Bluetooth device rapid pairing method and system based on multiple scenes

Abstract

The invention discloses a Bluetooth device rapid pairing method and system based on multiple scenes. The method comprises the steps that scene features, device historical pairing, demand features, an efficiency set and signal intensity data are acquired; performing time sequence sampling on the signal intensity, calculating a stable distance value in combination with the correction model, and weighting historical data by using a time decay function to generate a historical weight score; scene correlation scores are obtained through scene model matching, demand features are mapped into demand vectors, an efficiency set is subjected to standardization processing, and function matching scores are calculated through cosine similarity. And performing weighted summation and normalization processing by integrating the stable distance, the historical weight, the scene correlation and the function matching to obtain an equipment priority score. And finally, carrying out intelligent grouping and conflict resolution on the equipment library based on the scores, generating a dynamic pairing strategy and executing connection. According to the method, rapid pairing of the Bluetooth equipment can be realized.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of Bluetooth pairing, in particular to a Bluetooth device fast pairing method and system based on multiple scenes. BACKGROUND

[0002] At present, as a core support of short-distance wireless communication, Bluetooth technology occupies a key position in intelligent device interconnection, and its fast, stable and convenient pairing capability directly affects the user experience and the collaborative efficiency of the device ecology. Whether it is smart home, mobile terminal or vehicle-mounted system, the intelligentization and high efficiency of Bluetooth pairing have become an important driving force to improve the interactive experience.

[0003] In an existing technical method, after a device starts Bluetooth, a basic identifier and signal strength are broadcasted, a user needs to manually select a target device name in a list to initiate a pairing request, and the system only screens connectable devices according to a single signal strength threshold; in the pairing stage, static PIN code or one-key confirmation mode is adopted to complete identity verification and generate a fixed link key, and in subsequent reconnection, the stored key is directly called to establish connection. In the traditional Bluetooth pairing process, manual selection and single signal strength judgment are relied on, and the adaptability to complex scenes is lacked, and in the device selection, user habits and device characteristics are often ignored, resulting in low pairing efficiency, especially when functionally similar devices coexist, the user needs to spend extra time to confirm the target device.

[0004] In summary, in the prior art, the Bluetooth device pairing process lacks intelligent adaptive capability in a multi-device and multi-scene environment, resulting in the problem of low pairing efficiency. SUMMARY

[0005] The application provides a Bluetooth device fast pairing method and system based on multiple scenes to solve the problem of low pairing efficiency in a multi-device and multi-scene environment.

[0006] In a first aspect, to solve the above technical problems, the application provides a Bluetooth device fast pairing method and system based on multiple scenes, comprising: obtaining current scene characteristic data, historical pairing data of a device, demand characteristic data, an efficiency characteristic set and signal strength data; sampling the signal strength data to obtain a time series data set, and combining a preset multi-factor signal strength correction model to obtain a stable distance value of the device; weighting the historical pairing data through a preset time decay function, and performing weight distribution to obtain a historical weight score; matching a predefined scene model according to the scene characteristic data to obtain a scene correlation score; According to the demand feature data, a first demand vector is obtained by weighted summation combined with a preset function mapping table, the performance feature set is normalized to obtain a first function vector; A cosine similarity between the first demand vector and the first function vector is calculated, and a function matching score is obtained by mapping; According to the stable distance value, the historical use weight score, the scene correlation score and the function matching score, a comprehensive priority score of each device is calculated by weighted summation, and a final priority score is obtained by normalization processing; According to the final priority score, a preset device database is intelligently grouped and conflict is resolved to obtain a target device, and a dynamic pairing strategy is generated according to the target device and connection is performed.

[0007] In an optional implementation, the signal strength data is sampled to obtain a time series data set, and a stable distance value of the device is obtained by combining a preset multi-factor signal strength correction model, including: Obtain environmental noise level data; The signal strength data is sampled to generate a time series data set containing signal strength values at multiple time points; If the noise level in the environmental noise data set exceeds a preset threshold, the sampling frequency is increased to regenerate the time series data set; The time series data set is processed by a Kalman filter algorithm to obtain a filtered signal strength data set; The filtered signal strength data set is input into a preset multi-factor signal strength correction model to obtain a stable distance value of the device.

