Method and system for quickly connecting head-mounted Bluetooth headset to smart TV

Abstract

The present invention relates to a method and system for quickly connecting a head-mounted Bluetooth headset to a smart TV. The method comprises the following steps: collecting parameters of the connection between the head-mounted Bluetooth headset and the smart TV to generate a standardized feature matrix; inputting the parameters into a parallel signal processor, respectively calculating the time domain evaluation parameters and the frequency domain evaluation parameters through a time domain processing unit and a frequency domain processing unit, and generating connection quality evaluation data after fusion through time-frequency entropy constraints; performing neural network classification operations to generate a dynamic optimization parameter set; inputting the dynamic optimization parameter set into a fast device scanning module, using a two-stage scanning mechanism to obtain feature information of surrounding Bluetooth devices, and generating device feature correlation data; calculating the optimal connection cost value, establishing a Bluetooth connection when the optimal connection cost value meets a preset threshold condition, and generating a real-time performance evaluation result. The present invention ensures the reliability and stability of the connection process, and significantly improves the recognition efficiency and accuracy of the target device.

Description

Technical Field

[0001] The present invention relates to the technical field of Bluetooth connection, and in particular to a method and system for quickly connecting a head-mounted Bluetooth headset to a smart TV. Background Art

[0002] With the rapid development of smart home technology, the demand for wireless connection between Bluetooth headsets and smart TVs is growing, and users have higher requirements for audio transmission quality and connection convenience. However, in actual applications, since Bluetooth connections are affected by environmental interference, multi-device competition, and signal attenuation, problems such as unstable connection, high latency, and audio distortion often occur, which seriously affects the user experience.

[0003] At present, traditional Bluetooth connection methods mainly rely on a single signal strength indicator for device selection and connection, lacking a comprehensive evaluation and dynamic optimization mechanism for connection quality. In addition, in a complex environment where multiple devices coexist, simple device scanning and connection strategies often lead to low connection efficiency and even incorrect connections. Existing Bluetooth connection technology also has problems with fixed parameter configuration and poor adaptability. It cannot be dynamically adjusted according to the real-time connection status, resulting in large fluctuations in connection quality. At the same time, the lack of an effective performance evaluation and feedback mechanism makes it difficult for the system to promptly discover and solve problems in the connection process. Summary of the invention

[0004] The main purpose of the present invention is to provide a method and system for quickly connecting a head-mounted Bluetooth headset to a smart TV. The present invention ensures the reliability and stability of the connection process and significantly improves the recognition efficiency and accuracy of the target device.

[0005] To achieve the above object, the present invention provides a method for quickly connecting a head-mounted Bluetooth headset to a smart TV, comprising the following steps:

[0006] Collect parameters of the physical layer, protocol layer, and service layer of the connection between the Bluetooth headset and the smart TV to generate a standardized feature matrix;

[0007] The standardized feature matrix is ​​input into a parallel signal processor, and a time domain evaluation parameter and a frequency domain evaluation parameter are calculated by a time domain processing unit and a frequency domain processing unit respectively, and connection quality evaluation data is generated after fusion through time-frequency entropy constraints;

[0008] Performing a neural network classification operation on the connection quality assessment data, extracting target parameters that affect the connection state, and generating a dynamic optimization parameter set;

[0009] Input the dynamic optimization parameter set into the fast device scanning module, adopt a two-stage scanning mechanism to obtain the characteristic information of the surrounding Bluetooth devices, and generate device characteristic association data;

[0010] The optimal connection cost is calculated according to the device feature association data, a Bluetooth connection is established when the optimal connection cost meets a preset threshold condition, and connection status data is continuously collected to generate a real-time performance evaluation result.

[0011] The present invention also provides a head-mounted Bluetooth headset and a smart TV quick connection system, comprising:

[0012] The acquisition module is used to collect parameters of the physical layer, protocol layer and service layer of the connection between the head-mounted Bluetooth headset and the smart TV, and generate a standardized feature matrix;

[0013] A calculation module, used for inputting the standardized feature matrix into a parallel signal processor, calculating the time domain evaluation parameter and the frequency domain evaluation parameter through a time domain processing unit and a frequency domain processing unit respectively, and generating connection quality evaluation data after fusion through time-frequency entropy constraints;

[0014] An extraction module, used to perform a neural network classification operation on the connection quality assessment data, extract target parameters that affect the connection status, and generate a dynamic optimization parameter set;

[0015] A scanning module, used to input the dynamic optimization parameter set into a fast device scanning module, adopt a two-stage scanning mechanism to obtain characteristic information of surrounding Bluetooth devices, and generate device characteristic association data;

[0016] The connection module is used to calculate the optimal connection cost value according to the device feature association data, establish a Bluetooth connection when the optimal connection cost value meets a preset threshold condition, and continuously collect connection status data to generate real-time performance evaluation results.

[0017] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0018] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0019] In summary, the technical solution provided by the present invention realizes comprehensive perception and accurate evaluation of the connection status by establishing a multi-dimensional parameter collection mechanism for the physical layer, protocol layer and service layer, combined with time domain and frequency domain feature extraction technology; adopts parallel signal processing and time-frequency entropy constraint fusion technology to improve the accuracy and real-time performance of connection quality evaluation, uses neural network classification technology to intelligently analyze the connection status, and enhances the system's adaptability to environmental changes by dynamically optimizing the generation of parameter sets; innovatively proposes a two-stage device scanning mechanism, which significantly improves the recognition efficiency and accuracy of the target device in conjunction with the device feature correlation calculation; designs a connection decision mechanism and real-time performance evaluation method based on adaptive thresholds to ensure the reliability and stability of the connection process; and establishes a closed-loop connection quality optimization mechanism through dynamic parameter adjustment and performance monitoring feedback, which effectively improves the overall performance of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the steps of a method for quickly connecting a head-mounted Bluetooth headset to a smart TV in one embodiment of the present invention;

[0021] Figure 2 This is a structural block diagram of a system for quickly connecting a head-mounted Bluetooth headset to a smart TV in one embodiment of the present invention;

[0022] Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0023] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0025] Reference Figure 1 This embodiment provides a method for quickly connecting a head-mounted Bluetooth headset to a smart TV, comprising the following steps:

[0026] S1, collects parameters of the physical layer, protocol layer and service layer of the connection between the head-mounted Bluetooth headset and the smart TV, and generates a standardized feature matrix;

[0027] Among them, an interaction mechanism of the physical layer, protocol layer and service layer is constructed between the head-mounted Bluetooth headset and the smart TV, in which the physical layer is responsible for basic signal transmission and reception, the protocol layer is responsible for the analysis of the Bluetooth protocol and the execution of communication rules, and the service layer is responsible for the implementation of specific audio and video data transmission and functional services. At the physical layer, the signal transmission strength is sampled in real time through the hardware interface and the signal acquisition module, and the collected signal strength data reflects the stability and transmission quality of the signal between the devices. Based on the signal strength data, the connection stability data is obtained by calculating the difference in signal strength between adjacent time points, which shows the fluctuation characteristics of the signal during the transmission process. At the same time, based on the analysis of the connection stability data, the energy consumption during the signal transmission process is monitored to obtain power consumption control data, which helps to evaluate the energy efficiency of Bluetooth headsets and TVs in different connection states. In the protocol layer, the time measurement is performed for the transmission process of the data packet, and the transmission delay data is calculated. These delay data describe the time difference from the sender to the receiver. Based on the transmission delay data, the packet loss data is obtained by counting the sending and receiving of the data packet. The data records the number of data packets that failed to reach the receiver during the transmission process. On this basis, error analysis is performed on the packet loss data. By calculating the ratio of the number of erroneous bits in the received data packet to the total number of bits, the error data is obtained. This data is an important indicator for evaluating transmission reliability and can reflect problems such as signal interference and decoding errors. In terms of the service layer, the synchronization of the transmitted audio and video data is analyzed. By comparing the time difference between the audio signal and the video signal, the synchronization deviation data is obtained. This data is an important reference for ensuring a smooth user experience. In actual operation, synchronization deviation will cause the phenomenon of audio and video being out of sync. The signal strength data, connection stability data, power consumption control data, transmission delay data, packet loss data, error data and synchronization deviation data are combined to form an initial acquisition matrix. The matrix contains key performance parameters of various aspects of the physical layer, protocol layer and service layer. The data in the initial acquisition matrix is ​​subjected to time domain feature extraction and frequency domain feature extraction. The time domain feature extraction mainly analyzes the characteristics of data changes over time, such as the amplitude fluctuation and delay change of the signal, while the frequency domain feature extraction obtains the frequency distribution characteristics of the data through spectrum analysis, such as the main frequency component and interference frequency of the signal. The extracted time domain and frequency domain features are processed by a nonlinear signal enhancer, which removes noise and amplifies key features, making the data more suitable for subsequent analysis and decision-making. On this basis, the data is normalized to generate a standardized feature matrix.

