Bluetooth headset connection control method, device, equipment and storage medium

Abstract

The present application relates to the technical field of data transmission control, and discloses a connection control method, device, equipment and storage medium for a Bluetooth headset. The method includes: constructing an initial Bluetooth connection topology network of a target Bluetooth headset and multiple local terminal devices; collecting environmental data and creating an initial connection decision list; performing a connection decision test between the target Bluetooth headset and each local terminal device, and collecting the data rate and transmission power of each local terminal device; performing feature extraction and coding vector conversion to generate a connection test coding vector; performing connection quality analysis through a connection quality monitoring model to obtain a connection quality index; performing network variable weight optimization and decision list optimization on the initial Bluetooth connection topology network according to the connection quality index to generate a target connection decision list. The present application realizes intelligent selection of connections between Bluetooth headsets and local terminal devices, thereby improving the data rate and transmission power of Bluetooth headsets.

Description

Technical Field

[0001] The present application relates to the technical field of data transmission control, and in particular to a connection control method, device, equipment and storage medium for a Bluetooth headset. Background Art

[0002] Bluetooth headsets are becoming more and more widely used in daily life. However, how to effectively manage the connections between Bluetooth headsets and multiple devices has become a challenge. Existing connection management mechanisms often cannot adapt well to complex and changing connection environments, which leads to instability and inefficiency when Bluetooth devices connect.

[0003] Specifically, existing Bluetooth connection control methods have limitations when dealing with multi-device connections and switching. In actual use, Bluetooth headsets may face problems such as signal interference, insufficient compatibility between devices, and inflexible connection settings, all of which will affect the user's connection experience. Previous technologies often required users to manually switch connections between devices, which was not only cumbersome but also inefficient, resulting in a poor user experience in multi-device usage scenarios. In addition, the connection stability between Bluetooth headsets and devices is also an important factor affecting user experience. However, traditional connection methods often cannot effectively solve this problem, resulting in intermittent connections and affecting user use. Summary of the invention

[0004] The present application provides a connection control method, apparatus, device and storage medium for a Bluetooth headset, which are used to realize intelligent selection of a connection between a Bluetooth headset and a local terminal device, thereby improving the data rate and transmission power of the Bluetooth headset.

[0005] In a first aspect, the present application provides a connection control method for a Bluetooth headset, the connection control method for a Bluetooth headset comprising:

[0006] Searching for multiple local terminal devices within the connection range of the target Bluetooth headset, and constructing an initial Bluetooth connection topology network of the target Bluetooth headset and the multiple local terminal devices;

[0007] Collecting environmental data of the initial Bluetooth connection topology network to obtain multi-dimensional network environmental data, and creating an initial connection decision list according to the multi-dimensional network environmental data;

[0008] According to the initial connection decision list, performing connection decision tests on the target Bluetooth headset and each local terminal device in turn, and collecting the data rate and transmission power of each local terminal device;

[0009] Performing feature extraction and coding vector conversion on the data rate and the transmission power to generate a connection test coding vector for each local terminal device;

[0010] Inputting the connection test coding vector into a preset connection quality monitoring model to perform connection quality analysis to obtain a connection quality indicator for each local terminal device;

[0011] The initial Bluetooth connection topology network is optimized for network weight change and decision list according to the connection quality index to generate a target connection decision list.

[0012] In a second aspect, the present application provides a connection control device for a Bluetooth headset, the connection control device for a Bluetooth headset comprising:

[0013] A search module, used to search for multiple local terminal devices within the connection range of the target Bluetooth headset, and to construct an initial Bluetooth connection topology network of the target Bluetooth headset and the multiple local terminal devices;

[0014] A creation module, used to collect environment data of the initial Bluetooth connection topology network to obtain multi-dimensional network environment data, and create an initial connection decision list according to the multi-dimensional network environment data;

[0015] A test module, used to perform a connection decision test on the target Bluetooth headset and each local terminal device in turn according to the initial connection decision list, and collect the data rate and transmission power of each local terminal device;

[0016] A conversion module, used for performing feature extraction and coding vector conversion on the data rate and the transmission power to generate a connection test coding vector for each local terminal device;

[0017] An analysis module, used for inputting the connection test coding vector into a preset connection quality monitoring model to perform connection quality analysis to obtain a connection quality indicator for each local terminal device;

[0018] The optimization module is used to perform network weight optimization and decision list optimization on the initial Bluetooth connection topology network according to the connection quality index to generate a target connection decision list.

[0019] The third aspect of the present application provides a connection control device for a Bluetooth headset, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions corresponding to the memory so that the connection control device of the Bluetooth headset executes the above-mentioned connection control method for the Bluetooth headset.

[0020] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned method for controlling the connection of a Bluetooth headset.