[0008] In an optional implementation, the historical pairing data is weighted by a preset time decay function, and a historical weight score is obtained by weight distribution, including: The historical pairing data is weighted by a preset time decay function to obtain a spatiotemporal feature data set, and the preset time decay function calculation formula is: Wherein, is a time decay weight, is the current time, is the recording time, is a decay coefficient; The spatiotemporal feature data set is linearly weighted to calculate the historical weight score of each device.

[0009] In an optional implementation, the first demand vector is obtained by weighted summation combined with a preset function mapping table according to the demand feature data, including: preprocessing the demand feature data and quantifying the functional demand to obtain a functional demand degree; mapping the functional demand degree in combination with a preset functional mapping table to obtain a demand weight; According to the demand weight and the demand feature data, a first demand vector is calculated by weighted summation.

[0010] In an optional implementation, the cosine similarity between the first demand vector and the first function vector is calculated, and a function matching score is obtained by mapping, comprising: The cosine similarity between the first demand vector and the first function vector is calculated according to the following formula: wherein, is the cosine similarity between the first demand vector and the first function vector, is the first demand vector, is the first function vector; The cosine similarity is mapped into the interval [0, 100] by linear mapping method to obtain the function matching score.

[0011] In an optional implementation, the comprehensive priority score of each device is calculated by weighted summation according to the stable distance value, the historical use weight score, the scene correlation score and the function matching score, comprising: The comprehensive priority score is calculated by the following formula: wherein, is the comprehensive priority score of the device, is the stable distance value, is the basic distance value, is the historical use weight score, is the scene correlation score, is the function matching score, , , , is a preset weight coefficient, satisfying .

[0012] In an optional implementation, the target device is obtained by intelligently grouping and conflict resolution of the preset device database according to the final priority score, comprising: obtaining device function description features and unique features from a preset device database; According to the device function description features, the devices are clustered and analyzed to obtain device grouping; obtaining priority values of devices in a group, when the number of devices with the highest priority value is greater than one and the priority difference is less than a preset difference threshold, extracting historical pairing records of the devices, and calculating a stability analysis result; According to the stability analysis result and the uniqueness feature, a weighted sum is obtained to obtain a uniqueness feature score of the device.

[0013] When the uniqueness feature score of the device is the highest, the device is determined as the target device.

[0014] In an optional embodiment, the generating a dynamic pairing strategy according to the target device and performing connection includes: obtaining a priority score of the target device from a preset device database; determining a connection timeout threshold and a number of retries according to the priority score to obtain a dynamic pairing strategy; According to the dynamic pairing strategy, a connection request is generated using a Bluetooth protocol stack to initiate an active connection to the target device.

[0015] In an optional embodiment, after the generating a dynamic pairing strategy according to the target device and performing connection, includes: Recording the pairing results, connection quality and usage time data of the connection; Using a time series database to index by device ID and timestamp to form a structured historical record table; Real-time monitoring of the total amount of data in the historical usage record; When the amount of data in the historical usage record table exceeds a preset data threshold, the device feature vector in the preset device database is updated through incremental learning, the weight coefficient is adjusted, the direction of weight update is determined, and the optimized pairing decision parameters are obtained.

[0016] In a second aspect, the application provides a Bluetooth device fast pairing system based on multiple scenes, which includes: A data acquisition module acquires current scene characteristic data, historical pairing data of the device, demand feature data, an efficiency feature set, and signal strength data; A stable distance module samples the signal strength data to obtain a time series data set, and combines a preset multi-factor signal strength correction model to obtain a stable distance value of the device; A weight score module weights the historical pairing data through a preset time decay function and performs weight distribution to obtain a historical weight score; A scene matching module matches a predefined scene model according to the scene characteristic data to obtain a scene correlation score; a function vector calculation module, configured to obtain a first demand vector by weighted summation according to the demand characteristic data and a preset function mapping table, and to obtain a first function vector by standardizing the performance characteristic set; a function matching module, configured to calculate a cosine similarity between the first demand vector and the first function vector, and to obtain a function matching score by mapping; a priority calculation module, configured to calculate a comprehensive priority score of each device by weighted summation according to the stable distance value, the historical use weight score, the scene correlation score and the function matching score, and to obtain a final priority score by normalization processing; a pairing connection module, configured to obtain a target device by intelligent grouping and conflict resolution of a preset device database according to the final priority score, and to generate a dynamic pairing strategy and perform connection according to the target device.

