual applications or based on factors such as historical data, the weighting coefficients can be dynamically adjusted. Time t represents the cumulative time from the moment the portable Wi-Fi device is turned on to the current moment.

[0026] In this embodiment, step S3 predicts the user's data traffic demand by combining time series analysis with a deep learning model, and calculates the traffic demand peak, including the following steps: Collect historical time series data sets, including: historical traffic consumption time series, device movement GPS coordinate time series, environmental noise intensity, temperature time series data, humidity time series data, number of online devices, and channel occupancy rate; preprocess the collected historical time series data, including data cleaning and standardization, and aligning by timestamp to form a unified time step sequence; Combining time series analysis with deep learning models, we constructed a spatiotemporal attention network model to predict user data traffic needs. In the temporal dimension, we used a long short-term memory (LSTM) network to process traffic sequences and extract long-term dependency features. In the spatial dimension, we converted the device's GPS coordinate time series into a trajectory graph by connecting consecutive GPS points as edges, with edge weights calculated as the inverse of the Euclidean distance. We then used a graph neural network to model device movement trajectories and capture spatial correlations. The preprocessed time series data set is input into the spatiotemporal attention network model for training. Based on the trained spatiotemporal attention network model, the time series data collected and preprocessed through real-time monitoring is input into the trained spatiotemporal attention network model, and the output prediction result is the data traffic demand prediction value and timestamp. The prediction results are smoothed using the moving average method or the exponential smoothing method. Based on the smoothed prediction results, the peak point of traffic demand, that is, the time point when the traffic demand reaches its maximum value and the corresponding traffic value, are identified, and the peak traffic demand value is finally calculated.

[0027] Specifically, historical time series datasets are collected, including the following types of data: historical data usage time series (recording user data usage at past points in time); device movement GPS coordinate time series (recording device GPS coordinates at past points in time); ambient noise intensity (recording ambient noise intensity); temperature time series (recording temperature); humidity time series (recording humidity); online device count (recording the number of online devices); and channel occupancy (recording channel occupancy). Preprocessing of the collected historical time series data involves data cleaning to remove outliers, filling in missing values, and normalizing the data to a consistent scale. All data is aligned by timestamp to form a unified time step sequence. Combining time series analysis with deep learning models, a spatiotemporal attention network model is constructed: a long short-term memory network is used to process the traffic series and extract long-term dependency features. Spatial dimension processing involves converting the device movement GPS coordinate time series into a trajectory graph by connecting consecutive GPS points as edges, with edge weights calculated as the inverse of the Euclidean distance. Device movement trajectories are then modeled using a graph neural network to capture spatial correlations. The preprocessed time series data set is fed into the spatiotemporal attention network model for training. The model uses the preprocessed time series data as input, and the model outputs the predicted data traffic demand value and the corresponding timestamp. The model parameters are adjusted using the backpropagation algorithm to optimize prediction accuracy. The trained spatiotemporal attention network model is used for real-time prediction. Time series data is collected and preprocessed in real time and fed into the trained model. The model outputs the predicted data traffic demand value and timestamp. The prediction results are smoothed and traffic demand peaks are identified. The moving average or exponential smoothing method is used to smooth the prediction results to reduce fluctuations. Based on the smoothed prediction results, the time point at which traffic demand reaches its maximum and the corresponding traffic value are identified, and the peak traffic demand value is calculated.

[0028] In this embodiment, the reinforcement learning algorithm is used in step S4 to dynamically select the optimal channel based on the comprehensive channel quality score and the peak traffic demand, including the following steps: Based on the traffic demand peak predicted and calculated by the spatiotemporal attention network model, channels with bandwidth capacity lower than the traffic demand peak are excluded from the portable Wi-Fi channels. According to the peak traffic demand, the weight of the comprehensive channel quality score is dynamically adjusted; during the peak traffic demand period, 0.6, 0.3, is 0.1; during non-peak traffic demand periods, 0.4, 0.4, is 0.2; Based on the assigned weights, the comprehensive channel quality scores for different time periods are calculated and ranked in descending order as candidate channels. Among the candidate channels, the channel with the highest comprehensive channel quality score is selected as the optimal channel. The Q-learning algorithm, a reinforcement learning algorithm, is used to switch to the optimal channel. Each portable Wi-Fi device acts as an intelligent agent. In the Q-learning algorithm, the action-value function Q(s, a) is learned for each state and action pair. The agent then chooses action a to switch to the optimal channel based on the current channel state s. The agent performs the selected action and switches the portable WiFi device to the optimal channel; after switching the portable WiFi device to the selected optimal channel, communication begins; During the switching process, communication on the current channel is maintained and unsent data packets are cached. After the switch, the optimal channel is monitored in real time, and the immediate reward and long-term reward are calculated based on the reward function of the Q-learning algorithm. Based on the immediate reward and long-term reward, the corresponding Q value in the Q-value table is updated. By evaluating the convergence of the Q-value table, training is stopped when convergence is achieved, otherwise iterative learning continues.