[0021] In the technical solution provided by the present application, by constructing the initial Bluetooth connection topology network and collecting multi-dimensional environmental data, the environmental information is fully utilized to optimize the connection decision. This strategy significantly improves the intelligence and adaptability of the connection decision, and in a complex environment, the optimal connection object can be selected more accurately, thereby improving the stability and efficiency of the connection. The feature extraction and coding vector conversion of data rate and transmission power provide a basis for monitoring the connection quality. Through such information processing, the obtained coding vector is more representative, and the influence of each link on the connection quality can be more accurately reflected in the connection quality monitoring model, which is conducive to real-time monitoring and optimization of the Bluetooth connection status, and ensuring the continuity and quality of audio transmission. The connection quality index obtained after further analysis using the connection quality monitoring model can guide the network variable weight optimization and decision list optimization. This not only enhances the dynamic adjustment ability of the method, but also responds quickly to real-time network status and environmental changes. This optimization strategy can reduce the fluctuation of connection quality caused by environmental changes, and ensure the high-quality connection experience of users when moving or using multiple devices alternately. The present application realizes the intelligent selection of the connection between the Bluetooth headset and the local terminal device, thereby improving the data rate and transmission power of the Bluetooth headset. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings corresponding to the description below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0023] Figure 1 A schematic diagram of an embodiment of a method for controlling the connection of a Bluetooth headset in an embodiment of the present application;

[0024] Figure 2 This is a schematic diagram of an embodiment of a connection control device for a Bluetooth headset in an embodiment of the present application. DETAILED DESCRIPTION

[0025] The embodiments of the present application provide a method, device, equipment and storage medium for controlling the connection of a Bluetooth headset. The terms "first", "second", "third", "fourth", etc. (if any) corresponding to the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 , an embodiment of the connection control method of the Bluetooth headset in the embodiment of the present application includes:

[0027] Step S101, searching for multiple local terminal devices within the connection range of the target Bluetooth headset, and constructing an initial Bluetooth connection topology network of the target Bluetooth headset and the multiple local terminal devices;

[0028] It is understandable that the execution subject of the present application may be a connection control device of a Bluetooth headset, or a terminal or a server, which is not limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0029] Specifically, when the Bluetooth headset enters the discovery mode, the headset will send a broadcast signal, which includes the device ID of the headset and other necessary connection parameters, such as device category and technical specifications. At the same time, its receiving module is activated to listen for responses from other devices. On this basis, the headset uses the broadcast and discovery mechanism defined in the Bluetooth core specification to search for all local terminal devices within the range that are turned on and in a discoverable state, such as smartphones, tablets or other Bluetooth accessories. The distance to each device is estimated based on the signal strength indication (RSSI) of the device response. A higher RSSI value indicates a closer distance, which helps to initially establish the relative position relationship between devices. The Bluetooth headset uses this information to build an initial Bluetooth connection topology network. In this network, each node represents a terminal device, and the connection between nodes represents a feasible Bluetooth connection path. Collect basic connection data for each device, such as supported Bluetooth versions, channel usage, and historical connection quality.

[0030] Step S102: collecting environmental data of the initial Bluetooth connection topology network to obtain multi-dimensional network environmental data, and creating an initial connection decision list according to the multi-dimensional network environmental data;

[0031] Specifically, multiple sensors are used to collect real-time data on the operating environment of each node in the network. These sensors can detect and record a variety of information including Bluetooth signal strength, surrounding electromagnetic interference, physical location of the device and its operating status. This information is combined to form the original network environment data. The original network environment data is cleaned to obtain multi-dimensional network environment data, including filtering out noise, eliminating irrelevant data and correcting erroneous data records. The multi-dimensional network environment data is classified to obtain Bluetooth signal strength data, interference frequency data and device status data. Each of these data reflects different aspects of network connection: Bluetooth signal strength data reveals the communication quality between devices, interference frequency data helps identify external interference sources that may affect the stability of the connection, and device status data provides information on the current operating status of each device. An initial connection decision list is created based on the Bluetooth signal strength data, interference frequency data and device status data. The list is sorted according to the priority of the device connection, and the priority is determined based on a comprehensive evaluation of the device's signal strength, the degree of interference and its operating status. For example, a device with high signal strength, low interference and good status will be ranked higher, so as to ensure that the Bluetooth headset can give priority to trying to establish a stable connection with the device with the best conditions during the actual connection process.

[0032] Collect Bluetooth signal strength data of each local terminal device. These data reflect the distance and signal quality between each device and the Bluetooth headset. The stronger the signal, the closer the device is to the headset, and the higher the connection possibility and stability. Prioritize multiple local terminal devices according to the Bluetooth signal strength data to obtain the first device priority sequence to ensure that the device with the strongest signal gets a higher connection priority. Consider interference factors and collect and analyze the interference frequency data of each device. These data involve electromagnetic interference sources in the surrounding environment that may affect Bluetooth connection, such as the activities of other wireless devices. By analyzing the interference frequency, sort the devices and generate the second device priority sequence, giving priority to devices with less interference, thereby reducing potential problems in the connection process. Prioritize multiple local terminal devices according to the device status data to obtain the third device priority sequence. Including information such as the power status of the device, the currently running application, and the historical connection quality. By evaluating these status data, determine which devices are more suitable for stable connection under current conditions. Priority weight analysis is performed on the collected Bluetooth signal strength, interference frequency, and device status data. Assign appropriate weights to different priority indicators. For example, if the environment is severely interfered, a higher weight may be required for the interference frequency data. After setting weights for these three types of data, use these weights to perform weighted fusion on the device priority sequence. Based on the comprehensive weights and priority sequence, the target device priority sequence is obtained. This sequence is based on the results of comprehensive analysis to ensure that the selected device has a strong signal and is in an environment with less interference, and the device status is also suitable for connection. Create an initial connection decision list corresponding to the target Bluetooth headset based on the target device priority sequence. This list guides the Bluetooth headset to try to connect to each device in order of priority during the actual connection process, thereby optimizing the overall connection effect and user experience.