[0017] In a third aspect, the present application further provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the current application when executing the computer program.

[0018] In a fourth aspect, the present application further provides a computer readable storage medium, comprising a stored computer program, wherein the computer readable storage medium controls the device where the computer readable storage medium is located to implement the current application when the computer program is running.

[0019] Compared with the prior art, the present application has the following beneficial effects: (1) The present application realizes full-dimensional perception of device state by simultaneously obtaining scene characteristics, historical pairing records, demand characteristics, performance characteristics and signal strength data, avoiding the problem of insufficient scene adaptability caused by single signal strength or manual selection in traditional methods.

[0020] (2) The present application accurately calculates the stable distance value of the device based on time series signal sampling and multi-factor correction model (such as environmental obstruction compensation and dynamic attenuation compensation), effectively reducing the influence of signal jitter on connection stability.

[0021] (3) The present application uses a time attenuation function to weight the historical pairing data, strengthens the weight distribution of high-frequency use devices, realizes adaptive learning of user behavior habits, and reduces the probability of mispairing when multiple devices coexist.

[0022] (4) The application significantly improves the matching degree of device recommendation and the current use scene by matching real-time scene characteristics with pre-defined scene models (such as home / office / public places), and solves the problem of response lag during cross-scene switching.

[0023] (5) The application maps the demand characteristics into a demand vector, and combines the standardized performance characteristics to perform cosine similarity calculation, which can effectively distinguish the core differences of similar function devices (such as audio device sound quality priority and delay sensitivity), and avoid the burden of manual screening by the user.

[0024] (6) The application breaks through the limitations of traditional signal strength or name sorting by using a weighted scoring mechanism that comprehensively considers stable distance, historical weight, scene relevance and function matching, and forms an objective and quantitative device priority ranking.

[0025] (7) Based on the normalized priority score, the device library is grouped and optimized (such as primary and backup device groups, mutually exclusive device groups), which automatically avoids channel occupation conflicts and function repeated connection problems, and realizes reliable pairing decision in complex environment. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 is a flowchart of a multi-scene-based Bluetooth device fast pairing method provided by the first embodiment of the application; Figure 2 is a system structure diagram of a multi-scene-based Bluetooth device fast pairing system provided by the second embodiment of the application. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0028] With reference to Figure 1 , the first embodiment of the application provides a multi-scene-based Bluetooth device fast pairing method, which includes the following steps: S11, obtaining current scene characteristic data, historical pairing data of the device, demand characteristic data, a performance characteristic set and signal strength data; S12, sampling the signal strength data to obtain a time series data set, and combining a pre-set multi-factor signal strength correction model to obtain a stable distance value of the device; S13, weighting the historical pairing data by a pre-set time decay function, and performing weight distribution to obtain a historical weight score; S14, match a predefined scene model according to the scene characteristic data to obtain a scene correlation score; S15, obtain a first demand vector by weighted summation according to the demand characteristic data and a preset function mapping table, and obtain a first function vector by standardizing the performance characteristic set; S16, calculate a cosine similarity between the first demand vector and the first function vector, and perform mapping to obtain a function matching score; S17, calculate a comprehensive priority score of each device by weighted summation according to the stable distance value, the historical use weight score, the scene correlation score and the function matching score, and perform normalization processing to obtain a final priority score; S18, intelligently group and conflict-resolve a preset device database according to the final priority score to obtain a target device, generate a dynamic pairing strategy according to the target device, and perform connection.

[0029] In step S11, current scene characteristic data, historical pairing data of the device, demand characteristic data, performance characteristic set and signal strength data are obtained.

