Channel selection method based on time dimension statistics and stability sorting

Abstract

The invention relates to the technical field of wireless communication, and discloses a channel selection method based on time dimension statistics and stability sorting, which comprises the following steps: screening out channels conforming to local communication specifications from a preset region-channel mapping table according to a country or region where equipment is located currently, and forming an alternative channel list; monitoring the communication state of the current channel in real time, and if any one of the following conditions is satisfied, triggering channel switching: the channel interference intensity is lower than a first threshold for a first preset time; the communication rate is lower than a second threshold value for a second preset time; the number of times that the data rate of the channel is accumulatively lower than the third threshold value reaches the preset number. According to the method, illegal or unmatched channels are automatically screened out through matching of the preset regional specifications and the real-time geographic positions, it is ensured that equipment can strictly follow local radio management laws and regulations in different countries and regions in the world, meanwhile, the compliance alternative list is dynamically generated, and legal, safe and efficient utilization of channel resources is achieved.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a channel selection method based on time-dimensional statistics and stability ranking. Background Technology

[0002] In existing technologies, wireless channel selection often adopts fixed channel allocation or simple random selection methods, without combining national and regional channel specifications and historical usage data for intelligent optimization.

[0003] The existing technology has three major flaws: First, it is not adapted to the channel usage rules of different countries / regions, which can easily lead to illegal channel use or waste of effective channels; second, it lacks a full-dimensional record of the channel usage process, making it impossible to select stable channels based on historical performance, resulting in frequent switching and poor communication stability; third, it has not established a scientific channel conflict avoidance mechanism, and there is no control over channels repeatedly selected in a short period of time, which can easily cause communication interruptions.

[0004] Currently, there is no integrated solution that simultaneously achieves "regional compliance adaptation, historical data recording, intelligent conflict avoidance, and stable channel selection." Existing technologies are insufficient to meet the multiple requirements of wireless display devices for channel selection accuracy, stability, and compliance. Summary of the Invention

[0005] This invention provides a channel selection method based on time-dimensional statistics and stability ranking to solve existing technical problems, thereby addressing the difficulty of meeting the multiple requirements of wireless display devices for channel selection accuracy, stability, and compliance.

[0006] To address the aforementioned technical problems, according to one aspect of the present invention, more specifically, a channel selection method based on time-dimensional statistics and stability ranking, comprising the following steps: S1. Based on the country or region where the device is currently located, select channels that conform to local communication standards from the preset region-channel mapping table to form a candidate channel list; S2. Monitor the current channel's communication status in real time. If any of the following conditions are met, trigger a channel switch: S201, The channel interference intensity remains below a first threshold for a first preset time; S202, The communication rate remains below a second preset time for a second threshold; S203, The number of times the channel's data rate has fallen below the third threshold has reached a preset number; S204, Channel occupancy conflict triggers alarm; S3. During channel switching, execute an intelligent channel selection strategy, including: If there are channels in the candidate channel list that are not recorded in the historical channel list, the unrecorded channels will be selected for switching first. If all channels in the candidate channel list are already recorded in the historical channel list, then perform the following sub-steps: S301. Filter conflicting channels. If a channel is selected twice consecutively within a preset time interval, the channel is marked as an ignored channel and will not participate in the selection during the ignored period. If all candidate channels are ignored channels, the channel status list is cleared and step S3 is repeated. S302. Calculate the average dwell time of each channel based on the historical dwell time data of each channel in the historical channel list; S303. Filter out the channel with the longest average dwell time and obtain a set of candidate channels whose average dwell time does not differ from the channel by a preset deviation range; if the channel exceeds the preset deviation range, select the channel with the longest average dwell time as the target channel for switching. S304. Calculate the Sigma value of the dwell time of each channel in the candidate channel set, and select the channel with the smallest Sigma value as the target channel for switching. S4. After switching to the target channel, update the historical channel list and record the basic information, entry time, exit time and dwell time of the channel.

[0007] Furthermore, the region-channel mapping table is two-dimensional structured data, including: region identifier, channel frequency band, channel number, maximum transmit power, and supported WiFi protocol version.

[0008] Furthermore, the historical channel linked list is used to store channel usage records, and each node includes: channel value, channel entry time, channel exit time, channel dwell time, and average dwell time; The linked list data is stored for a preset number of days, and is automatically cleaned up after the timeout period.

[0009] Furthermore, the average dwell time is calculated as follows: ; In the above formula, Indicates the average length of stay; Indicates the first Duration of stay; This indicates the total number of times the channel has been selected.