[0033] Step S103: According to the initial connection decision list, a connection decision test is performed on the target Bluetooth headset and each local terminal device in turn, and the data rate and transmission power of each local terminal device are collected;

[0034] Specifically, the test sequence is arranged according to the initial connection decision list. The list sorts all local terminal devices to be connected according to the signal strength, interference and status data of the devices to ensure that the test starts with the device that is most likely to successfully connect. The target Bluetooth headset starts the connection test program, which automatically attempts to establish a connection with each terminal device in the list in turn. In each connection attempt, the data transmission test is activated to evaluate the actual performance of the connection. This includes measuring the data transmission rate, that is, the rate at which data is transmitted from one end to the other in the connected state, and the transmission power of the headset and the terminal device during the transmission process. The transmission power reflects the energy consumed by the device when sending and receiving data, which helps to evaluate the battery life and communication efficiency of the device. During each connection test, the monitoring software automatically records the relevant data of the data rate and transmission power, which will then be stored and analyzed to determine whether the connection meets the expected performance standards. If a device shows superior data transmission rate and reasonable power consumption in the test, it indicates that its connection with the Bluetooth headset is of high quality and reliability, and it will be given higher priority in the final connection decision.

[0035] Step S104: extracting features and converting coding vectors on data rate and transmission power to generate a connection test coding vector for each local terminal device;

[0036] Specifically, the rate numerical feature of the data rate and the power numerical feature of the transmission power are calculated respectively. The rate numerical feature of the data rate refers to the actual rate of data transmission during the connection process, which is usually measured in Mbps; while the power numerical feature of the transmission power reflects the energy consumption of the device during the data transmission process, which is usually measured in watts. Next, the rate numerical feature and the power numerical feature are normalized to eliminate the differences in test conditions and performance standards of different devices, so that the feature values ​​can be compared and analyzed on the same scale, and the normalized rate feature and the normalized power feature are generated. According to the degree of influence of the data rate and the transmission power on the connection quality, the influence intensity order of the two features is obtained. Based on this order, the normalized rate and power features are serialized and converted, and these features are sorted according to their importance to form a comprehensive feature sequence. The sequence comprehensively reflects the performance of each device in the actual connection process. The comprehensive feature sequence is sequence encoded and vector converted to convert the serialized features into a form that can be directly processed by the machine learning model to obtain the connection test encoding vector. The comprehensive feature sequence is compressed or converted into an encoding vector with high information density using methods such as principal component analysis or autoencoder.

[0037] Step S105: input the connection test coding vector into a preset connection quality monitoring model to perform connection quality analysis to obtain a connection quality indicator for each local terminal device;

[0038] Specifically, a connection quality monitoring model is created, which includes three main parts: a first threshold recurrent network, a second threshold recurrent network, and a softmax layer. Through the first threshold recurrent network, a high-dimensional hidden feature extraction is performed on the connection test encoding vector, and a hidden high-dimensional feature vector is extracted from the high-dimensional data. The first threshold recurrent network can effectively process sequence data by using its cyclic structure, ensuring that the dynamically changing pattern is captured from the connection test data of the time series, thereby generating a hidden feature high-dimensional vector representing the connection characteristics of each terminal device. Through the second threshold recurrent network, the hidden feature high-dimensional vector is mapped to a low-dimensional hidden feature to obtain a hidden state feature vector. The feature space is simplified and the generalization ability of the model is enhanced, while the computational complexity is reduced. Through the low-dimensional mapping, the model can focus on the most influential features and improve the accuracy of predicting the connection quality. Through the softmax layer, the connection quality analysis of the hidden state feature vector is performed. The softmax layer is a commonly used classifier, and the connection quality of each terminal device is finally evaluated and classified according to the softmax layer. By calculating the probability distribution of the hidden state feature vector, the connection quality indicators of each terminal device are output, which indicate the quality of the connection with the device.

[0039] Step S106: Perform network weight optimization and decision list optimization on the initial Bluetooth connection topology network according to the connection quality index to generate a target connection decision list.

[0040] Specifically, target network weight parameters of the initial Bluetooth connection topology network are generated according to the connection quality index. These parameters reflect the priority and stability of different device connections. The weight allocation is based on the connection quality score of each device to ensure that devices with higher quality connections receive higher priority. The initial Bluetooth connection topology network is feature clustered according to the target network weight parameters, and the devices in the network are grouped according to their connection characteristics to form a target clustering result. Clustering analysis helps identify groups of devices with similar connection properties. For example, devices with high signal strength and stability are clustered together to enhance the overall connection efficiency and reliability of the network. According to the target clustering results, the shortest path algorithm is used to calculate the subordination relationship of the initial Bluetooth connection topology network to generate a target subordination tree structure. Considering the physical distance and signal quality between devices, and through the intra-group relationship determined by clustering, it is ensured that the data transmission between devices passes through the optimal path as much as possible to reduce transmission delay and interference. After obtaining the subordination tree structure, the initial Bluetooth connection topology network is optimized for network variable weights, and the weight and connection strategy of each connection node are readjusted to ensure that the network operates efficiently according to the new subordination structure. According to the newly generated target Bluetooth connection topology network and the updated multi-dimensional network environment data, the initial connection decision list is adjusted to generate a target connection decision list. This list guides the Bluetooth headset to select the most appropriate device for connection in actual operation and optimizes the pairing and data exchange process between devices.