[0030] It should be noted that the scene feature data includes sensor data, geographic location, time period and Wi-Fi network information, the corresponding information is obtained through the sensors of the device itself, the sensor data usually includes the output of accelerometer, gyroscope, etc., the geographic location is obtained through GPS to get the latitude and longitude, the time period records the time range of the use of the device, and the Wi-Fi network information contains the connected SSID and signal strength. The historical pairing data of the device is obtained from the historical database, the original pairing data set containing device identification, timestamp, place coordinates and use frequency is extracted through the database query module, the records are read in segments in a batch processing manner to obtain the original pairing data set, data preprocessing is performed on the original pairing data set, the place coordinates are denoised through data cleaning, if the place coordinates deviation exceeds the preset threshold, the mean filtering algorithm is used to generate the clean place data set, and the historical pairing data is obtained. For example, the coordinates recorded by a device in succession are (100, 200), (102, 201) and (150, 250). If the preset deviation threshold is 10 meters, the coordinate (150, 250) is determined to be abnormal. The mean filtering algorithm takes the average of the first two coordinates to generate the clean coordinate (101, 200.5). This method can smooth the abnormal points and generate a reliable clean place data set. The demand feature data includes application type, application function demand and device historical use scene data, the application type refers to social, game, office and the like, the function demand includes whether high graphic rendering or network connection is needed, and the historical use scene data records the use mode of the device at a specific time and place. For example, a mobile phone records that the recently started application program is social, the currently active application needs real-time video stream, and the historical data shows that the user often uses social application at home at 19:00. These data are collected through the API of the user device to provide input for data fusion. After obtaining the application type, application function demand and device historical use scene data, data fusion processing is performed to integrate the application type, function demand and historical use scene to generate structured demand feature data. The fusion process needs to ensure data time consistency, for example, the application startup record within 1 minute is aligned with the historical scene data to generate a data set containing application type, function demand and scene feature. The performance feature set is obtained from the device to obtain performance parameters and function set, such as processor speed, memory size and supported 5G connection. The standardized processing normalizes the performance parameters to the range of 0 to 1, for example, the processor speed of 2GHz is mapped to 0.8, and the memory of 8GB is mapped to 0.7. The signal strength data is usually represented in the form of received signal strength indication (RSSI) with the unit of decibel-milliwatt (dBm), which reflects the quality of signal transmission between devices. The signal strength data is obtained by periodically receiving the RSSI value through the Bluetooth adapter.

[0031] In step S12, the signal strength data is sampled to obtain a time series data set, and a preset multi-factor signal strength correction model is combined to obtain a stable distance value of the device.

[0032] In an embodiment, the sampling of the signal strength data to obtain a time series data set, and the combination of the preset multi-factor signal strength correction model to obtain a stable distance value of the device, comprises: obtaining environmental noise level data; sampling the signal strength data to generate a time series data set containing signal strength values at multiple time points; if the noise level in the environmental noise data set exceeds a preset threshold, increasing the sampling frequency to regenerate the time series data set; processing the time series data set using a Kalman filter algorithm to obtain a filtered signal strength data set; inputting the filtered signal strength data set into a preset multi-factor signal strength correction model to obtain a stable distance value of the device.

[0033] It is worth noting that while collecting signal strength data, noise parameters caused by surrounding electromagnetic interference and physical obstacles are recorded to obtain environmental noise level data. If the noise level in the environmental noise data set exceeds a preset threshold, the number of samplings is increased, and the threshold judgment module controls the data acquisition module to increase the sampling frequency to generate the time series data set containing signal strength values at multiple time points. For example, for a scenario where the environmental noise level exceeds the preset threshold, assuming the threshold is -80 dBm, if the noise value reaches -75 dBm, the threshold judgment module is triggered to increase the number of samplings. The time series data set is processed using a Kalman filter algorithm, and noise is eliminated through the filter algorithm combined with device transmit power and signal attenuation mode parameters to obtain the filtered signal strength data set. Finally, the filtered signal strength value is input into a multi-factor signal strength correction model, which considers the differences in transmit power (e.g., device A is 20 dBm and device B is 18 dBm), environmental attenuation mode (e.g., walls cause an increase in signal attenuation coefficient), and calibration parameters to calculate the actual stable distance of the device. For example, the corrected RSSI value -63 dBm may correspond to an actual distance of 5 meters in an office environment, while the original value without correction may result in a distance estimation error of up to 2 meters due to noise interference. This process significantly improves the accuracy of distance measurement by dynamically adjusting the sampling strategy and filter processing, especially in a multi-device dense or complex electromagnetic environment.