[0028] The initial acquisition matrix is ​​analyzed in the time domain. By calculating the difference sequence of signal strengths at adjacent time points, the signal's time domain change data is extracted to reflect the dynamic change characteristics of the signal in time, which is used to capture the jitter amplitude and short-term fluctuation law of the signal. Based on the time domain change data, trend analysis is carried out. By performing sliding average calculation on N sampling points, short-term noise interference is effectively removed and the long-term trend of the signal is extracted, and a time domain feature vector is generated, which contains the main change trend of the signal. At the same time, the initial acquisition matrix is ​​analyzed in the frequency domain. The fast Fourier transform algorithm is used to convert the time domain signal into a frequency domain representation to obtain a spectrum data sequence, which reveals the energy distribution of the signal at different frequencies. The spectrum data sequence is subjected to frequency domain feature extraction, and the energy distribution of the frequency components is analyzed to extract key indicators such as the main frequency components and bandwidth characteristics. The results of the frequency domain analysis are presented in the form of frequency domain feature vectors, which contain the frequency distribution characteristics and energy concentration trend of the signal. The time domain feature vector and the frequency domain feature vector are simultaneously input into the nonlinear signal enhancer for processing. The nonlinear signal enhancer uses an adaptive gain control algorithm to adjust the gain according to the dynamic changes of signal characteristics, so that weak features can be amplified, while effectively suppressing noise components to obtain enhanced feature data. On this basis, the enhanced feature data is standardized and transformed to eliminate the dimensional differences and range inconsistencies in the data, and to generate standardized feature data. The feature mapping relationship is constructed based on the standardized feature data. By calculating the correlation coefficient matrix of time domain features and frequency domain features, the intrinsic correlation between the two types of features is revealed. The calculation results of the correlation coefficient matrix can provide rich information for describing the coupling relationship and complementary characteristics of signals in different dimensions. On this basis, the feature correlation data is reconstructed into a matrix, and the standardized feature matrix is ​​generated by effectively combining the standardized features of the time domain and frequency domain.

[0029] S2, input the standardized feature matrix into the parallel signal processor, calculate the time domain evaluation parameters and the frequency domain evaluation parameters through the time domain processing unit and the frequency domain processing unit respectively, and generate the connection quality evaluation data after the time-frequency entropy constraint fusion;

[0030] Specifically, the standardized feature matrix is ​​input into a parallel signal processor, wherein the parallel signal processor includes a time domain processing unit and a frequency domain processing unit. Through an efficient parallel computing architecture, the standardized feature matrix is ​​divided into a time domain data stream and a frequency domain data stream. The stability analysis of the time domain data stream is performed. The stability characteristics of the time domain data are evaluated by calculating three key indicators: topological stability, connectivity availability, and failure controllability. Topological stability measures the timing reliability of the signal by analyzing the distribution consistency of data at different time points, connectivity availability reflects the robustness of the network structure by evaluating the accessibility of the data transmission path, and failure controllability is used to measure the size of the system recovery capability in the case of data loss or interference. These three indicators together constitute the time domain stability index. At the same time, the characteristics of the frequency domain data stream are analyzed. By calculating the three indicators of transmission stability, packet loss availability, and error controllability, the dynamic characteristics of the frequency domain signal can be effectively described. Transmission stability is used to reflect the consistency and energy concentration of data in the frequency domain. Packet loss availability measures the availability of frequency domain data by analyzing the packet loss ratio during data transmission, while error controllability is used to describe the controllability of frequency domain signals when errors occur and their impact on overall performance. These calculation results form frequency domain characteristic indicators. The time domain stability indicator is input into the time domain correlation analysis unit. By calculating the linear correlation coefficient of the time domain data, the correlation of the time domain data in different dimensions is revealed, and the time domain evaluation parameters are generated. Similarly, the frequency domain characteristic indicator is input into the frequency domain correlation analysis unit. By calculating the spectral correlation coefficient of the frequency domain data, the coupling relationship between the frequency components of the frequency domain signal can be effectively captured, and the frequency domain evaluation parameters can be generated. The entropy values ​​of the time domain evaluation parameters and the frequency domain evaluation parameters are calculated. The time-frequency entropy constraint value can be obtained by calculating the information entropy. This constraint value plays a key role in the fusion process, describing the uncertainty and information density of the data. Based on the time-frequency entropy constraint value, the least square method is used to solve the fusion weight coefficient through the time-frequency weight calculation method, and the contribution weight of the time domain and frequency domain data in the final quality assessment is optimized, making the overall assessment result more accurate and reliable. According to the fusion weight coefficient, the time domain evaluation parameters and the frequency domain evaluation parameters are linearly weighted to finally generate the connection quality assessment data.

[0031] S3, performing neural network classification operation on the connection quality assessment data, extracting target parameters that affect the connection status, and generating a dynamic optimization parameter set;

[0032] It should be noted that the connection quality assessment data is input into the neural network classification model, which consists of an input layer, three hidden layers and an output layer. The number of nodes in the input layer is consistent with the dimension of the connection quality assessment data to ensure that all assessment data features can be fully input. The first hidden layer contains 128 neurons, and the ReLU activation function is used to enhance the nonlinear expression ability of the model; the second hidden layer is reduced to 64 neurons, and the ReLU activation function is used to extract intermediate features; the third hidden layer contains 32 neurons, and the ReLU activation function is used to refine and compress high-level features. The output layer contains 2 neurons, and the Softmax activation function is used to classify the connection state, generate classification results, and identify whether the current connection quality belongs to a high-performance state or a low-performance state. The reverse feature importance calculation is performed on the connection state classification results. By evaluating the contribution of each input feature to the classification result, a feature importance vector is generated. The gradient backpropagation technology is used to trace back from the classification result of the output layer to the input layer, and the absolute value of the gradient of each input feature is calculated to quantify its influence on the classification result. The size of this vector reflects the relative importance of each feature in the connection quality assessment. Based on the feature importance vector, the features are sorted from high to low according to the importance score, and the key features that have a significant impact on the classification results are screened out. In order to ensure that the screened features can effectively reflect the changes in the connection state, a preset threshold is set, and the features with importance scores higher than the threshold are selected as target parameters to form a target parameter sequence. The target parameter sequence represents the main variables that affect the performance during the connection process. A sensitivity analysis is performed on each parameter in the target parameter sequence to evaluate the specific impact of each parameter change on the connection state. By making a small adjustment to each parameter and observing the changes in the classification results, the sensitivity of the parameter to the connection state is quantified, and a parameter sensitivity matrix is ​​generated, in which each element represents the intensity of the impact of a specific parameter on the connection state under different conditions. Based on the parameter sensitivity matrix, the parameter adjustment range is calculated. By analyzing the actual impact range of each parameter change on the connection state, the upper and lower limits of the parameter are set to form parameter constraints. On this basis, an optimization objective function is constructed, the goal of which is to minimize the weighted sum of the connection state classification error and the parameter adjustment amplitude, thereby improving the classification accuracy while reducing the disturbance of the parameter adjustment to the system. Apply the gradient descent optimization method to the optimization objective function, and iteratively update the parameter values ​​until convergence to obtain the optimal parameter combination. In each iteration, adjust the parameter value according to the gradient direction so that the objective function gradually approaches the global optimal value. When the objective function converges, the final parameter combination is the optimization result. The optimal parameter combination is grouped according to different functional groups, including signal gain group, power control group, and coding mode group. The grouped dynamic optimization parameter set can be flexibly applied to different connection scenarios to achieve comprehensive optimization of signal transmission quality, power consumption efficiency, and coding efficiency.