[0041] Based on the connection quality index of each local terminal device, the network weight is initialized, and a preliminary weight parameter is assigned to each connection in the initial Bluetooth connection topology network to obtain the first network weight parameter. These parameters reflect the initial connection preference of each device, where higher weights are assigned to devices with better connection quality. The effectiveness of the first network weight parameter is evaluated, and the first fitness value of each weight parameter is calculated. The fitness value is a standard for measuring the effectiveness of the parameter, based on the overall performance of the network, including connection stability and data transmission efficiency. According to the first fitness value, the first network weight parameter is divided into scheme groups to form different target network weight parameter groups. The target network weight parameter group is subjected to genetic, crossover and mutation processing, and a new generation of network weight parameters is generated by selecting the optimal solution, crossover combination and random mutation to obtain the second network weight parameter. Each second network weight parameter is evaluated again and its second fitness value is calculated. This evaluation more accurately reflects the actual performance of each parameter in the simulated network environment. According to the second fitness value, the second network weight parameters are optimized to determine which weight parameters can best improve the connection quality and efficiency of the entire network. The generated target network weight parameters are used to update the initial Bluetooth connection topology network to ensure that the network structure can achieve optimal connection performance and stability in actual operation.

[0042] In the embodiment of the present application, by constructing the initial Bluetooth connection topology network and collecting multi-dimensional environmental data, environmental information is fully utilized to optimize the connection decision. This strategy significantly improves the intelligence and adaptability of the connection decision, and in a complex environment, the optimal connection object can be selected more accurately, thereby improving the stability and efficiency of the connection. The feature extraction and coding vector conversion of data rate and transmission power provide a basis for monitoring the connection quality. Through such information processing, the obtained coding vector is more representative, and the influence of each link on the connection quality can be more accurately reflected in the connection quality monitoring model, which is conducive to real-time monitoring and optimization of the Bluetooth connection state, and ensuring the continuity and quality of audio transmission. The connection quality index obtained after further analysis using the connection quality monitoring model can guide the network variable weight optimization and decision list optimization. This not only enhances the dynamic adjustment ability of the method, but also can respond quickly to real-time network status and environmental changes. This optimization strategy can reduce the connection quality fluctuations caused by environmental changes, and ensure the user's high-quality connection experience when moving or using multiple devices alternately. The present application realizes the intelligent selection of the connection between the Bluetooth headset and the local terminal device, thereby improving the data rate and transmission power of the Bluetooth headset.

[0043] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0044] (1) Through multiple sensors, multi-dimensional operating environment data of the initial Bluetooth connection topology network is collected to obtain the original network environment data;

[0045] (2) Clean the original network environment data to obtain multi-dimensional network environment data;

[0046] (3) Classify the multi-dimensional network environment data to obtain Bluetooth signal strength data, interference frequency data, and device status data;

[0047] (4) Create an initial connection decision list based on Bluetooth signal strength data, interference frequency data, and device status data.

[0048] Specifically, sensors such as wireless signal sensors, location sensors, and environmental noise sensors installed on Bluetooth headsets and related devices are used to collect multi-dimensional operating environment data including signal strength, device location, and electromagnetic interference in real time. For example, signal sensors can measure the Bluetooth signal strength between surrounding devices, location sensors provide physical location information of devices, and environmental noise sensors detect the background electromagnetic noise level. These data together constitute the original network environment data set. Next, the original data is cleaned. Data cleaning includes removing outliers and noise introduced by sensor errors, data transmission errors, or external interference, and filling in missing data to ensure the accuracy and reliability of subsequent analysis. By applying data smoothing techniques and outlier detection algorithms, these data can be effectively purified to obtain multi-dimensional network environment data. For example, if a location sensor suddenly reports an outlier, this may be caused by temporary signal interference. By comparing with historical data and data from adjacent sensors, this can be verified and appropriate adjustments can be made. Multi-dimensional network environment data is classified and processed. The data is decomposed into more specific and useful information categories, such as Bluetooth signal strength data, interference frequency data, and device status data. Classification helps to apply different analysis strategies and decision logic to different types of data. Bluetooth signal strength data reflects the direct quality of communication between devices, interference frequency data helps identify external factors that may affect signal transmission, and device status data (such as battery level, software version, etc.) provides information on the current operating status of each device. Create an initial connection decision list based on Bluetooth signal strength data, interference frequency data, and device status data. The list is built by evaluating the connection priority of each device based on the analysis results of the above data. For example, the system may give priority to connecting to devices with high signal strength, low interference frequency, and sufficient battery power, ensuring that Bluetooth headsets can establish a stable connection with other devices under the most favorable conditions. For example, if the signal strength between a smartphone and a Bluetooth headset is higher than a certain threshold, and the phone is in a relatively low electromagnetic interference area and has sufficient battery, it will be ranked higher in the connection priority list.

[0049] In a specific embodiment, the process of executing the step of creating an initial connection decision list according to the Bluetooth signal strength data, the interference frequency data, and the device status data may specifically include the following steps:

[0050] (1) Prioritizing multiple local terminal devices according to the Bluetooth signal strength data to obtain a first device priority sequence;

[0051] (2) Prioritizing the plurality of local terminal devices according to the interference frequency data to obtain a second device priority sequence;

[0052] (3) Prioritizing the plurality of local terminal devices according to the device status data to obtain a third device priority sequence;

[0053] (4) performing a priority weight analysis on the Bluetooth signal strength data to obtain a first priority weight, performing a priority weight analysis on the interference frequency data to obtain a second priority weight, and performing a priority weight analysis on the device status data to obtain a third priority weight;

[0054] (5) performing priority-weighted fusion on the first device priority sequence, the second device priority sequence, and the third device priority sequence according to the first priority weight, the second priority weight, and the third priority weight to obtain a target device priority sequence;

[0055] (6) Create an initial connection decision list corresponding to the target Bluetooth headset according to the target device priority sequence.