[0029] Specifically, based on the traffic demand peaks predicted and calculated by the spatiotemporal attention network model, portable Wi-Fi channels with bandwidth capacities lower than the peak traffic demand are excluded. Based on the predicted peak traffic demand, the weights of the comprehensive channel quality score are dynamically adjusted. During peak traffic demand periods, the weights are set to 0.6 for the channel performance index, 0.3 for the network performance index, and 0.1 for the environmental impact index. During non-peak traffic demand periods, the weights are set to 0.4 for the channel performance index, 0.4 for the network performance index, and 0.2 for the environmental impact index. Based on the assigned weights, a comprehensive score is calculated for each channel. Channels are sorted in descending order by comprehensive score to form a candidate channel list. The channel with the highest comprehensive score is selected as the optimal channel from the candidate channels. A Q-learning algorithm is used to switch to the optimal channel. Each portable Wi-Fi device is configured as an agent. The state represents the current channel state, and the action represents whether to switch to the optimal channel. The immediate reward in the reward function is the comprehensive channel quality score after the switch, and the long-term reward is set to the average traffic transmission rate or the average comprehensive channel quality score over a period of time or the past 10 channel switches. The Q-value table is updated based on the reward function. The agent selects an action based on its current state, evaluates the reward after executing the action, and updates the Q-value table until the Q-value table converges. The agent executes the selected action, switching the portable WiFi device to the optimal channel. During the switch, communication on the current channel is maintained and unsent packets are buffered. After the switch, the quality of the optimal channel is monitored in real time, and the immediate and long-term rewards are calculated based on the reward function. Based on the immediate and long-term rewards, the corresponding Q-values ​​in the Q-value table are updated. The Q-value table is evaluated for convergence, and training is terminated when convergence is achieved; otherwise, iterative learning continues.

[0030] In this embodiment, step S4 uses reinforcement learning technology combined with a deep learning model based on the selected optimal channel to identify whether it is congested and dynamically adjust the channel, including the following steps: Based on the selected optimal channel, channel scanning technology is used to collect the optimal channel status information in real time, including: channel utilization, bandwidth utilization, channel capacity, switching delay, and data loss; the collected data is pre-processed, including data cleaning and normalization; Combining reinforcement learning technology with deep learning models, a congestion identification model based on deep reinforcement learning is constructed. A historical optimal channel state information set, labeled with whether channel congestion occurs, is input into the congestion identification model based on deep reinforcement learning for training. The optimal channel state information set, collected and preprocessed in real time, is input into the trained congestion identification model based on deep reinforcement learning to output the congestion identification results. If channel congestion is not detected, communication continues on the current channel. If channel congestion is detected, a binary exponential backoff algorithm is used. Based on the binary exponential backoff algorithm, an adaptive mechanism is introduced to dynamically adjust the backoff window size based on the real-time channel status information and congestion level. The congestion level is calculated by taking the weighted average of channel utilization, bandwidth utilization, channel capacity, switching delay, and data loss. Backoff window dynamic adjustment formula: in, is the size of the backoff window when the kth channel conflict or congestion occurs; is the level of congestion; When the congestion level is greater than the congestion level threshold, the dynamic adjustment mechanism of the backoff window size is triggered, and the backoff window size is adjusted to ; Otherwise, adjust the backoff window size to ;in, The congestion threshold is 0.65, is the maximum congestion level.