[0033] S4, inputting the dynamically optimized parameter set into the fast device scanning module, using a two-stage scanning mechanism to obtain feature information of surrounding Bluetooth devices, and generating device feature correlation data;

[0034] Specifically, in the first stage of scanning, the signal gain parameter in the dynamic optimization parameter set is adjusted in strength, and the scanning range is expanded or contracted by changing the transmission power, so as to perform a broadcast scan on the surrounding devices. The scanning target of this stage is to quickly collect the original scan list including MAC address, signal strength value and device type. The original scan list is a preliminary device set in the Bluetooth environment, which contains all Bluetooth device information that can be detected by the headset. The original scan list is screened for smart TV devices, and the target TV device list is obtained by querying whether the device type field is a smart TV identifier. A temporary connection channel is established for each device in the target TV device list. On the basis of the temporary connection channel, the service type supported by the device is obtained by sending a service query request, and a service feature set is generated, which includes the core service capability description of each device, such as whether it supports audio transmission or remote control function. On this basis, the service feature set is analyzed for audio and video transmission capabilities, especially to detect whether the device supports A2DP (Advanced Audio Distribution Protocol) and AVRCP (Audio and Video Remote Control Protocol). These two protocols are key standards for audio and video data transmission, and their compatibility directly affects the interaction quality and function support between the device and the headset, thereby generating protocol compatibility data. At the same time, based on the power control parameters in the dynamic optimization parameter set, the signal quality of each target device is measured. The mean and variance of the received signal strength indication (RSSI) are calculated by continuous N sampling to evaluate the stability of the signal. The RSSI mean reflects the average strength of the signal, while the RSSI variance describes the fluctuation of the signal. These two indicators together constitute the signal stability data. In order to improve the evaluation of the device connection quality, the signal stability data is input into the channel quality evaluation unit. The channel quality evaluation unit describes the channel status of the device by calculating the link quality index (LQI) and the packet error rate (PER). Among them, the LQI value is used to measure the overall health of the channel, and the PER value describes the reliability of the channel by counting the error rate in the transmission of statistical packets. These indicators jointly generate channel status data, which provides a quantitative basis for the device connection performance. The protocol compatibility data, signal stability data and channel status data are combined to generate a device feature vector. The device feature vector is a high-dimensional feature representation that integrates the protocol support capabilities, signal quality characteristics and channel health of the target device. The matching score between the device and the smart TV is calculated based on the device feature vector. The weighted summation method is adopted to assign weights to each feature according to its importance and then perform comprehensive summation to generate equipment feature correlation data.

[0035] S5, calculate the optimal connection cost value according to the device feature correlation data, establish a Bluetooth connection when the optimal connection cost value meets the preset threshold condition, and continuously collect connection status data to generate real-time performance evaluation results.

[0036] Among them, the device feature correlation data is normalized, and the differences between different feature dimensions are eliminated by unifying the scale and range of the data to generate standardized correlation data. The standardized correlation data is input into the connection cost calculation function, and the expression of the function is: .in, , , are weight coefficients, which are used to balance the signal quality values. , link quality value and bit error rate value the relative importance of They are the signal benchmark value, link benchmark value, and bit error benchmark value, which are used to standardize the calculation benchmarks of various features. Through this function, the signal strength, link health status, and bit error impact are comprehensively considered to generate connection cost data. The smaller the value of the cost data, the better the connectivity of the current device. The generated connection cost data is thresholded. In order to adapt to dynamically changing environmental conditions, an adaptive threshold function is used. ,in is the baseline threshold, is the volatility coefficient, The introduction of the adaptive threshold function enables the system to adjust the threshold according to time and environmental fluctuations, thereby improving the flexibility of connection judgment. On this basis, it is determined whether Condition, when the cost data is less than the threshold, it means that the device meets the connection requirements. When the connection decision result shows that a Bluetooth connection can be established, a Bluetooth communication channel is constructed based on this result. In the configuration process of the communication channel, the signal gain parameter in the dynamic optimization parameter set is applied to optimize the signal strength to ensure the stability and efficiency of the communication link. At the same time, the communication link data is sampled for performance, and the connection performance data is obtained by measuring the connection availability, service quality and response time every other dynamic sampling period. These data describe the real-time performance of the Bluetooth connection from different dimensions, such as whether the connection is stable, whether the service transmission is smooth, and whether the response time is within an acceptable range. Based on the above collected connection performance data, the performance score calculation is performed. The score calculation is based on multiple performance indicators, and the performance score data is generated by comprehensively analyzing key factors such as connection availability, service quality and response time. In order to ensure the timeliness and dynamics of the performance score data, the performance score is continuously updated according to the newly collected data, and a dynamic performance evaluation is performed in combination with historical data to generate real-time performance evaluation results to promptly reflect changes in the connection status.

[0037] In one example, the parameters of the physical layer, protocol layer, and service layer are collected for the connection between the head-mounted Bluetooth headset and the smart TV to generate an initial collection matrix, including:

[0038] Construct the physical layer, protocol layer and service layer between the Bluetooth headset and the smart TV, and sample the signal transmission strength in the physical layer to obtain signal strength data;

[0039] Calculate the signal strength difference at adjacent time points based on the signal strength data to obtain connection stability data, and monitor the energy consumption during signal transmission based on the connection stability data to obtain power consumption control data;

[0040] The time of data packet transmission in the protocol layer is measured to obtain transmission delay data, and data packet statistics are performed based on the transmission delay data to obtain packet loss data;

[0041] Perform bit error analysis on packet loss data, calculate the ratio of the number of erroneous bits in the received data packet to the total number of bits, obtain bit error data, and perform synchronization analysis on the audio and video data in the service layer to obtain synchronization deviation data;

[0042] The signal strength data, connection stability data, power consumption control data, transmission delay data, packet loss data, bit error data and synchronization deviation data are combined to obtain an initial acquisition matrix;

[0043] The data in the initial acquisition matrix are subjected to time domain feature extraction and frequency domain feature extraction, and a standardized feature matrix is ​​obtained after being processed by a nonlinear signal enhancer.