[0056] Specifically, all local terminal devices are sorted according to the Bluetooth signal strength to generate a first device priority sequence. In this sequence, the device with the strongest signal is ranked first. High signal strength usually means better connection quality and more stable data transmission. The devices are sorted according to the interference frequency data to form a second device priority sequence. The device with the least interference is ranked first. A low interference environment helps reduce the possibility of connection failure. The devices are sorted according to the device status data to generate a third device priority sequence. The devices with high priority are usually devices with sufficient batteries and good system status, ensuring that the device will not be disconnected due to insufficient power or system problems during the connection process. Priority weight analysis is performed on the Bluetooth signal strength data to obtain a first priority weight, and priority weight analysis is performed on the interference frequency data to obtain a second priority weight, and priority weight analysis is performed on the device status data to obtain a third priority weight. Determine the relative importance of various types of data in the connection decision. For example, if the user scenario pays more attention to the stability of the connection, the weight of the Bluetooth signal strength may be set higher; if the environmental interference is particularly severe, the weight of the interference frequency data will increase; and in scenarios where the device is frequently used and needs to maintain the connection for a long time, the weight of the device status data will also increase accordingly. Next, the first, second, and third device priority sequences are weighted and fused according to the set weights. A weighted average method is used, where the final priority of each device is determined by its position in the three sequences and the corresponding weights. For example, if a device ranks very high in the signal strength sequence, but ranks low in the interference frequency sequence, and ranks in the middle in the device status sequence, the system will calculate a comprehensive score based on the weights to determine the final ranking of the device. Based on the weighted fused target device priority sequence, a corresponding initial connection decision list is created. This list guides the Bluetooth headset to first try to connect to the top-ranked device during the actual connection process.

[0057] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0058] (1) Calculate the rate numerical characteristics of the data rate and the power numerical characteristics of the transmission power respectively;

[0059] (2) Performing feature normalization processing on the rate numerical features and the power numerical features to generate normalized rate features and normalized power features;

[0060] (3) Obtaining the order of influence strength of data rate and transmission power, and performing serial conversion on the normalized rate characteristics and normalized power characteristics based on the order of influence strength to generate a comprehensive characteristic sequence for each local terminal device;

[0061] (4) Perform sequence encoding and vector conversion on the comprehensive feature sequence to generate a connection test coding vector for each local terminal device.

[0062] Specifically, the rate numerical feature of the data rate and the power numerical feature of the transmission power are calculated respectively. The data rate feature reflects the speed of data transmission under a specific connection, usually measured in Mbps (megabits per second), while the transmission power feature describes the amount of power consumed during data transmission, usually in watts. Next, the rate numerical feature and the power numerical feature are normalized. The data rate and power output between different devices may vary greatly, and directly comparing these values ​​may lead to inaccurate data analysis. Normalization is usually achieved by subtracting the minimum value of each feature value from its range (i.e., the maximum value minus the minimum value), so that all data are adjusted to the same scale between 0 and 1. For example, if the minimum value of the data rate in a series of tests is 5Mbps and the maximum value is 15Mbps, the normalized value of the data rate of 10Mbps is (10-5) / (15-5) = 0.5. Obtain the order of the intensity of the impact of data rate and transmission power on connection quality. Based on historical data and a pre-set model, determine which feature has a greater impact on connection quality. For example, if the data shows that the impact of data rate on connection stability is greater than power consumption, then the rate feature will be given a higher priority in subsequent processing. According to the order of influence intensity, the normalized rate features and power features are serialized and converted, and these features are sorted according to their importance to form a comprehensive feature sequence. The sequence comprehensively reflects the performance of each device in the actual connection process. For example, if the data rate is regarded as a more critical feature, the normalized rate feature will be placed at the front of the sequence, and the normalized power feature will be placed at the back. The comprehensive feature sequence is sequence encoded and vector converted, and the serialized features are converted into a form that can be directly processed by the machine learning model to obtain a connection test coding vector. For example, using methods such as principal component analysis or autoencoders, the comprehensive feature sequence is compressed or converted into a coding vector with a higher information density. These coding vectors are used as input in the prediction model to evaluate and predict the connection quality of each terminal device, thereby guiding the Bluetooth headset to make the best connection decision in actual use. In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0063] (1) creating a connection quality monitoring model, wherein the connection quality monitoring model includes: a first threshold recurrent network, a second threshold recurrent network, and a softmax layer;

[0064] (2) extracting high-dimensional hidden features from the connection test encoding vector through the first threshold recurrent network to obtain a high-dimensional vector of hidden features;

[0065] (3) Through the second threshold recurrent network, the hidden feature high-dimensional vector is mapped to a hidden feature low-dimensional vector to obtain a hidden state feature vector;

[0066] (4) Through the softmax layer, the connection quality analysis is performed on the hidden state feature vector to obtain the connection quality index of each local terminal device.