[0034] In step S13, the historical pairing data is weighted by a preset time decay function, and a weight score is obtained by weight distribution.

[0035] In an embodiment, the historical pairing data is weighted by a preset time decay function, and a weight distribution is performed to obtain a historical weight score, comprising: The historical pairing data is weighted by a preset time decay function to obtain a spatiotemporal feature dataset, and the preset time decay function calculation formula is: Among them, is the time decay weight, is the current time, is the record time, is the decay coefficient; The spatiotemporal feature dataset is linearly weighted to calculate the historical weight score of each device.

[0036] It should be noted that first, the system extracts the pairing records of the device from the historical database, including device identification, timestamp, usage frequency, and location distribution, etc. For example, the historical record of a certain Bluetooth headset shows that it has been paired 50 times in the past year, and the last use time is December 1, 2023. The system uses a time decay function , wherein is the current time (e.g. December 31, 2023), is the record time (e.g. December 1, 2023), and the time difference is 30 days. The decay coefficient is the inverse of the time unit, for example, set to 0.05 (1 / day), and the time decay weight is calculated as , indicating that the weight of this record is attenuated compared to recent records, but still retains some importance. If the of another record is June 1, 2023 (time difference 180 days), its weight is , almost negligible, reflecting the weakening effect of the time decay function on distant records. Next, the system linearly weights the spatiotemporal feature dataset. For example, the spatiotemporal features of a certain device include usage frequency (10 times per week), recent use time (30 days ago), time decay weight (0.223), and location distribution (90% concentrated in home scenarios). The system calculates the historical weight score according to the preset weight distribution proportion, such as usage frequency accounting for 60%, recent use time accounting for 20%, time decay weight accounting for 10%, and location consistency accounting for 10%. The specific calculation is: This score reflects the high usage frequency and recent activity of the device, although the time decay weight is low, but the overall other features still give a high priority. If another device has a usage frequency of 5 times per week, but the recent use time is 3 days ago , its score can be higher, reflecting the reinforcement of the time decay function on recent records. Through such multi-dimensional weighting, the system can dynamically quantify the historical use value of the device, providing accurate basis for subsequent intelligent pairing.

[0037] In step S14, the scene relevance score is obtained by matching the scene feature data with predefined scene models.

[0038] It is worth noting that the process of matching the scene feature data with predefined scene models to generate the scene relevance score is as follows: First, the system obtains the user's current environmental context information through device sensors and network interfaces, including activity status (such as stationary, walking, driving), geographic location (GPS coordinates), time period (such as 10 am), and surrounding Wi-Fi network SSID and signal strength. For example, the user is at home at 8 pm, the phone accelerometer shows a stationary state (acceleration change less than 0.2 m / s 2 ), the GPS coordinates are 39.9° N, 116.3° E, the connected Wi-Fi network is "Home-WiFi" and the signal strength is -50 dBm. These heterogeneous data are integrated into a structured environmental feature set through data fusion technology, such as aligning acceleration data, geographic location, timestamp and Wi-Fi information to generate input data in a unified format.

[0039] Next, the system matches the environmental feature set with predefined scene models (such as home, office, car, outdoor). Each scene model contains a typical feature combination, for example, a home scene is usually associated with a specific Wi-Fi SSID, a stationary state, and a late-night time period; an office scene may correspond to a weekday daytime time period, an office area GPS coordinate, and a corporate Wi-Fi network. The system uses a decision tree algorithm to classify the preliminary scene. For example, if the user's current Wi-Fi SSID is "Home-WiFi" and the time period is 20:00, the decision tree will preferentially match the home scene, giving a preliminary classification confidence of 0.8; if the geographic location is also detected to match the home coordinates, the confidence may be increased to 0.9.