[0044] In this example, a three-layer connection architecture is built between the Bluetooth headset and the smart TV, in which the physical layer is responsible for basic signal transmission, the protocol layer provides data transmission rules, and the service layer realizes efficient interaction of audio and video data. At the physical layer, the signal transmission strength is sampled through the hardware interface, and the collected data is used Indicates that It's time It is at the moment The sampling is completed through the built-in signal receiving module at a certain sampling rate. Get the signal strength data stream. Based on the collected signal strength data , calculate the difference in signal strength at adjacent time points. The calculation formula for the difference is:

[0045] ;

[0046] in, Indicates the change in signal strength. is the sampling interval. Through this formula, the fluctuation of the signal at different time points is evaluated to obtain the connection stability data . Based on connection stable data , monitor the energy consumption during signal transmission. Energy consumption is measured by transmission power The relationship between the signal strength is calculated as follows:

[0047] ;

[0048] in, is energy consumption, It is at the moment Combined with the stable data connection , estimate power consumption control data In the protocol layer, the transmission delay data is calculated by measuring the transmission time of the data packet. Assumptions and are the timestamps of data packet sending and receiving respectively, then the transmission delay is expressed as:

[0049] ;

[0050] Based on the transmission delay data, data packet statistics are performed to calculate the packet loss data. The calculation formula is:

[0051] ;

[0052] in, is the total number of packets sent, is the total number of packets received. Perform bit error analysis on the packet loss data and calculate the bit error rate The formula for bit error rate is:

[0053] ;

[0054] in, is the number of received error bits, is the total number of bits. At the service layer, the synchronization deviation data is calculated by analyzing the synchronization of audio and video data. The formula is:

[0055] ;

[0056] in, and are the arrival times of the audio and video signals respectively. , connect to stable data , power consumption control data , transmission delay data , Packet loss data , Error data and synchronization deviation data Combine and construct the initial acquisition matrix , which is of the form:

[0057] ;

[0058] For the initial acquisition matrix Perform time domain and frequency domain feature extraction. In the time domain, calculate the sliding average As a time domain feature, the formula is:

[0059] ;

[0060] in, is the size of the sliding window. In the frequency domain, the main frequency component of the signal is extracted by fast Fourier transform to generate spectrum feature data. The time domain and frequency domain features are input into the nonlinear signal enhancer for processing, and the standardized feature matrix is ​​obtained by enhancing the key features of the signal and suppressing noise.

[0061] In one example, the data in the initial acquisition matrix is ​​subjected to time domain feature extraction and frequency domain feature extraction, and a standardized feature matrix is ​​obtained after being processed by a nonlinear signal enhancer, including:

[0062] Perform time domain analysis on the initial acquisition matrix, obtain time domain variation data by calculating the signal strength difference sequence of adjacent time points, and perform trend analysis based on the time domain variation data, and obtain the time domain feature vector by calculating the sliding average of N sampling points;

[0063] Perform fast Fourier transform on the initial acquisition matrix, convert the time domain signal into frequency domain representation to obtain a spectrum data sequence, perform frequency domain feature extraction on the spectrum data sequence, and obtain a frequency domain feature vector by analyzing the energy distribution of the frequency component;

[0064] Inputting the time domain feature vector and the frequency domain feature vector into a nonlinear signal enhancer, obtaining enhanced feature data through an adaptive gain control algorithm, and performing a standardized transformation on the enhanced feature data to obtain standardized feature data;

[0065] A feature mapping relationship is constructed based on the standardized feature data. The feature correlation data is obtained by calculating the correlation coefficient matrix of the time domain and frequency domain features. The feature correlation data is reconstructed into a matrix. The standardized feature matrix is ​​obtained by combining the standardized features of the time domain and frequency domain.

[0066] In this example, the initial acquisition matrix Represented as a set of sampled signals, where each column corresponds to a feature and the number of sampling points is . Assume that a column of signals Expressed as ,in It is In order to obtain the time domain variation characteristics of the signal, the difference sequence of the signal strength at adjacent time points is calculated. , the formula is:

[0067] ;

[0068] in, Indicates the amplitude of the signal change between consecutive time points. This difference sequence reflects the short-term dynamic change trend of the signal. For trend analysis, the sliding average method is used to smooth signal changes and extract long-term trends. The sliding average calculation formula is:

[0069] ;

[0070] in, is the width of the sliding window, It is the average signal value centered at the current time point. The sliding average can remove short-term fluctuations in the signal, highlight the long-term change trend, and generate a time domain feature vector. Perform frequency domain analysis on the initial acquisition matrix. Through fast Fourier transform, the time domain signal Convert to frequency domain representation , the formula is:

[0071] ;

[0072] in, is the first The complex value of the frequency components, is the corresponding frequency, is the total sampling time length of the signal. The result of the fast Fourier transform provides a spectral data sequence of the signal, describing the amplitude and phase of different frequency components. Based on the spectral data sequence , extract frequency domain features and calculate the energy distribution of frequency components. The energy of each frequency component is expressed as:

[0073] ;

[0074] in, is the frequency The signal energy at is the modulus of the corresponding complex amplitude. By normalizing and statistically analyzing the energy distribution, the frequency domain feature vector is extracted , which contains the characteristics of the main frequency component, bandwidth and energy distribution. The extracted time domain feature vector and frequency domain eigenvector The input is processed by the nonlinear signal enhancer. The nonlinear signal enhancer uses an adaptive gain control algorithm to dynamically adjust the gain according to the characteristics of the input data. The gain calculation formula is:

[0075] ;

[0076] in, is the signal gain after enhancement, is the parameter that controls the enhancement rate, is the mean value of the signal. Through this algorithm, weak signals are enhanced while suppressing high-amplitude noise to generate enhanced feature data. The enhanced feature data is standardized to eliminate the difference in data scale. The standardized formula is:

[0077] ;

[0078] in, is the standardized eigenvalue, and are the minimum and maximum values ​​of the input data respectively. Standardized feature data More suitable for subsequent analysis. Based on standardized feature data , construct the mapping relationship between time domain and frequency domain features. By calculating the time domain features and frequency domain characteristics The correlation coefficient matrix , quantifying the degree of correlation between the two, the formula is:

[0079] ;

[0080] in, is the time domain feature and frequency domain characteristics The covariance of and are their standard deviations respectively. Correlation coefficient matrix Describe the interaction between different features. Correlation coefficient matrix Perform matrix reconstruction, combine the standardized features of the time domain and frequency domain into a unified feature matrix, and obtain the standardized feature matrix .

[0081] The time domain feature vector and the frequency domain feature vector are input into a nonlinear signal enhancer, and enhanced feature data are obtained through an adaptive gain control algorithm, and the enhanced feature data are standardized and transformed to obtain standardized feature data, including: feature quantization of the time domain feature vector and the frequency domain feature vector, and obtaining a feature importance vector by calculating the weight coefficient of each feature dimension; dynamic threshold calculation of the feature importance vector, and obtaining a hierarchical gain control parameter by setting threshold boundaries of K different gain levels; nonlinear transformation of the time domain feature vector based on the hierarchical gain control parameter, and adaptive adjustment of the parameters of the sigmoid function to obtain a time domain enhancement. Coefficient; perform product operation on the time domain enhancement coefficient and the time domain feature vector, and obtain enhanced time domain feature data by dynamically adjusting the signal amplitude; perform gain adjustment on the frequency domain feature vector based on the hierarchical gain control parameter, and obtain the frequency domain enhancement coefficient through the frequency band energy redistribution algorithm; perform product operation on the frequency domain enhancement coefficient and the frequency domain feature vector, and obtain enhanced frequency domain feature data through frequency domain energy compensation; fuse the enhanced time domain feature data and the enhanced frequency domain feature data, and obtain enhanced feature data through a weighted combination algorithm; perform zero-mean normalization on the enhanced feature data, and obtain standardized feature data by calculating the feature mean and variance.