[0067] Specifically, a connection quality monitoring model is created, including a first threshold recurrent network, a second threshold recurrent network, and a softmax layer. Through the first threshold recurrent network, high-dimensional hidden features are extracted from the connection test encoding vector, and the architecture of the recurrent neural network is used to process time series data or data features with time dependence. These hidden features can capture complex patterns and long-term dependencies in connection dynamics, which helps to understand and predict the interaction between Bluetooth devices. For example, if a Bluetooth headset performs multiple connection tests with the same smartphone under different environmental conditions, the first threshold recurrent network will analyze the changes in these connection tests and extract the key hidden factors that affect the stability and quality of the connection. Next, through the second threshold recurrent network, the hidden feature high-dimensional vector is mapped to a low-dimensional hidden feature to obtain a hidden state feature vector. The low-dimensional mapping helps to simplify the model complexity, improve processing efficiency and prediction accuracy. The low-dimensional feature vector more concentratedly represents the core attributes of the data, reducing the interference of noise and irrelevant information. For example, through the complex high-dimensional data extracted from multiple connection tests, the second threshold recurrent network can identify the most critical factors that best predict the quality of the connection. Through the softmax layer, the connection quality analysis of the hidden state feature vector is performed to obtain the connection quality index of each local terminal device. The softmax layer is an output layer applied to classification tasks, which can map the output of one or more linear models into a probability distribution. The softmax layer scores and classifies the connection quality of each local terminal device based on the hidden state feature vector, and outputs the probability index of the connection quality of each device. These indicators reflect the performance of each device in future connections and help the decision-making system optimize the connection strategy, such as selecting the device with the highest expected connection quality. For example, suppose a Bluetooth headset needs to be connected to multiple devices, such as smartphones, tablets, and laptops. After the above model analysis, the connection quality index of each device may be 0.9, 0.75, and 0.6. This shows that in the current environment and device status, the connection with the smartphone is the most stable and reliable, so the system will give priority to smartphones for connection.

[0068] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0069] (1) Generate target network weight parameters of the initial Bluetooth connection topology network based on the connection quality index;

[0070] (2) Perform feature clustering on the initial Bluetooth connection topology network according to the target network weight parameters to obtain the target clustering results;

[0071] (3) Based on the target clustering results, the shortest path algorithm is used to calculate the subordinate relationship of the initial Bluetooth connection topology network to obtain the target subordinate relationship tree structure;

[0072] (4) According to the target subordinate relationship tree structure, the initial Bluetooth connection topology network is optimized with network variable weights to obtain the target Bluetooth connection topology network;

[0073] (5) According to the target Bluetooth connection topology network and multi-dimensional network environment data, the initial connection decision list is adjusted to generate a target connection decision list.

[0074] Specifically, a target network weight parameter of the initial Bluetooth connection topology network is generated based on the connection quality index. The connection quality index is usually based on a comprehensive evaluation of multiple factors, including signal strength, interference level, device response time, etc. The weight parameter is the basis for adjusting the connection topology network, and the purpose is to give priority to devices with high connection quality, thereby improving the overall network performance. Next, based on the generated target network weight parameter, the initial Bluetooth connection topology network is feature clustered. Devices are grouped by a clustering algorithm so that the devices in each group have similar network weights and connection characteristics. This can be achieved through various clustering techniques, such as K-means or hierarchical clustering. For example, assuming there are five devices, three of which have similar connection quality indicators and the other two are lower, the clustering results may divide the first three devices into one group and the remaining two into another group. Based on the clustering results, the affiliation of the initial Bluetooth connection topology network is calculated using the shortest path algorithm to form a target affiliation tree structure. This tree structure helps to determine the hierarchy and dependency between devices in the network and ensure efficient data transmission between devices. For example, if device C is in a central position in the tree structure, it may become a relay point for other devices to connect, especially when the direct connection between devices is poor. Next, according to the target affiliation tree structure, the initial Bluetooth connection topology network is optimized by network variable weights, including readjusting the connection weights and path selection in the network to improve the overall efficiency and stability of the network. For example, if the original connection path of a device is no longer optimal due to a change in position, the network variable weight optimization will find a new optimal path and adjust the connection structure to adapt to this change. Based on the optimized target Bluetooth connection topology network and the latest multi-dimensional network environment data, the initial connection decision list is adjusted to generate the final target connection decision list. This list guides the Bluetooth headset to select the most appropriate device for connection in actual operation. For example, if the final network optimization result shows that a new device has the highest connection quality, then this device will be placed at the top of the connection decision list and become the preferred device for connection.

[0075] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0076] (1) Initializing the network weight of the initial Bluetooth connection topology network according to the connection quality indicator to obtain a plurality of first network weight parameters;

[0077] (2) calculating the first fitness value of each first network weight parameter respectively, and dividing the first network weight parameters into scheme groups according to the first fitness values ​​to obtain a target network weight parameter group;

[0078] (3) performing heredity, crossover and mutation processing on the target network weight parameter group to obtain a plurality of second network weight parameters, and respectively calculating the second fitness value of each second network weight parameter;

[0079] (4) Optimizing the plurality of second network weight parameters according to the second fitness value to generate target network weight parameters of the initial Bluetooth connection topology network.

[0080] Specifically, based on the connection quality index, the network weight of the initial Bluetooth connection topology network is initialized. These indicators may include factors such as signal strength, connection stability, and historical connection success rate between devices. For example, if a device has shown high stability and high data transmission rate in previous connection tests, it will be given a higher weight in the initial network weight. The system assigns an initial weight to each connection in the network, and these weights constitute the first network weight parameter. Next, the first fitness value of each first network weight parameter is calculated. The fitness value is an important indicator for evaluating the effectiveness of a certain network configuration, which can be measured based on multiple dimensions, such as the overall transmission efficiency of the network, the average connection delay, and the load balancing of the network. According to these fitness values, the first network weight parameters are divided into scheme groups, and network weight configurations with similar performance are grouped together to form multiple different target network weight parameter groups. The target network weight parameter groups are subjected to genetic, crossover and mutation processing. The genetic process simulates the natural selection mechanism in biological evolution, and those network weight parameters with high fitness are preferred. The crossover process simulates the genetic recombination of organisms, and some parameters of the two network weight configurations are exchanged in order to generate a new network weight configuration. Mutation processing is to make small random adjustments to the existing network weight parameters to explore possible new effective configurations. For example, if two network configurations A and B perform well under different environmental conditions, a new network configuration C may be generated through crossover and mutation, which inherits the advantages of A and B, but may also have new adaptive characteristics. After each inheritance, crossover and mutation, the second fitness value is calculated for the newly generated second network weight parameters to evaluate the network performance of the new configuration and ensure that the optimization direction is correct. The second fitness value reflects the potential performance of the network configuration after genetic algorithm processing in actual operation. According to the second fitness value, the optimization solution is performed, and the network weight configuration with the highest fitness is selected as the target network weight parameter of the initial Bluetooth connection topology network. Ensure that the network configuration can provide the best connection quality and performance in actual use. For example, through multiple rounds of genetic algorithm optimization, a network configuration may be determined in which the weights of certain key connections are significantly increased to optimize data flow and reduce connection delay.