[0031] Specifically, channel scanning technology is used to collect optimal channel state information in real time, including channel utilization, bandwidth utilization, channel capacity, handover delay, and data loss. The collected data is cleaned and normalized to ensure accuracy and consistency. A congestion identification model based on deep reinforcement learning is constructed by combining reinforcement learning techniques with deep learning models. A set of historical optimal channel state information, labeled with whether or not channel congestion has occurred, is input into the model for training, enabling it to learn the characteristics and patterns of channel congestion. The optimal channel state information, collected and preprocessed in real time, is then fed into the trained congestion identification model, which outputs a congestion identification result, determining whether the current channel is congested. If channel congestion is not detected, communication continues on the current channel. If channel congestion is detected, channel adjustments are made using an improved binary exponential backoff algorithm. Building on the traditional binary exponential backoff algorithm, an adaptive mechanism is introduced to dynamically adjust the backoff window size based on real-time channel status information and the degree of congestion. The congestion level is calculated by taking the weighted average of channel utilization, bandwidth utilization, channel capacity, handover delay, and data loss. The weight coefficients for channel utilization, bandwidth utilization, channel capacity, handover delay, and data loss are 0.3, 0.25, 0.2, 0.1, and 0.15, respectively, and are dynamically adjusted based on actual conditions and historical data. A dynamic backoff window adjustment mechanism is triggered based on the congestion threshold. The backoff window size is adjusted based on the backoff window adjustment formula and the congestion level. The congestion threshold is 0.65 and is dynamically adjusted based on historical data and actual conditions. Based on the adjusted backoff window, a new channel is selected for handoff. During the handoff process, communication on the current channel is maintained, and unsent data packets are buffered. After switching to the new channel, communication is resumed, and channel status monitoring continues. Based on real-time monitoring of channel status and handoff results, the congestion identification model and backoff algorithm are continuously updated and optimized to improve model accuracy and algorithm efficiency.

[0032] In this embodiment, the step S5 includes the following steps: monitoring the adjusted optimal channel in real time, calculating the effect score, and evaluating the effect level of the optimal channel. Monitor the channel monitoring data and environmental impact data of the adjusted optimal channel in real time, and substitute them into the channel quality comprehensive scoring formula to calculate the channel quality comprehensive score; set the dynamic evaluation cycle to trigger once every 5 seconds to evaluate the effect; calculate the effect score formula: Get performance rating ;in, Indicates the comprehensive score of channel quality before channel adjustment; Indicates the comprehensive score of channel quality after channel adjustment; is the degree of congestion; is the congestion level threshold; According to the effect score, it is divided into three levels: excellent, good and poor; When the grade is excellent; when When When the evaluation level is poor, return to step S4 and re-adjust the channel selection.

[0033] Specifically, the channel monitoring data and environmental impact data of the adjusted optimal channel are monitored in real time, and substituted into the channel quality comprehensive scoring formula to calculate the channel quality comprehensive score; the dynamic evaluation cycle is set to trigger once every 5 seconds, and the effect evaluation is performed regularly. In each evaluation cycle, channel status data and environmental impact data are collected to calculate the channel quality comprehensive score. The channel quality comprehensive score before and after the channel adjustment, the congestion level and the congestion level threshold are substituted into the effect scoring formula to calculate the effect score. The effect score threshold is divided according to the actual situation and the mean and variance of historical data combined with dynamic adjustment. When the effect level is excellent or good, continue to use the current channel and keep monitoring; when the effect level is poor, return to step S4 and re-select and adjust the channel.

[0034] See also Figure 2 As shown, the present invention is an adaptive adjustment system based on portable WiFi channels, including the following modules: Monitoring and collection module: real-time monitoring and collection of channel monitoring data and environmental impact data of portable Wi-Fi channels; Data preprocessing module: performs data cleaning and standardization preprocessing on the collected channel monitoring data and environmental impact data; Data Analysis Module: Calculates the channel performance index, network performance index, and environmental impact index based on pre-processed channel monitoring data and environmental impact data. It comprehensively evaluates channel quality in real time based on the channel performance index, network performance index, and environmental impact index to generate a comprehensive score for the portable Wi-Fi channel quality. It also predicts user data traffic demands through time series analysis combined with deep learning models and calculates peak traffic demand. Selection and Adjustment Module: This module uses a reinforcement learning algorithm to dynamically select the optimal channel based on channel quality scores and predicted peak traffic demand. Based on the selected optimal channel, it combines reinforcement learning technology with a deep learning model to identify congestion and dynamically adjust the channel. Monitoring and evaluation module: monitors the adjusted optimal channel in real time, calculates the effect score and evaluates the effect level of the optimal channel.

[0035] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for adaptive adjustment of portable Wi-Fi channels, characterized in that: The following steps are involved: S1: Real-time monitoring and collection of channel monitoring data and environmental impact data of the portable Wi-Fi channel; S2: Preprocess the collected channel monitoring data and environmental impact data by cleaning and standardizing them respectively; S3: Analyze and calculate the channel performance index, network performance index, and environmental impact index based on preprocessed channel monitoring data and environmental impact data. Perform a real-time assessment of channel quality based on the channel performance index, network performance index, and environmental impact index to generate a comprehensive score for the portable Wi-Fi channel quality. Use time series analysis combined with deep learning models to predict user data traffic demands and calculate peak traffic demand. S4: Utilizes a reinforcement learning algorithm to dynamically select the optimal channel based on channel quality scores and predicted peak traffic demand. Based on the selected optimal channel, reinforcement learning technology combined with a deep learning model is used to identify congestion and dynamically adjust the channel. S5: Monitor the adjusted optimal channel in real time, calculate the effect score and evaluate the effect level of the optimal channel.