[0082] In one example, the standardized feature matrix is ​​input into a parallel signal processor, and the time domain evaluation parameter and the frequency domain evaluation parameter are calculated by the time domain processing unit and the frequency domain processing unit respectively, and the connection quality evaluation data is generated after the time-frequency entropy constraint fusion, including:

[0083] The standardized characteristic matrix is ​​input into the time domain processing unit and the frequency domain processing unit of the parallel signal processor, and a time domain data stream and a frequency domain data stream are obtained through parallel calculation;

[0084] The stability of the time domain data stream is analyzed. The time domain stability index is obtained by calculating the three indicators of topological stability, connectivity availability and failure controllability. The characteristics of the frequency domain data stream are analyzed. The frequency domain characteristic index is obtained by calculating the three indicators of transmission stability, packet loss availability and error controllability.

[0085] Input the time domain stability index into the time domain correlation analysis unit, and obtain the time domain evaluation parameter by calculating the linear correlation coefficient of the time domain data; and input the frequency domain characteristic index into the frequency domain correlation analysis unit, and obtain the frequency domain evaluation parameter by calculating the spectral correlation coefficient of the frequency domain data;

[0086] The entropy values ​​of the time domain evaluation parameters and the frequency domain evaluation parameters are calculated, and the time-frequency entropy constraint value is obtained by information entropy calculation. The time-frequency weight is calculated based on the time-frequency entropy constraint value, and the weight coefficient is solved by the least square method to obtain the fusion weight coefficient.

[0087] The time domain evaluation parameters and the frequency domain evaluation parameters are linearly weighted according to the fusion weight coefficient to obtain the connection quality evaluation data.

[0088] In this example, the feature matrix is ​​normalized Input to a parallel signal processor, the matrix is ​​represented as ,in is the first Line The architecture of the parallel signal processor allows the partitioning of different eigenvectors in the matrix to be calculated simultaneously. Split into time domain feature parts And the frequency domain characteristics . Time domain feature matrix Represents the time series data of the signal, the frequency domain feature matrix It represents the feature representation after frequency domain conversion. The processor generates time domain data streams respectively and frequency domain data stream For time domain data stream Stability analysis is performed to obtain the time domain stability index by calculating the three indicators of topological stability, connectivity availability, and failure controllability. Topological stability is measured by the rate of change of signal connections, and its calculation formula is:

[0089] ;

[0090] in, represents the topological stability, is the total number of sampling points, For the moment The time domain data value of . Connectivity availability measures whether the signal transmission is continuous, and its formula is:

[0091] ;

[0092] in, represents the connectivity availability, is the signal strength threshold, which is used to determine whether the signal is available. Failure controllability reflects the system's ability to recover when the signal is lost or abnormal, and its calculation formula is:

[0093] ;

[0094] in, is the failure controllability, is the maximum allowed interval of the recovery time. , the focus of characteristic analysis is to calculate transmission stability, packet loss availability and error controllability. Transmission stability is evaluated by calculating the concentration of frequency domain signal energy, and its formula is:

[0095] ;

[0096] in, For transmission stability, and are the upper and lower limits of the main frequency band, and is the frequency range of the signal. Packet loss availability is calculated based on the packet loss rate , the formula is:

[0097] ;

[0098] in, is the packet loss availability, Indicates the packet loss rate. Error controllability is measured by bit error rate. The inverse function evaluation of is:

[0099] ;

[0100] in, is the error controllability. The above time domain stability index is input into the time domain correlation analysis unit, and the time domain evaluation parameter is obtained by calculating the linear correlation coefficient of the time domain data. The formula of the linear correlation coefficient is:

[0101] ;

[0102] in, is the time domain correlation coefficient, Cov represents the covariance, is the standard deviation. Similarly, the frequency domain characteristic index is input into the frequency domain correlation analysis unit, and the frequency domain evaluation parameter is obtained by calculating the spectral correlation coefficient of the frequency domain data. The formula of the spectral correlation coefficient is:

[0103] ;

[0104] in, is the frequency domain spectral correlation coefficient, and are two frequency domain signals. The entropy value is calculated for the time domain evaluation parameters and the frequency domain evaluation parameters. The entropy value calculation formula is:

[0105] ;

[0106] in, is the information entropy, is the probability distribution of the evaluation parameters. Based on the entropy value calculation results, the time-frequency entropy constraint value is obtained, and the weight coefficient is solved by the least squares method. The objective function of weight optimization is:

[0107] ;

[0108] in, and are the weights in the time domain and frequency domain, is the target fitting value. According to the obtained fusion weight coefficient, the time domain evaluation parameters and frequency domain evaluation parameters are linearly weighted, and the formula is:

[0109] ;

[0110] in, It is the final connection quality assessment data.

[0111] In one example, a neural network classification operation is performed on the connection quality assessment data to extract target parameters that affect the connection status and generate a dynamic optimization parameter set, including:

[0112] The connection quality assessment data is input into the neural network classification model, which includes an input layer, three hidden layers and an output layer, wherein the number of nodes in the input layer is the same as the dimension of the connection quality assessment data, the first hidden layer contains 128 neurons and uses the ReLU activation function, the second hidden layer contains 64 neurons and uses the ReLU activation function, the third hidden layer contains 32 neurons and uses the ReLU activation function, and the output layer contains 2 neurons and uses the Softmax activation function to obtain the connection status classification result;

[0113] Perform reverse feature importance calculation on the connection status classification results, and obtain the feature importance vector by calculating the contribution of each input feature to the classification result;

[0114] Feature sorting is performed based on the feature importance vector. By sorting the features from high to low according to the importance score, the features with importance scores greater than a preset threshold are selected as target parameters to obtain a target parameter sequence.

[0115] Perform sensitivity analysis on each parameter in the target parameter sequence, and obtain the parameter sensitivity matrix by calculating the impact of parameter changes on the connection status;

[0116] The parameter adjustment interval is calculated based on the parameter sensitivity matrix. By analyzing the effective range of parameter changes, the upper and lower limits of the parameters are set to obtain the parameter constraints. The optimization objective function is constructed based on the parameter constraints. The parameter optimization objective is obtained by minimizing the weighted sum of the connection state classification error and the parameter adjustment amplitude.

[0117] The parameter optimization target is optimized by gradient descent calculation, and the optimal parameter combination is obtained by iteratively updating the parameter value until convergence. The optimal parameter combination is grouped according to the signal gain group, power control group and coding mode group to form a dynamic optimization parameter set.

[0118] In this example, we will connect the quality assessment data Input into the neural network classification model, the number of nodes in the input layer of the neural network is equal to the dimension of the data The classification model consists of three hidden layers. The first hidden layer contains 128 neurons, the second hidden layer contains 64 neurons, and the third hidden layer contains 32 neurons. Each layer uses the ReLU activation function, and its formula is:

[0119] ;

[0120] in is the linear combination value input to the activation function. The ReLU activation function has the advantages of high computational efficiency and the ability to process sparse data, making it easier for the network to capture important features during training. The output layer contains two neurons and uses the Softmax activation function, whose formula is:

[0121] ;

[0122] in Represents the output layer The Softmax function converts the output into the probability value of two categories, which is used to predict the classification result of the connection status. ,in and They represent the probability that the connection state belongs to the first category or the second category respectively. After obtaining the connection state classification result, the reverse feature importance calculation is used to evaluate the contribution of each input feature to the classification result. This is achieved through the gradient back propagation algorithm. The specific formula is:

[0123] ;

[0124] in It is a feature The importance of is the loss function of the classification model, such as the cross entropy loss:

[0125] ;

[0126] in is the predicted class probability value. By calculating each input feature pair The partial derivative of quantifies the contribution of features to the classification results and generates feature importance vectors . Based on the feature importance vector, the features are sorted from high to low. Select the importance score Greater than the preset threshold The characteristics are taken as the target parameters, and the formula is:

[0127] ;