[0081] The above describes the connection control method of the Bluetooth headset in the embodiment of the present application. The following describes the connection control device of the Bluetooth headset in the embodiment of the present application. Figure 2 , an embodiment of the connection control device of the Bluetooth headset in the embodiment of the present application includes:

[0082] A search module 201 is used to search for multiple local terminal devices within the connection range of the target Bluetooth headset and to construct an initial Bluetooth connection topology network of the target Bluetooth headset and the multiple local terminal devices;

[0083] A creation module 202 is used to collect environment data of the initial Bluetooth connection topology network, obtain multi-dimensional network environment data, and create an initial connection decision list according to the multi-dimensional network environment data;

[0084] The test module 203 is used to perform a connection decision test on the target Bluetooth headset and each local terminal device in turn according to the initial connection decision list, and collect the data rate and transmission power of each local terminal device;

[0085] The conversion module 204 is used to extract features and convert coding vectors of data rate and transmission power to generate a connection test coding vector for each local terminal device;

[0086] An analysis module 205 is used to input the connection test coding vector into a preset connection quality monitoring model to perform connection quality analysis to obtain a connection quality indicator for each local terminal device;

[0087] The optimization module 206 is used to perform network weight optimization and decision list optimization on the initial Bluetooth connection topology network according to the connection quality index, and generate a target connection decision list.

[0088] Through the collaboration of the above components, by constructing the initial Bluetooth connection topology network and collecting multi-dimensional environmental data, full use is made of environmental information to optimize connection decisions. This strategy significantly improves the intelligence and adaptability of connection decisions. In complex environments, the optimal connection object can be selected more accurately, thereby improving the stability and efficiency of the connection. The feature extraction and coding vector conversion of data rate and transmission power provide a basis for monitoring the connection quality. Through such information processing, the obtained coding vector is more representative, and the influence of each link on the connection quality can be more accurately reflected in the connection quality monitoring model, which is conducive to real-time monitoring and optimization of the Bluetooth connection status, and ensuring the continuity and quality of audio transmission. The connection quality index obtained after further analysis using the connection quality monitoring model can guide network variable weight optimization and decision list optimization. This not only enhances the dynamic adjustment ability of the method, but also responds quickly to real-time network status and environmental changes. This optimization strategy can reduce the fluctuation of connection quality caused by environmental changes, and ensure the high-quality connection experience of users when moving or using multiple devices alternately. This application realizes the intelligent selection of the connection between the Bluetooth headset and the local terminal device, thereby improving the data rate and transmission power of the Bluetooth headset.

[0089] The present application also provides a connection control device for a Bluetooth headset, the connection control device for the Bluetooth headset comprising a memory and a processor, the memory storing computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the steps of the connection control method for the Bluetooth headset corresponding to the above-mentioned embodiments.

[0090] The present application also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the connection control method of the Bluetooth headset.

[0091] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes of the aforementioned method embodiments, and will not be repeated here.

[0092] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0093] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A connection control method for a Bluetooth headset, characterized in that: The connection control method of the Bluetooth headset includes: Searching for multiple local terminal devices within the connection range of the target Bluetooth headset, and constructing an initial Bluetooth connection topology network of the target Bluetooth headset and the multiple local terminal devices; Collecting environmental data of the initial Bluetooth connection topology network to obtain multi-dimensional network environmental data, and creating an initial connection decision list according to the multi-dimensional network environmental data; According to the initial connection decision list, performing connection decision tests on the target Bluetooth headset and each local terminal device in turn, and collecting the data rate and transmission power of each local terminal device; Performing feature extraction and coding vector conversion on the data rate and the transmission power to generate a connection test coding vector for each local terminal device; Inputting the connection test coding vector into a preset connection quality monitoring model to perform connection quality analysis to obtain a connection quality indicator for each local terminal device; The initial Bluetooth connection topology network is optimized for network weight change and decision list according to the connection quality index to generate a target connection decision list.

2. The connection control method of the Bluetooth headset according to claim 1, characterized in that: The collecting of environment data of the initial Bluetooth connection topology network to obtain multi-dimensional network environment data, and creating an initial connection decision list according to the multi-dimensional network environment data, includes: Through multiple sensors, multi-dimensional operating environment data collection is performed on the initial Bluetooth connection topology network to obtain original network environment data; Performing data cleaning on the original network environment data to obtain multi-dimensional network environment data; Classifying the multi-dimensional network environment data to obtain Bluetooth signal strength data, interference frequency data, and device status data; An initial connection decision list is created according to the Bluetooth signal strength data, the interference frequency data, and the device status data.