[0010] Furthermore, the formula for calculating the Sigma value is as follows: ; In the above formula, Indicates the first Duration of stay; This indicates the average dwell time of the channel; This indicates the total number of times the channel has been selected.

[0011] Furthermore, in step S3, the specific steps for prioritizing the switching of unrecorded channels are as follows: A. Obtain the channel environment quality prediction index and potential collision probability of channels in the candidate channel list that are not recorded in the historical channel chain; B. Input the obtained channel environment quality prediction index and channel potential conflict probability into the exploration priority comprehensive judgment model; C. Based on the magnitude of the priority coefficient output by the comprehensive judgment model according to the exploration priority, the unrecorded channel is selected for switching first.

[0012] Furthermore, the specific steps for constructing a comprehensive judgment model for exploration priorities are as follows: Step 1: Using the controlled variable method, obtain the channel environment quality prediction index, channel potential collision probability, and comprehensive stability index of the same type of channel in subsequent use, which are recorded in the historical channel chain. Step 2: Construct independent sub-models based on the relationship between the comprehensive stability index and the channel environment quality prediction index and the channel potential collision probability, respectively. Step 3: Determine the relationship between the comprehensive stability index and the channel environment quality prediction index and the channel potential collision probability, respectively. Step 4: Construct a comprehensive judgment model for exploration priorities based on independent sub-models and relationships.

[0013] Furthermore, the exploration priority comprehensive judgment model is a comprehensive evaluation model based on historical channel usage data, used to prioritize channels that are not recorded in the historical channel list. This exploration priority comprehensive judgment model integrates the channel environment quality prediction index and the channel potential conflict probability, and constructs a correlation based on the comprehensive stability performance of historical channels of the same type. It then outputs a priority coefficient that reflects the expected stability of the channel, so as to guide the equipment to prioritize the channel with the best comprehensive evaluation result among multiple unused channels for switching.

[0014] Furthermore, the preset time interval is 120 seconds, the ignore duration is 5 minutes; the first preset time is 3 seconds, the first threshold is 20dB; the second preset time is 10 seconds, the second threshold is 8Mbps; the third threshold is 58Mbps, the preset number of times is 3; and the preset deviation range is the average dwell time ±30 seconds.

[0015] This invention provides a channel selection method based on time-dimensional statistics and stability ranking. Compared with existing technologies, this method achieves the following advantages: 1. This invention automatically filters out illegal or incompatible channels by matching pre-set regional specifications with real-time geographical locations, ensuring that the equipment can strictly comply with local radio management regulations in different countries and regions around the world. At the same time, it dynamically generates a list of compliant alternatives, realizing the legal, safe and efficient use of channel resources, and avoiding communication interruptions or compliance risks caused by mismatched regional rules from the source.

[0016] 2. This invention selects a set of candidate channels with reliable long-term performance by calculating the average dwell time, and further uses the dispersion index of dwell time to accurately locate the channel with the least fluctuation. Thus, in multiple handovers, it systematically tends to select the channel with a continuously stable communication state, which significantly improves the overall reliability of wireless connection and user experience.

[0017] 3. By introducing a conflict channel marking and ignoring mechanism, this invention intelligently avoids channels that are repeatedly selected in a short period of time. Combined with periodic data cleaning and list reset strategies, it effectively breaks the vicious cycle of repeated oscillations between limited channels, reduces the number of invalid handovers caused by channel contention or momentary interference, and improves the efficiency of channel selection decisions and the overall utilization of network resources.

[0018] 4. This invention adopts a priority evaluation model that integrates multi-dimensional parameters. It correlates real-time sensing data such as channel environment quality, physical characteristics and potential conflict probability with the stability performance of historical channels of the same type, thereby enabling scientific prediction and ranking of unknown channel performance. This guides the equipment to make risk-controlled and better-expected choices when exploring new channels, and enhances the system's adaptive and learning capabilities in dynamic environments. Attached Figure Description

[0019] Figure 1 This is the upper part of the flowchart of Embodiment 1 of the present invention; Figure 2 This is the lower half of the flowchart of Embodiment 1 of the present invention. Figure 3 This is a flowchart of Embodiment 2 of the present invention; Figure 4 This is a graph showing the relationship between the comprehensive stability coefficient and the channel environment quality prediction index in this invention. Figure 5 This is a graph showing the relationship between the overall stability coefficient and the potential channel collision probability in this invention. Figure 6 This is a graph showing the relationship between the channel environment quality prediction index and the channel potential conflict probability in this invention. Detailed Implementation

[0020] To make the technical solution of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] Example 1 like Figure 1 , Figure 2 As shown, this invention provides a channel selection method based on time-dimensional statistics and stability ranking, applicable to wireless display devices in embedded multimedia products, specifically relating to the field of channel selection and switching technology in wireless communication. This embodiment uses the chip models am8270, am8360, am8268 and their corresponding SDKs (am1229_linux, 826x_new_linux, am8270) as examples for illustration.