[0128] Screen out key target parameters and form a target parameter sequence ,in For each parameter in the target parameter sequence A sensitivity analysis is performed to quantify the impact of parameter changes on the connection status. The formula for sensitivity analysis is:

[0129] ;

[0130] in Indicates the target parameter For Category By analyzing the changes of all target parameters, a parameter sensitivity matrix is ​​constructed. , whose elements Reflects the contribution of the parameter to the probability value of each category. Based on the parameter sensitivity matrix, the parameter adjustment range is calculated to determine the upper and lower limits of each parameter. By analyzing the effective range of the parameter, for example, by optimizing the following constraint problem:

[0131] ;

[0132] in and The parameters are The minimum and maximum values ​​of is the classification probability of the baseline state, is the allowable classification error range. These upper and lower limits constitute parameter constraints Using the parameter constraints, an optimization objective function is constructed to minimize the weighted sum of the connection state classification error and the parameter adjustment amplitude. The objective function is defined as:

[0133] ;

[0134] in, is the optimization objective function, is the weight of the adjustment amplitude, is a parameter The objective function is optimized by the gradient descent method, and the parameter value is updated using the following iterative formula:

[0135] ;

[0136] in, is the learning rate, is the number of iterations. When the gradient change or the objective function change is less than the preset convergence threshold, the iteration is stopped to obtain the optimal parameter combination. The optimized parameter combinations are classified into signal gain group, power control group and coding mode group according to their functions. For example, signal gain parameter optimization is mainly aimed at signal strength adjustment, power control parameter optimization is used to balance energy consumption and transmission quality, and coding mode parameter optimization is aimed at transmission efficiency and data reliability. These groups form a dynamic optimization parameter set.

[0137] In one example, the dynamically optimized parameter set is input into the fast device scanning module, and the characteristic information of the surrounding Bluetooth devices is obtained by using a two-stage scanning mechanism to generate device characteristic association data, including:

[0138] In the first stage of scanning, the signal gain parameter in the dynamic optimization parameter set is adjusted in strength, and the surrounding device broadcast scan is performed by changing the transmission power to obtain the original scan list containing MAC addresses, signal strength values ​​and device types;

[0139] The original scan list is screened for smart TV devices, and a target TV device list is obtained by querying whether the device type field is a smart TV identifier for matching;

[0140] In the second stage of scanning, a temporary connection channel is established for each device in the target TV device list, and the service type supported by the device is obtained by sending a service query request to obtain a service feature set;

[0141] Analyze the audio and video transmission capabilities of the service feature set, obtain protocol compatibility data by detecting the support of A2DP and AVRCP protocols, measure the signal quality of each device based on the power control parameters in the dynamic optimization parameter set, and calculate the RSSI mean and variance through N consecutive sampling to obtain signal stability data;

[0142] The signal stability data is input into the channel quality assessment unit, and the channel status data is obtained by calculating the link quality index LQI value and the packet error rate PER value, and the protocol compatibility data, signal stability data and channel status data are combined to obtain the device feature vector;

[0143] The matching score with the smart TV is calculated based on the device feature vector, and the device feature association data is obtained through weighted summation.

[0144] In this example, in the first phase of scanning, the signal gain parameter in the dynamic optimization parameter set is used to control the scanning range by adjusting the strength of the Bluetooth transmission power. The adjustment formula is:

[0145] ;

[0146] in, is the initial transmit power, is the dynamically optimized gain adjustment. , expand the scanning range, and ensure more comprehensive discovery of surrounding Bluetooth devices. During the broadcast scanning process, each scanned device will return a MAC address (device unique identifier), signal strength value, (Received Signal Strength Indicator) and information about the device type, which constitutes the original scan list ,in It is The MAC address of the device, is its signal strength value, is the device type. After the original scan list is generated, the device type field Filter to match Smart TV identifiers The filtering rule is expressed by the following expression:

[0147] ;

[0148] Filter the results to form a target TV device list , which only contains smart TV devices. When entering the second stage of scanning, a temporary connection channel is established for each device in the target TV device list. After the channel is established, a service query request is sent , the device will return a set of supported service types These service types include audio transmission protocols (such as A2DP) and remote control protocols (such as AVRCP). By checking whether these protocols are included in the service set, the device's protocol compatibility is evaluated. Protocol compatibility data The calculation formula is:

[0149] ;

[0150] Using the power control parameters in the dynamic optimization parameter set The signal quality of each device is measured. Sub-sampling calculation of received signal strength indicator The mean and variance of are used to evaluate signal stability. The formula is:

[0151] ;

[0152] in, is the mean signal strength, is the variance of the signal strength. The signal stability data is input into the channel quality assessment unit for further analysis. The link quality index (LQl) is calculated as follows:

[0153] ;

[0154] in, and are the minimum and maximum thresholds of signal strength respectively. At the same time, the channel error performance is calculated by statistically analyzing the packet error rate (PER), and the formula is:

[0155] ;

[0156] in, is the number of error packets, is the total number of packets. The channel status data consists of the results of LQI and PER. , Signal stability data Combined with channel status data (LQI, PER) into a device feature vector , which is of the form:

[0157] ;

[0158] Calculate the matching score with smart TV by using the feature vector of each device , using the weighted summation formula:

[0159] ;

[0160] in, It is the weight coefficient of the feature, reflecting the relative importance of each feature to the matching degree. The device feature association data is constructed to guide the Bluetooth headset to prioritize the connection target with the smart TV.

[0161] Before calculating the optimal connection cost value according to the device feature correlation data, it also includes: combining the device feature correlation data and the performance evaluation results, and obtaining the agent network structure by constructing a double-layer agent network model of the physical layer agent unit and the protocol layer agent unit; inputting the signal strength data and stability data in the performance evaluation results into the physical layer agent unit, calculating the physical layer state space transition probability through the deep reinforcement learning algorithm Q(s,a), and obtaining the physical layer optimal agent strategy; constructing the protocol layer state space based on the physical layer optimal agent strategy, calculating the optimization direction of the protocol layer transmission parameters through the policy gradient algorithm π(a|s), and obtaining the protocol layer optimal agent strategy; combining the physical layer optimal agent strategy and the protocol layer optimal agent strategy, and constructing a double-layer reward function R(s,a)=w1×Rphy +w2×Rpro, where Rphy represents the physical layer connection quality reward, and Rpro represents the protocol layer transmission efficiency reward, to obtain the two-layer strategy evaluation parameters; interactively optimize the two-layer strategy evaluation parameters, and obtain the optimized two-layer proxy model by updating the physical layer Q value and the protocol layer policy gradient in parallel; classify the device feature association data according to physical features and protocol features, input the optimized two-layer proxy model, and obtain the two-layer joint decision result by calculating the action space output of the two layers in parallel; perform consistency constraint calculation based on the two-layer joint decision result, optimize the physical layer action aphy and the protocol layer action apro by constructing the minimum conflict objective function min{||aphy-apro||2}, and obtain the final decision parameters, which are used to calculate the optimal connection cost value.

[0162] In one example, the optimal connection cost is calculated based on the device feature association data, a Bluetooth connection is established when the optimal connection cost meets a preset threshold condition, and connection status data is continuously collected to generate real-time performance evaluation results, including:

[0163] Normalizing the equipment feature correlation data to obtain standardized correlation data;

[0164] Input the standardized correlation data into the connection cost calculation function C(t)=w1×(1-exp(-S / S0))+w2×log(1+Q / Q0)+w3×(1-1 / (1+exp(-E / E0))), where w1, w2, w3 are weight coefficients, S is the signal quality value, S0 is the signal reference value, Q is the link quality value, Q0 is the link reference value, E is the bit error rate value, and E0 is the bit error reference value, to obtain the connection cost data;

[0165] The connection cost data is subjected to threshold judgment. By constructing an adaptive threshold function δ(t)=δ0×(1+β×sin(2πt / T)), where δ0 is the reference threshold, β is the fluctuation coefficient, and T is the time period, it is determined whether the condition C(t)<δ(t) is satisfied, and the connection decision result is obtained.