2. The method for adaptively adjusting the portable Wi-Fi channel according to claim 1, characterized in that: The channel monitoring data in step S1 includes: the number of devices on the channel, device movement speed, signal strength, channel occupancy, device battery level, device temperature and network traffic data; the environmental impact data includes: ambient temperature, ambient humidity, ambient noise value, air quality index and light intensity; among them, the network traffic data includes: bandwidth utilization, packet loss rate, network throughput and delay, and transmission efficiency.

3. The method for adaptively adjusting the portable Wi-Fi channel according to claim 2, characterized in that: The step S3 includes the following steps: Calculate the channel performance index formula: Get the channel performance index Y, where is the signal strength of the i-th device; and Respectively represent the maximum and minimum values ​​of signal strength; is the total number of devices; is the moving speed of the i-th device; is the device temperature of the i-th device; is the maximum value of the device temperature; Z is the channel occupancy rate; Indicates the battery power usage of the i-th device; is the maximum value of battery power consumption; is the battery life of the i-th device; Calculate the network performance index formula by: The network performance index W is obtained, where k is the bandwidth utilization; d is the packet loss rate; L is the network throughput; c is the delay; x is the transmission efficiency; By calculating the environmental impact index formula: The environmental impact index H is obtained, where is the ambient temperature; is the ambient humidity; β is the ambient noise value; I is the air quality index, and Q is the light intensity.

4. The method for adaptively adjusting the portable Wi-Fi channel according to claim 3, characterized in that: In step S3, the channel quality is comprehensively evaluated in real time based on the channel performance index, the network performance index, and the environmental impact index to generate a comprehensive score of the portable Wi-Fi channel quality, including the following steps: Formula for calculating comprehensive channel quality score: Get the comprehensive channel quality score , where Y, W and H represent the channel performance index, network performance index and environmental impact index respectively; 、 、 Represent the weight coefficients of channel performance index, network performance index and environmental impact index respectively; is the channel occupancy; is the adjustment coefficient of channel occupancy, which is 0.6, and t is the time.

5. The method for adaptively adjusting the portable Wi-Fi channel according to claim 1, characterized in that: Step S3 predicts the user's data traffic demand by combining time series analysis with a deep learning model, and calculates the traffic demand peak, including the following steps: Collect historical time series data sets, including: historical traffic consumption time series, device movement GPS coordinate time series, environmental noise intensity, temperature time series data, humidity time series data, number of online devices, and channel occupancy rate; preprocess the collected historical time series data, including data cleaning and standardization, and aligning by timestamp to form a unified time step sequence; Combining time series analysis with deep learning models, we constructed a spatiotemporal attention network model to predict user data traffic needs. In the temporal dimension, we used a long short-term memory (LSTM) network to process traffic sequences and extract long-term dependency features. In the spatial dimension, we converted the device's GPS coordinate time series into a trajectory graph by connecting consecutive GPS points as edges, with edge weights calculated as the inverse of the Euclidean distance. We then used a graph neural network to model device movement trajectories and capture spatial correlations. The preprocessed time series data set is input into the spatiotemporal attention network model for training. Based on the trained spatiotemporal attention network model, the time series data collected and preprocessed through real-time monitoring is input into the trained spatiotemporal attention network model, and the output prediction result is the data traffic demand prediction value and timestamp. The prediction results are smoothed using the moving average method or the exponential smoothing method. Based on the smoothed prediction results, the peak point of traffic demand, that is, the time point when the traffic demand reaches its maximum value and the corresponding traffic value, are identified, and the peak traffic demand value is finally calculated.