[0022] I. Core Algorithm Principles and Execution Steps Step S1: Determine the candidate channel list based on country / region. The device has a built-in library of wireless channel specifications for major countries / regions worldwide, namely a region-channel mapping table. This table is a two-dimensional structured data table containing the following core fields: Regional designations (in accordance with ISO 3166-1alpha-2 standards, such as CN, US, EU); Channel frequency band (2.4GHz / 5GHz); Channel number (e.g., 2.4GHz: 1-13, 5GHz: 36-165); Maximum transmit power (unit: dBm); Supported WiFi protocol versions (such as 802.11n / ac / ax).

[0023] When the device starts up, it reads the preset region parameters (configurable by the user) to identify the current country and region, and automatically filters out channels that conform to the specifications of that region from the mapping table to form a list of candidate channels.

[0024] Step S2: Channel switching condition judgment The system monitors the current channel's communication status in real time and triggers a channel switch if any of the following conditions are met: (1) The channel interference intensity (signal-to-noise ratio) remains below 20dB for 3 seconds; (2) The communication rate remains below 8 Mbps for 10 seconds; (3) The number of times the channel data rate is lower than 58Mbps reaches 3 times; (4) Alarm triggered by channel occupancy conflict.

[0025] If none of the above conditions are met, the current channel operation will continue.

[0026] Step S3, Channel Data Recording Mechanism After each channel switch, the channel usage record is automatically stored in the historical channel linked list. Each node in the linked list contains the following information: Channel value; Channel entry time (accurate to milliseconds); Channel departure time (accurate to milliseconds); Channel dwell time (departure time - entry time); Average length of stay (calculated based on historical length of stay).

[0027] The linked list data storage period is set to 7 days, and it will be automatically cleaned up after the timeout to ensure storage efficiency.

[0028] Step S4, Intelligent Channel Selection Strategy When a channel switch is triggered, the following logic is executed: If there are channels in the candidate channel list that are not recorded in the historical channel chain, then an unused channel will be selected for switching, ensuring that all compliant channels have the opportunity to be evaluated.

[0029] If all candidate channels are already recorded in the history list, then perform the following sub-steps: (1) Conflicting channel screening: If a channel is selected twice consecutively within 120 seconds, it is marked as an "ignore channel" and will not participate in the selection for the next 5 minutes. If all candidate channels are ignored, the channel state list is cleared and this step is repeated.

[0030] (2) Calculate the average dwell time: Based on the historical dwell time of each channel in the historical list, calculate its average dwell time, as shown in the following formula: ; In the above formula, Indicates the average length of stay; Indicates the first Duration of stay; This indicates the total number of times the channel has been selected.

[0031] (3) Screening the candidate channel set: Find the channel with the longest average dwell time, and screen other channels whose average dwell time differs from the channel by no more than ±30 seconds, and together they form the candidate channel set.

[0032] (4) Calculate the Sigma value and select the most stable channel: Calculate the Sigma value (sample standard deviation) of the dwell time of each channel in the candidate set, as follows: ; In the above formula, Indicates the first Duration of stay; This indicates the average dwell time of the channel; This indicates the total number of times the channel has been selected; Select the channel with the smallest Sigma value as the target channel for switching to ensure the most stable channel usage.

[0033] Step S5: Update History After switching to the target channel, update the historical channel list, record the entry time, exit time, and dwell time of the channel, and recalculate the average dwell time.

[0034] II. Key Technology Description (1) Implementation of regional channel adaptation The mapping table is constructed using multi-source rules, including standards from national communications management agencies, WiFi Alliance guidelines, and chip manufacturer compatibility data.

[0035] Supports dynamic updates: The mapping table can be updated via firmware upgrades or SDK patches to adapt to changes in channel rules in various countries.

[0036] Automatic configuration process: region identification → rule matching → channel filtering → parameter writing, requiring no manual intervention throughout.

[0037] (2) Linked list storage mechanism Historical channel data is stored using a linked list structure, which supports fast insertion, querying, deletion and updating, thereby improving the efficiency of algorithm execution.