[0166] A Bluetooth communication channel is constructed based on the connection decision result, and the communication link data is obtained by configuring the signal gain parameter in the dynamic optimization parameter set. The communication link data is sampled for performance, and the connection performance data is obtained by measuring the connection availability, service quality and response time at each dynamic sampling period.

[0167] A performance score calculation is performed based on the connection performance data to obtain performance score data, and the performance score data is dynamically updated and performance evaluated to obtain a real-time performance evaluation result.

[0168] In this example, the device feature association data Normalization is performed to ensure that the values ​​of all features fall into a uniform range (such as [0,1]) to avoid the impact of differences in feature scales on subsequent calculations. The normalization formula is:

[0169] ;

[0170] in, Represents the normalized eigenvalues, and are the minimum and maximum values ​​of the original data respectively. After normalization, the standardized correlation data is obtained . Input the standardized correlation data into the connection cost calculation function , which is used to comprehensively evaluate the connection cost of each device. The formula for the connection cost calculation function is:

[0171] ;

[0172] in, is the connection cost; , , are weight coefficients, representing the signal quality , Link Quality and bit error rate The relative importance of connection cost; is the signal quality value, is the signal quality benchmark value; is the link quality value, is the link quality benchmark value; is the bit error rate value, is the bit error rate benchmark value. After calculating the connection cost, the threshold is determined. In order to dynamically adjust the connection standard, an adaptive threshold function is constructed. , the formula is:

[0173] ;

[0174] in, is the baseline threshold; is the fluctuation coefficient, which is used to control the fluctuation amplitude of the threshold; is the time period. The conditions are used to determine whether to allow the device to establish a Bluetooth connection. If the conditions are met, a connection decision result is generated to approve the establishment of a communication channel. Based on the connection decision result, the signal gain parameter in the dynamic optimization parameter set is used to Configure the communication channel and obtain the communication link data. During the communication process, Collect connection performance data, including connection availability , Service Quality and response time These data are calculated by the following formulas:

[0175] ;

[0176] ;

[0177] ;

[0178] in, is the packet loss rate. Based on the collected performance data, a performance score is calculated to comprehensively evaluate the quality of the connection status. The performance score formula is:

[0179] ;

[0180] in, , , are the weights corresponding to connection availability, service quality, and response time, respectively. The final performance score data is used to dynamically update the system status and generate performance evaluation results in combination with real-time data.

[0181] Reference Figure 2 This embodiment provides a system for quickly connecting a head-mounted Bluetooth headset to a smart TV, including:

[0182] Collection module 1, used for collecting parameters of the physical layer, protocol layer and service layer of the connection between the head-mounted Bluetooth headset and the smart TV, and generating a standardized feature matrix;

[0183] Calculation module 2, used for inputting the standardized feature matrix into the parallel signal processor, calculating the time domain evaluation parameter and the frequency domain evaluation parameter through the time domain processing unit and the frequency domain processing unit respectively, and generating the connection quality evaluation data after the time-frequency entropy constraint fusion;

[0184] Extraction module 3, used to perform neural network classification operation on the connection quality assessment data, extract target parameters that affect the connection status, and generate a dynamic optimization parameter set;

[0185] Scanning module 4, used to input the dynamic optimization parameter set into the fast device scanning module, adopt a two-stage scanning mechanism to obtain the characteristic information of the surrounding Bluetooth devices, and generate device characteristic association data;

[0186] The connection module 5 is used to calculate the optimal connection cost value according to the device feature correlation data, establish a Bluetooth connection when the optimal connection cost value meets the preset threshold condition, and continuously collect connection status data to generate real-time performance evaluation results.

[0187] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to the above method embodiment, which will not be repeated here.

[0188] Reference Figure 3 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0189] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0190] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0191] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0192] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0193] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for quickly connecting a Bluetooth headset to a smart TV, characterized in that: The following steps are involved: Collect parameters of the physical layer, protocol layer, and service layer of the connection between the Bluetooth headset and the smart TV to generate a standardized feature matrix; Specifically, it includes: constructing a physical layer, a protocol layer and a service layer between a head-mounted Bluetooth headset and a smart TV, and sampling the signal transmission strength in the physical layer to obtain signal strength data; calculating the difference in signal strengths at adjacent time points based on the signal strength data to obtain connection stability data, and monitoring the energy consumption in the signal transmission process based on the connection stability data to obtain power consumption control data; measuring the time of data packet transmission in the protocol layer to obtain transmission delay data, and performing data packet statistics based on the transmission delay data to obtain packet loss data; performing bit error analysis on the packet loss data, calculating the ratio of the number of erroneous bits in the received data packet to the total number of bits to obtain bit error data, and performing synchronization analysis on the audio and video data in the service layer to obtain synchronization deviation data; combining the signal strength data, the connection stability data, the power consumption control data, the transmission delay data, the packet loss data, the bit error data and the synchronization deviation data to obtain an initial acquisition matrix; performing time domain feature extraction and frequency domain feature extraction on the data in the initial acquisition matrix, and obtaining a standardized feature matrix after processing by a nonlinear signal enhancer; The standardized feature matrix is ​​input into a parallel signal processor, and a time domain evaluation parameter and a frequency domain evaluation parameter are calculated by a time domain processing unit and a frequency domain processing unit respectively, and connection quality evaluation data is generated after fusion through time-frequency entropy constraints; Performing a neural network classification operation on the connection quality assessment data, extracting target parameters that affect the connection state, and generating a dynamic optimization parameter set; Input the dynamic optimization parameter set into the fast device scanning module, adopt a two-stage scanning mechanism to obtain the characteristic information of the surrounding Bluetooth devices, and generate device characteristic association data; The optimal connection cost value is calculated according to the device feature association data, a Bluetooth connection is established when the optimal connection cost value meets a preset threshold condition, and connection status data is continuously collected to generate a real-time performance evaluation result.

2. The method for quickly connecting a Bluetooth headset to a smart TV according to claim 1, characterized in that: The step of extracting time domain features and frequency domain features from the data in the initial acquisition matrix and obtaining a standardized feature matrix after processing by a nonlinear signal enhancer includes: Performing time domain analysis on the initial acquisition matrix, obtaining time domain variation data by calculating a sequence of signal strength differences between adjacent time points, and performing trend analysis based on the time domain variation data, obtaining a time domain feature vector by calculating a sliding average of N sampling points; Performing a fast Fourier transform on the initial acquisition matrix, converting the time domain signal into a frequency domain representation to obtain a spectrum data sequence, performing frequency domain feature extraction on the spectrum data sequence, and obtaining a frequency domain feature vector by analyzing the energy distribution of the frequency component; Inputting the time domain feature vector and the frequency domain feature vector into a nonlinear signal enhancer, obtaining enhanced feature data through an adaptive gain control algorithm, and performing a standardized conversion on the enhanced feature data to obtain standardized feature data; A feature mapping relationship is constructed based on the standardized feature data, feature correlation data is obtained by calculating the correlation coefficient matrix of time domain and frequency domain features, and the feature correlation data is matrix reconstructed to obtain a standardized feature matrix by combining the standardized features of time domain and frequency domain.