3. The connection control method of the Bluetooth headset according to claim 2, characterized in that: The creating an initial connection decision list according to the Bluetooth signal strength data, the interference frequency data, and the device status data includes: Prioritize the multiple local terminal devices according to the Bluetooth signal strength data to obtain a first device priority sequence; Prioritize the multiple local terminal devices according to the interference frequency data to obtain a second device priority sequence; Prioritize the multiple local terminal devices according to the device status data to obtain a third device priority sequence; Performing a priority weight analysis on the Bluetooth signal strength data to obtain a first priority weight, performing a priority weight analysis on the interference frequency data to obtain a second priority weight, and performing a priority weight analysis on the device status data to obtain a third priority weight; According to the first priority weight, the second priority weight and the third priority weight, priority-weighted fusion is performed on the first device priority sequence, the second device priority sequence and the third device priority sequence to obtain a target device priority sequence; An initial connection decision list corresponding to the target Bluetooth headset is created according to the target device priority sequence.

4. The connection control method of the Bluetooth headset according to claim 1, characterized in that: The feature extraction and coding vector conversion of the data rate and the transmission power to generate a connection test coding vector for each local terminal device includes: respectively calculating a rate numerical characteristic of the data rate and a power numerical characteristic of the transmission power; Performing feature normalization processing on the rate numerical feature and the power numerical feature to generate a normalized rate feature and a normalized power feature; Obtaining the order of influence strength of the data rate and the transmission power, and performing serial conversion on the normalized rate feature and the normalized power feature based on the order of influence strength to generate a comprehensive feature sequence for each local terminal device; The comprehensive feature sequence is subjected to sequence encoding and vector conversion to generate a connection test coding vector for each local terminal device.

5. The connection control method of the Bluetooth headset according to claim 1, characterized in that: The step of inputting the connection test coding vector into a preset connection quality monitoring model to perform connection quality analysis to obtain a connection quality indicator for each local terminal device includes: Creating a connection quality monitoring model, wherein the connection quality monitoring model includes: a first threshold recurrent network, a second threshold recurrent network and a softmax layer; Performing high-dimensional hidden feature extraction on the connection test encoding vector through the first threshold recurrent network to obtain a hidden feature high-dimensional vector; Performing hidden feature low-dimensional mapping on the hidden feature high-dimensional vector through the second threshold recurrent network to obtain a hidden state feature vector; Through the softmax layer, the connection quality analysis is performed on the hidden state feature vector to obtain the connection quality index of each local terminal device.

6. The connection control method of the Bluetooth headset according to claim 1, characterized in that: The performing network weight optimization and decision list optimization on the initial Bluetooth connection topology network according to the connection quality indicator to generate a target connection decision list includes: Generating a target network weight parameter of the initial Bluetooth connection topology network according to the connection quality indicator; Performing feature clustering on the initial Bluetooth connection topology network according to the target network weight parameter to obtain a target clustering result; According to the target clustering result, a shortest path algorithm is used to calculate the subordination relationship of the initial Bluetooth connection topology network to obtain a target subordination relationship tree structure; According to the target affiliation tree structure, performing network weight optimization on the initial Bluetooth connection topology network to obtain a target Bluetooth connection topology network; According to the target Bluetooth connection topology network and the multi-dimensional network environment data, the initial connection decision list is adjusted to generate a target connection decision list.

7. The connection control method of the Bluetooth headset according to claim 6, characterized in that: The step of generating a target network weight parameter of the initial Bluetooth connection topology network according to the connection quality indicator includes: Initializing the network weight of the initial Bluetooth connection topology network according to the connection quality indicator to obtain a plurality of first network weight parameters; Calculating the first fitness value of each first network weight parameter respectively, and dividing the first network weight parameters into scheme groups according to the first fitness values ​​to obtain a target network weight parameter group; Performing genetic, crossover and mutation processing on the target network weight parameter population to obtain a plurality of second network weight parameters, and respectively calculating a second fitness value of each second network weight parameter; According to the second fitness value, the plurality of second network weight parameters are optimized to generate target network weight parameters of the initial Bluetooth connection topology network.

8. A connection control device for a Bluetooth headset, characterized in that: The connection control device of the Bluetooth headset comprises: A search module, used to search for multiple local terminal devices within the connection range of the target Bluetooth headset, and to construct an initial Bluetooth connection topology network of the target Bluetooth headset and the multiple local terminal devices; A creation module, used to collect environment data of the initial Bluetooth connection topology network to obtain multi-dimensional network environment data, and create an initial connection decision list according to the multi-dimensional network environment data; A test module, used to perform a connection decision test on the target Bluetooth headset and each local terminal device in turn according to the initial connection decision list, and collect the data rate and transmission power of each local terminal device; A conversion module, used for performing feature extraction and coding vector conversion on the data rate and the transmission power to generate a connection test coding vector for each local terminal device; An analysis module, used for inputting the connection test coding vector into a preset connection quality monitoring model to perform connection quality analysis to obtain a connection quality indicator for each local terminal device; The optimization module is used to perform network weight optimization and decision list optimization on the initial Bluetooth connection topology network according to the connection quality index to generate a target connection decision list.

9. A connection control device for a Bluetooth headset, characterized in that: The connection control device of the Bluetooth headset includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor calls the instruction corresponding to the memory, so that the connection control device of the Bluetooth headset executes the connection control method of the Bluetooth headset according to any one of claims 1 to 7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the connection control method of the Bluetooth headset according to any one of claims 1 to 7 is implemented.

Images