[0038] (3) Stability assessment of Sigma value The standard deviation of the sample is used to measure the dispersion of channel dwell time. The smaller the Sigma value, the more stable the channel usage and the more reliable the communication quality.

[0039] The core innovation of this embodiment lies in establishing a channel stability quantitative assessment and selection mechanism based on historical dwell time statistics. The principle is to continuously record the actual usage time of each channel and calculate the average dwell time to screen for long-term reliable candidate channels. Furthermore, the dispersion of dwell time (Sigma value) is introduced as a precise stability index, thereby systematically selecting the channel with the least fluctuation and best sustainability during multiple handovers.

[0040] The advantage of this method is that it upgrades channel selection from the traditional mode that relies on instantaneous states to a decision-making mode based on big data analysis in the time dimension. This effectively avoids misjudgments and frequent jumps caused by instantaneous interference, significantly improves the long-term steady state of wireless communication and user experience, and, combined with an intelligent conflict avoidance mechanism, further ensures the orderly utilization efficiency of network resources.

[0041] Example 2 like Figure 3 As shown, this invention provides a channel selection method based on time-dimensional statistics and stability ranking, applicable to wireless display devices in embedded multimedia products, specifically relating to the field of channel selection and switching technology in wireless communication. This embodiment uses the chip models am8270, am8360, am8268 and their corresponding SDKs (am1229_linux, 826x_new_linux, am8270) as examples for illustration.

[0042] I. Core Algorithm Principles and Execution Steps Step S1: Based on the country or region where the device is currently located, select channels that conform to local communication standards from the preset region-channel mapping table to form a candidate channel list; Step S2: Monitor the current channel's communication status in real time. If any of the following conditions are met, trigger channel switching: (1) The channel interference intensity remains below the first threshold for a first preset time; (2) The communication rate remains below the second preset time for a second time threshold; (3) The number of times the channel's data rate has fallen below the third threshold reaches a preset number; (4) Alarm triggered by channel occupancy conflict; Step S3: During channel switching, if there are channels in the candidate channel list that are not recorded in the historical channel chain, then the unrecorded channels will be selected for switching.

[0043] The specific steps for prioritizing the switching of unrecorded channels in this step are as follows: (1) Obtain the channel environment quality prediction index and channel potential collision probability of channels that exist in the candidate channel list but are not recorded in the historical channel chain; (2) Input the obtained channel environment quality prediction index and channel potential conflict probability into the exploration priority comprehensive judgment model; (3) Based on the magnitude of the priority coefficient output by the exploration priority comprehensive judgment model, the unrecorded channels are selected for switching.

[0044] Step S4: After switching to the target channel, update the historical channel list and record the basic information, entry time, exit time and dwell time of the channel.

[0045] II. Key Technology Description like Figures 4 to 6 As shown, the specific steps for constructing a comprehensive judgment model for exploration priorities are as follows: (1) By using the control variable method, the channel environment quality prediction index, channel potential conflict probability, and comprehensive stability index of the same type of channel in subsequent use were obtained from the historical channel chain list. The channel environment quality prediction index c is obtained during channel usage, specifically as follows: ; in, This indicates the number of other Wi-Fi devices (APs / STAs) detected on this channel; This indicates the received signal strength of the i-th device (absolute value, in dBm). This indicates the reference maximum interference intensity (e.g., -20dBm).

[0046] Its constraint is: if (If the channel is idle, then) ;like Then take .

[0047] Among them, the fewer interfering devices in the environment and the weaker the signal, the closer the channel environment quality prediction index is to 1, indicating that the channel environment quality is better.

[0048] The potential collision probability p of the channel is obtained during channel usage, specifically as follows: ; in, This represents the channel occupancy rate (values ​​range from 0 to 1), measured through idle channel evaluation. This indicates the observation time window (in seconds), with a recommended range of 30-60 seconds. This represents the time decay constant (in seconds), with 300 seconds recommended. This represents the device density on the channel (number of devices ÷ unit time). This represents the maximum equipment density reference value; This represents the time weighting coefficient, with higher values ​​during peak periods (e.g., 1.2) and lower values ​​during off-peak periods (e.g., 0.8).

[0049] Among them, the higher the channel occupancy rate and the greater the equipment density, the closer the potential channel collision probability is to 1, indicating a higher collision probability.