[0040] Subsequently, the system calculates a final scene relevance score by weighting the preliminary classification result, time period, and environmental features. For example, the preliminary classification confidence of a home scene is 0.8, the time period weight is 0.3 (because the probability of home activity is high at night), and the Wi-Fi signal strength weight is 0.2 (a strong signal indicates stable connection). Assuming the weight distribution ratio is 50% for preliminary classification, 30% for time period, and 20% for environmental features, the weighted calculation is: 0.8x0.5+0.3x0.3+0.2x0.2=0.4+0.09+0.04=0.53. After normalization, the relevance score of the home scene can be 53 points (assuming a total score of 100), which is significantly higher than other scenes (e.g., an office may only score 20 points). This scoring mechanism improves accuracy through multi-dimensional feature verification. For example, when the user's GPS shows that they are in a car and the accelerometer detects high-speed motion, even if the home Wi-Fi is connected, the system may lower the home score due to scene conflict and increase the weight of the in-car scene. Ultimately, the scene with the highest relevance score will guide the device pairing strategy, such as connecting smart lights and sound boxes preferentially in a home scene, and automatically pairing car Bluetooth devices in a car scene.

[0041] In step S15, a first demand vector is obtained by weighted summation according to the demand feature data combined with a preset function mapping table, and a first function vector is obtained by standardizing the set of performance features.

[0042] In an embodiment, the obtaining of the first demand vector by weighted summation according to the demand feature data combined with the preset function mapping table comprises: The demand feature data is preprocessed and the function demand is quantified to obtain a function demand degree; The function demand degree is mapped with the preset function mapping table to obtain a demand weight; The first demand vector is calculated by weighted summation according to the demand weight and the demand feature data.

[0043] It should be noted that the process of generating the first demand vector by combining the demand feature data with the preset function mapping table weighted summation and normalizing the performance feature set to obtain the first function vector is as follows: first, the system obtains the demand feature data from the user equipment, including the type of the recently started application (such as social, game or office), the functional demand of the currently active application (such as real-time video streaming requiring high network bandwidth) and the device historical use scene correlation degree (such as the user frequently uses a specific device in the evening period at home). For example, the user is currently using a video conference application (the application type is “office type”), and its functional demand includes high network stability (requires 10 Mbps bandwidth) and low delay audio transmission (requires <50 ms delay). These demand feature data are quantified by the preprocessing module, such as dividing the functional demand into three grades of high, medium and low, or converting into specific numerical values. For example, the network stability demand “high” is quantified as 0.9, and the audio delay demand “low” is quantified as 0.8, forming a function demand degree data set.

[0044] Then, the system matches the quantized function demand degree with the preset function mapping table to obtain the demand weight. The function mapping table defines the weight allocation rules corresponding to different application types or functional demands. For example, the function mapping table of the office application may allocate a network stability weight of 0.6, an audio delay weight of 0.3, and a device compatibility weight of 0.1. Assuming that the user's current demand includes network stability quantified as 0.9 and audio delay as 0.8, the first demand vector is calculated by weighted summation: 0.9x0.6+0.8x0.3+0.0x0.1=0.54+0.24=0.78. This demand vector reflects the user's strong dependence on network performance and the secondary priority of audio quality.

[0045] At the same time, the system obtains the performance feature set from the target device, including hardware parameters (such as processor speed 2GHz, memory 8GB) and functional support (such as supporting Bluetooth 5.2, low power mode). For example, the performance features of a certain Bluetooth earphone are processor speed 1.5GHz, memory 4GB, support Bluetooth 5.0 and high-pass aptX audio encoding. These parameters are converted to a unified dimension range (such as 0 to 1) through standardization processing. For example, the processor speed of 2GHz is mapped to 0.8 after standardization (assuming the maximum reference value is 2.5GHz), and the memory of 8GB is mapped to 0.7 (assuming the maximum reference value is 12GB), and the Bluetooth version 5.2 is mapped to 1.0 (supporting the latest protocol), generating the first function vector 0.8, 0.7, 1.0. Standardization processing ensures that the capabilities of different devices can be compared in the same dimension, for example, the processor speed of a low-end device is 0.5 (1.25GHz) and the processor speed of a high-end device is 0.8 (2GHz), which intuitively reflects the performance gap.