3. The method for quickly connecting a Bluetooth headset to a smart TV according to claim 2, characterized in that: The step of inputting the standardized feature matrix into a parallel signal processor, calculating a time domain evaluation parameter and a frequency domain evaluation parameter through a time domain processing unit and a frequency domain processing unit respectively, and generating connection quality evaluation data after fusion through time-frequency entropy constraints, comprises: Inputting the standardized characteristic matrix into a time domain processing unit and a frequency domain processing unit of a parallel signal processor, and obtaining a time domain data stream and a frequency domain data stream through parallel calculation; Performing stability analysis on the time domain data stream, obtaining a time domain stability index by calculating three indexes: topology stability, connectivity availability, and failure controllability, and performing characteristic analysis on the frequency domain data stream, obtaining a frequency domain characteristic index by calculating three indexes: transmission stability, packet loss availability, and error controllability; Input the time domain stability index into the time domain correlation analysis unit, and obtain the time domain evaluation parameter by calculating the linear correlation coefficient of the time domain data; and input the frequency domain characteristic index into the frequency domain correlation analysis unit, and obtain the frequency domain evaluation parameter by calculating the spectral correlation coefficient of the frequency domain data; Performing entropy calculation on the time domain evaluation parameter and the frequency domain evaluation parameter, obtaining a time-frequency entropy constraint value through information entropy calculation, performing time-frequency weight calculation based on the time-frequency entropy constraint value, solving the weight coefficient through the least square method, and obtaining a fusion weight coefficient; The time domain evaluation parameter and the frequency domain evaluation parameter are linearly weighted according to the fusion weight coefficient to obtain connection quality evaluation data.

4. The method for quickly connecting a Bluetooth headset to a smart TV according to claim 3, characterized in that: The performing of neural network classification operation on the connection quality assessment data, extracting target parameters affecting the connection state, and generating a dynamic optimization parameter set includes: Input the connection quality assessment data into a neural network classification model, the neural network classification model includes an input layer, three hidden layers and an output layer, wherein the number of nodes in the input layer is the same as the dimension of the connection quality assessment data, the first hidden layer contains 128 neurons and uses a ReLU activation function, the second hidden layer contains 64 neurons and uses a ReLU activation function, the third hidden layer contains 32 neurons and uses a ReLU activation function, and the output layer contains 2 neurons and uses a Softmax activation function, to obtain a connection state classification result; Performing reverse feature importance calculation on the connection state classification result, and obtaining a feature importance vector by calculating the contribution of each input feature to the classification result; Sorting features based on the feature importance vector, sorting the features from high to low according to the importance scores, selecting features with importance scores greater than a preset threshold as target parameters, and obtaining a target parameter sequence; Performing sensitivity analysis on each parameter in the target parameter sequence, and obtaining a parameter sensitivity matrix by calculating the influence of parameter changes on the connection state; Calculating the parameter adjustment interval based on the parameter sensitivity matrix, setting the parameter upper limit and the parameter lower limit by analyzing the effective range of the parameter change, obtaining the parameter constraint condition, and constructing the optimization objective function according to the parameter constraint condition, and obtaining the parameter optimization target by minimizing the weighted sum of the connection state classification error and the parameter adjustment amplitude; The parameter optimization target is subjected to gradient descent optimization calculation, and the optimal parameter combination is obtained by iteratively updating the parameter value until convergence, and the optimal parameter combination is grouped according to the signal gain group, the power control group and the coding mode group to form a dynamic optimization parameter set.

5. The method for quickly connecting a Bluetooth headset to a smart TV according to claim 4, characterized in that: The step of inputting the dynamic optimization parameter set into a fast device scanning module, using a dual-stage scanning mechanism to obtain feature information of surrounding Bluetooth devices, and generating device feature association data includes: In the first stage of scanning, the signal gain parameter in the dynamic optimization parameter set is strength-adjusted, and a broadcast scan of surrounding devices is performed by changing the transmission power to obtain an original scan list including MAC addresses, signal strength values, and device types; The original scan list is screened for smart TV devices, and a target TV device list is obtained by querying whether the device type field is a smart TV identifier for matching; In the second stage of scanning, a temporary connection channel is established for each device in the target TV device list, and a service type supported by the device is obtained by sending a service query request to obtain a service feature set; Performing an audio and video transmission capability analysis on the service feature set, obtaining protocol compatibility data by detecting the support of A2DP and AVRCP protocols, and performing signal quality measurement on each device based on the power control parameters in the dynamic optimization parameter set, calculating the RSSI mean and variance through continuous N sampling, and obtaining signal stability data; Input the signal stability data into a channel quality assessment unit, obtain channel status data by calculating a link quality index LQI value and a packet error rate PER value, and perform feature combination on the protocol compatibility data, the signal stability data and the channel status data to obtain a device feature vector; The matching score with the smart TV is calculated based on the device feature vector, and the device feature association data is obtained by weighted summation.

6. The method for quickly connecting a Bluetooth headset to a smart TV according to claim 5, characterized in that: The calculating the optimal connection cost value according to the device feature association data, establishing a Bluetooth connection when the optimal connection cost value meets a preset threshold condition, and continuously collecting connection status data to generate a real-time performance evaluation result, includes: Normalizing the device feature correlation data to obtain standardized correlation data; Input the standardized correlation data into the connection cost calculation function C(t)=w1×(1-exp(-S / S0))+w2×log(1+Q / Q0)+w3×(1-1 / (1+exp(-E / E0))), where w1, w2, and w3 are weight coefficients, S is the signal quality value, S0 is the signal reference value, Q is the link quality value, Q0 is the link reference value, E is the bit error rate value, and E0 is the bit error reference value, to obtain the connection cost data; Perform threshold determination on the connection cost data, and construct an adaptive threshold function δ(t)=δ0×(1+β×sin(2πt / T)), where δ0 is the reference threshold, β is the fluctuation coefficient, and T is the time period, to determine whether the condition C(t)<δ(t) is satisfied, and obtain a connection decision result; Building a Bluetooth communication channel based on the connection decision result, obtaining communication link data by configuring the signal gain parameter in the dynamic optimization parameter set, and performing performance sampling on the communication link data, and obtaining connection performance data by measuring connection availability, service quality and response time at each dynamic sampling period; A performance score calculation is performed based on the connection performance data to obtain performance score data, and the performance score data is dynamically updated and performance evaluated to obtain a real-time performance evaluation result.

7. A quick connection system between a head-mounted Bluetooth headset and a smart TV, characterized in that: The method for quickly connecting a Bluetooth headset to a smart TV according to any one of claims 1 to 6 is used to implement the steps, wherein the Bluetooth headset and the smart TV quick connection system comprises: The acquisition module is used to collect parameters of the physical layer, protocol layer and service layer of the connection between the head-mounted Bluetooth headset and the smart TV, and generate a standardized feature matrix; A calculation module, used for inputting the standardized feature matrix into a parallel signal processor, calculating the time domain evaluation parameter and the frequency domain evaluation parameter through a time domain processing unit and a frequency domain processing unit respectively, and generating connection quality evaluation data after fusion through time-frequency entropy constraints; An extraction module, used to perform a neural network classification operation on the connection quality assessment data, extract target parameters that affect the connection status, and generate a dynamic optimization parameter set; A scanning module, used to input the dynamic optimization parameter set into a fast device scanning module, adopt a two-stage scanning mechanism to obtain characteristic information of surrounding Bluetooth devices, and generate device characteristic association data; The connection module is used to calculate the optimal connection cost value according to the device feature association data, establish a Bluetooth connection when the optimal connection cost value meets a preset threshold condition, and continuously collect connection status data to generate real-time performance evaluation results.

8. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method for quickly connecting a head-mounted Bluetooth headset to a smart TV as described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for quickly connecting a head-mounted Bluetooth headset to a smart TV according to any one of claims 1 to 6 are implemented.

Images