[0050] (2) Construct independent sub-models based on the relationship between the comprehensive stability index and the channel environment quality prediction index and the channel potential conflict probability, respectively; The overall stability index s is obtained based on feedback from the use of the channel after its use, specifically as follows: ; in, Indicates the signal-to-noise ratio stability coefficient; Indicates the effective throughput retention rate; Indicates packet loss rate stability; Indicates the ratio of continuous, interference-free duration; , , , This represents the weighting coefficients, and each takes... (Signal-to-noise ratio stability weight) (Throughput remains weighted) (Packet loss rate stability weight) (Weight of continuous working time).

[0051] However, the overall stability index obtained after the channel is used is not consistent. For example, if 1000 sample channels of the same type are obtained, the overall stability index of these 1000 sample channels will not be expressed in a consistent manner as 1%, 2%, etc.

[0052] Therefore, these 1000 sample channels are sorted according to the magnitude of the comprehensive stability index, and then expressed sequentially according to the sort order. For example, if the sort order of a certain sample channel is 500, it represents 50.0%, which is denoted as the comprehensive stability coefficient.

[0053] 1) From the above 1000 sample channels, select 100 samples with a potential collision probability of 0.5, and denote the comprehensive stability coefficient of these 100 sample channels as s1.

[0054] Therefore, the output value of the independent sub-model based on the channel environment quality prediction index is... The mathematical relationship between the channel environment quality prediction index c and the index is as follows: (Formula 1); In Formula 1 above, , Used for control A constant that approximates the overall stability coefficient s1.

[0055] The above 100 samples were analyzed by Figure 4 Able to determine , When, in Formula 1 The overall stability coefficient s1 is approximately equal to ( express Figure 4 The curve in the figure shows the point representing the comprehensive stability coefficient s1.

[0056] 2) From the 1000 sample channels, select 100 samples with a channel environment quality prediction index of 0.5 each, and denote the comprehensive stability coefficient of these 100 sample channels as s2.

[0057] Then, the output value of the independent sub-model based on the potential channel collision probability. The mathematical relationship between the potential collision probability p of the channel is as follows: (Formula 2); In formula 2 above, , Used for control A constant that approximates the comprehensive stability coefficient s2 (p takes values ​​between 0 and 1).

[0058] Furthermore, through analysis of the aforementioned 100 samples, and through... Figure 5 Able to determine , When, in Formula 2 It approximates the overall stability coefficient s2 ( express Figure 5 The curve in the figure shows the point representing the comprehensive stability coefficient s2.

[0059] (3) Determine the relationship between the independent sub-models based on the pairwise relationships between the comprehensive stability index and the channel environment quality prediction index and the channel potential conflict probability; The 1000 sample channels are sorted by size, and the overall stability coefficient of these 1000 sample channels is denoted as s3.

[0060] Therefore, to determine the relationship between Formula 1 and Formula 2 above, we need to consider the relationship between the channel environment quality prediction index c, the channel potential collision probability p, and the comprehensive stability coefficient s3. Then we have: ; (4) Construct a comprehensive judgment model for exploration priorities based on independent sub-models and relational expressions, specifically as follows: ; In the above, A constant used to control the output value S of the comprehensive judgment model for exploration priority to approximate the comprehensive stability coefficient s3. And by... Figure 6 The data in the middle can be determined , At that time, the output value S of the priority comprehensive judgment model tends to be approximately the same as the comprehensive stability coefficient s3.

[0061] Therefore, the mathematical expression for the exploration priority comprehensive judgment model obtained from the above 1000 sample channels is: ; In the above formula, S represents the priority coefficient of the channel output by the exploration priority comprehensive judgment model. The larger S is, the higher the priority of the unrecorded channel for switching.

[0062] The core innovation of this embodiment lies in proposing an intelligent exploration strategy for unused channels, based on multi-dimensional parameter fusion and historical data correlation. The principle is to construct an exploration priority comprehensive judgment model. This model analyzes the correlation between various quality parameters (such as environmental quality index and potential conflict probability) and the final comprehensive stability in historical channels using the controlled variable method, and uses this correlation to predict and rank the expected performance of unrecorded channels.

[0063] The advantage of this method is that it solves the problems of blindness and uncertainty when the equipment explores new channels. By combining real-time sensing parameters with historical experience data, it enables controllable prediction of the performance of unknown channels, allowing the system to make more scientific and stable choices while expanding available channel resources, and enhancing the adaptability and foresight of the overall solution in dynamic and complex environments.