6. The method for adaptively adjusting the portable Wi-Fi channel according to claim 1, characterized in that: Step S4 uses a reinforcement learning algorithm to dynamically select the optimal channel based on the comprehensive channel quality score and the peak traffic demand, including the following steps: Based on the traffic demand peak predicted and calculated by the spatiotemporal attention network model, channels with bandwidth capacity lower than the traffic demand peak are excluded from the portable Wi-Fi channels. According to the peak traffic demand, the weight of the comprehensive channel quality score is dynamically adjusted; during the peak traffic demand period, 0.6, 0.3, is 0.1; during non-peak traffic demand periods, 0.4, 0.4, is 0.2; Based on the assigned weights, the comprehensive channel quality scores for different time periods are calculated and ranked in descending order as candidate channels. Among the candidate channels, the channel with the highest comprehensive channel quality score is selected as the optimal channel. The Q-learning algorithm, a reinforcement learning algorithm, is used to switch to the optimal channel. Each portable Wi-Fi device acts as an intelligent agent. In the Q-learning algorithm, the action-value function Q(s, a) is learned for each state and action pair. The agent then chooses action a to switch to the optimal channel based on the current channel state s. The agent performs the selected action and switches the portable WiFi device to the optimal channel; after switching the portable WiFi device to the selected optimal channel, communication begins; During the switching process, communication on the current channel is maintained and unsent data packets are cached. After the switch, the optimal channel is monitored in real time, and the immediate reward and long-term reward are calculated based on the reward function of the Q-learning algorithm. Based on the immediate reward and long-term reward, the corresponding Q value in the Q-value table is updated. By evaluating the convergence of the Q-value table, training is stopped when convergence is achieved, otherwise iterative learning continues.

7. The method for adaptively adjusting the portable Wi-Fi channel according to claim 6, characterized in that: In step S4, based on the selected optimal channel, reinforcement learning technology is combined with a deep learning model to identify whether it is congested and dynamically adjust the channel, including the following steps: Based on the selected optimal channel, channel scanning technology is used to collect the optimal channel status information in real time, including: channel utilization, bandwidth utilization, channel capacity, switching delay, and data loss; the collected data is pre-processed, including data cleaning and normalization; Combining reinforcement learning technology with deep learning models, a congestion identification model based on deep reinforcement learning is constructed. A historical optimal channel state information set, labeled with whether channel congestion occurs, is input into the congestion identification model based on deep reinforcement learning for training. The optimal channel state information set, collected and preprocessed in real time, is input into the trained congestion identification model based on deep reinforcement learning to output the congestion identification results. If channel congestion is not detected, communication continues on the current channel. If channel congestion is detected, a binary exponential backoff algorithm is used. Based on the binary exponential backoff algorithm, an adaptive mechanism is introduced to dynamically adjust the backoff window size based on the real-time channel status information and congestion level. The congestion level is calculated by taking the weighted average of channel utilization, bandwidth utilization, channel capacity, switching delay, and data loss. Backoff window dynamic adjustment formula: in, is the size of the backoff window when the kth channel conflict or congestion occurs; is the level of congestion; When the congestion level is greater than the congestion level threshold, the dynamic adjustment mechanism of the backoff window size is triggered, and the backoff window size is adjusted to ; Otherwise, adjust the backoff window size to ;in, The congestion threshold is 0.65, is the maximum congestion level.

8. The method for adaptively adjusting the portable Wi-Fi channel according to claim 1, wherein: The step S5 includes the following steps: monitoring the adjusted optimal channel in real time, calculating the effect score, and evaluating the effect level of the optimal channel. Monitor the channel monitoring data and environmental impact data of the adjusted optimal channel in real time, and substitute them into the channel quality comprehensive scoring formula to calculate the channel quality comprehensive score; set the dynamic evaluation cycle to trigger once every 5 seconds to evaluate the effect; calculate the effect score formula: Get performance rating ;in, Indicates the comprehensive score of channel quality before channel adjustment; Indicates the comprehensive score of channel quality after channel adjustment; is the degree of congestion; is the congestion level threshold; According to the effect score, it is divided into three levels: excellent, good and poor; When the grade is excellent; when When When the evaluation level is poor, return to step S4 and re-adjust the channel selection.

9. A portable Wi-Fi channel-based adaptive adjustment system, using a portable Wi-Fi channel-based adaptive adjustment method according to any one of claims 1 to 8, characterized in that: Includes the following modules: Monitoring and collection module: real-time monitoring and collection of channel monitoring data and environmental impact data of portable Wi-Fi channels; Data preprocessing module: performs data cleaning and standardization preprocessing on the collected channel monitoring data and environmental impact data; Data Analysis Module: Calculates the channel performance index, network performance index, and environmental impact index based on pre-processed channel monitoring data and environmental impact data. It comprehensively evaluates channel quality in real time based on the channel performance index, network performance index, and environmental impact index to generate a comprehensive score for the portable Wi-Fi channel quality. It also predicts user data traffic demands through time series analysis combined with deep learning models and calculates peak traffic demand. Selection and Adjustment Module: This module uses a reinforcement learning algorithm to dynamically select the optimal channel based on channel quality scores and predicted peak traffic demand. Based on the selected optimal channel, it combines reinforcement learning technology with a deep learning model to identify congestion and dynamically adjust the channel. Monitoring and evaluation module: monitors the adjusted optimal channel in real time, calculates the effect score and evaluates the effect level of the optimal channel.

Images