[0064] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A channel selection method based on time dimension statistics and stability ranking, characterized in that, The method comprises the following steps: S1, screening a channel list according to the local communication standard from a preset area-channel mapping table according to the country or region where the device is currently located; S2, monitoring the communication state of the current channel in real time, and triggering channel switching if any of the following conditions is met: S201, the channel interference intensity is lower than a first threshold value for a first preset time; S202, the communication rate is lower than a second threshold value for a second preset time; S203, the number of times that the data rate of the channel is lower than a third threshold value reaches a preset number of times; S204, the channel occupancy conflict triggers an alarm; S3, when the channel is switched, an intelligent channel selection strategy is executed, comprising: if there is a channel in the channel list that is not recorded in the historical channel link list, the channel not recorded is preferentially selected for switching; if all channels in the channel list are recorded in the historical channel link list, the following sub-steps are executed: S301, screening a conflict channel, if a channel is selected twice continuously within a preset time interval, the channel is marked as an ignored channel and is not selected within an ignored time period; if all channels in the channel list are ignored channels, the channel state link list is cleared and step S3 is repeated; S302, calculating the average residence time of each channel based on the historical residence time data of each channel in the historical channel link list; S303, screening a channel with the longest average residence time and obtaining a candidate channel set whose average residence time differs from that of the channel by not more than a preset deviation range; if the channel exceeds the preset deviation range, the channel with the maximum average residence time is selected as a target channel for switching; S304, calculating the Sigma value of the residence time of each channel in the candidate channel set, and selecting the channel with the minimum Sigma value as the target channel for switching; S4, after switching to the target channel, updating the historical channel link list to record the basic information, entry time, exit time and residence time of the channel.

2. The method of claim 1, wherein: The area-channel mapping table is two-dimensional structured data, including: area identifier, channel frequency band, channel number, maximum transmission power, and supported WiFi protocol version.

3. The method of claim 1, wherein: The historical channel link list is used to store channel usage records, and each node includes: channel value, channel entry time, channel exit time, channel residence time, and average residence time; The chain table data storage period is a preset number of days, and is automatically cleaned up when it expires.

4. The method of claim 1, wherein: The calculation method of the average residence time is: ； In the above formula, denotes the average residence time; denotes the residence time of the first residence time; denotes the total number of times the channel was selected.

5. The method of claim 1, wherein: The calculation formula of the Sigma value is: ； In the above formula, denotes the residence time; denotes the average residence time of the channel; denotes the total number of times the channel was selected.

6. The method of claim 1, wherein: In step S3, the specific steps of preferentially selecting the channel not recorded for switching are: A, obtaining the channel environment quality prediction index and the channel potential conflict probability of the channel not recorded in the historical channel link list in the channel list; B, inputting the obtained channel environment quality prediction index and channel potential conflict probability into an exploration priority comprehensive judgment model; C, preferentially selecting the channel not recorded for switching according to the size of the priority coefficient output by the exploration priority comprehensive judgment model.

7. The method of claim 6, wherein: The specific steps of constructing the exploration priority comprehensive judgment model are: Step one, obtain the channel environment quality prediction index, the channel potential conflict probability of the same type channel recorded in the historical channel link table, and the corresponding comprehensive stability index of the same type channel in the subsequent use process by controlling variable method respectively; Step two, construct independent sub-models according to the relationship between the comprehensive stability index and the channel environment quality prediction index and the channel potential conflict probability respectively; Step three, determine the relationship of the independent sub-models according to the relationship between the comprehensive stability index and the channel environment quality prediction index and the channel potential conflict probability respectively; Step four, construct the exploration priority comprehensive judgment model based on the independent sub-models and the relationship.

8. The method of claim 6, wherein: The exploration priority comprehensive judgment model is a comprehensive evaluation model established based on historical channel use data, which is used to prioritize channels not recorded in the historical channel link table; The exploration priority comprehensive judgment model integrates the channel environment quality prediction index and the channel potential conflict probability, and constructs the correlation relationship according to the comprehensive stability performance of the historical same type channel, and then outputs the priority coefficient reflecting the expected stability degree of the channel, so as to guide the device to preferentially select the channel with the optimal comprehensive evaluation result among multiple unused channels for switching.

9. The method of claim 1, wherein: The preset time interval is 120 seconds, the ignore duration is 5 minutes; the first preset time is 3 seconds, and the first threshold is 20 dB; the second preset time is 10 seconds, and the second threshold is 8 Mbps; the third threshold is 58 Mbps, and the preset number of times is 3; the preset deviation range is the average residence time ± 30 seconds.

Images