[0046] Finally, the first demand vector and the first function vector will calculate the function matching score through the cosine similarity. For example, the demand vector 0.78 and the function vector 0.8, 0.7, 1.0 calculate the similarity after dimension reduction, if the matching score is higher than the preset threshold (such as 0.8), it is determined that the device meets the current demand of the user. This process realizes accurate function adaptation by dynamically quantifying demand and device capability, for example, in a video conference scenario, the device supporting high bandwidth and low delay is preferentially selected to improve user experience.

[0047] In step S16, the cosine similarity of the first demand vector and the first function vector is calculated, and a function matching score is obtained by mapping.

[0048] In an embodiment, the calculation of the cosine similarity of the first demand vector and the first function vector, and the mapping to obtain the function matching score, comprises: The cosine similarity of the first demand vector and the first function vector is calculated according to the following formula: Wherein, is the cosine similarity of the first demand vector and the first function vector, is the first demand vector, is the first function vector; The cosine similarity is mapped into the interval [0, 100] by linear mapping method to obtain the function matching score.

[0049] Exemplarily, assuming that the user is currently using a video conference application, the demand vector is 0.9, 0.8, 0.6 (corresponding to the demand weight of network stability, audio delay, and device compatibility), and the function vector of a certain Bluetooth headset is 0.8, 0.7, 0.9 (corresponding to the standardized value of processor speed, memory, and Bluetooth version). The dot product is calculated as 0.9x0.8+0.8x0.7+0.6x0.9=0.72+0.56+0.54=1.82, the length of the demand vector is 0.92+0.82+0.62≈1.36, the length of the function vector is 0.82+0.72+0.92≈1.38, and the cosine similarity is 1.82 / (1.36x1.38)≈0.97.

[0050] It is worth mentioning that the system converts the cosine similarity value from the interval of 0, 1, 0 to the functional matching score of 0, 1000, 100 through linear mapping. For example, the cosine similarity of 0.97 is mapped to 0.97 x 100 = 97 points. If the cosine similarity of another device is 0.75, the score is 75. This mapping process intuitively reflects the matching degree of the device and the user's demand, and the higher the score, the more suitable the device is for the current demand.

[0051] In practical applications, if the dimensions of the demand vector and the function vector are inconsistent, the system will ensure comparability through feature alignment or dimension reduction processing. For example, the user demand contains two dimensions of "network stability" and "audio delay", and the device function vector contains "Bluetooth version" and "battery capacity". The system may map the device parameters to the demand dimensions through the function mapping table and then calculate the similarity. In addition, if there are missing values in the function vector (such as a device not providing a Bluetooth version parameter), the system will use a default value (such as 0) or ignore the dimension to avoid calculation errors.

[0052] Finally, the functional matching score is used for device priority sorting. For example, in a video conference scenario, a Bluetooth headset with a score of 97 is preferentially recommended because it supports high bandwidth and low delay, while a normal headset with a score of 75 is listed as a second choice because of insufficient performance. This mechanism quantifies the matching degree of demand and ability, helping users quickly select the optimal device and improving interaction efficiency.

[0053] In step S17, according to the stable distance value, the historical use weight score, the scene relevance score and the functional matching score, the comprehensive priority score of each device is calculated by weighted summation, and normalized processing is performed to obtain the final priority score.

[0054] In one embodiment, the comprehensive priority score of each device is calculated by weighted summation according to the stable distance value, the historical use weight score, the scene relevance score and the functional matching score, comprising: The comprehensive priority score is calculated by the following formula: wherein, is the comprehensive priority score of the device, is the stable distance value, is the basic distance value, is the historical use weight score, is the scene relevance score, is the functional matching score, , , , is a preset weight coefficient, satisfying .

[0055] It should be noted that the process of calculating the comprehensive priority score of the device by weighted summation and normalization is as follows: the system first obtains the characteristic data of the device from four dimensions, including the stable distance value (the actual distance of the device after calibration by the multi-factor signal strength correction model), the historical use weight score (the device activity calculated based on the time decay function and the user's historical pairing behavior), the scene correlation score (the adaptation degree of the environment context matching the pre-defined scene model), and the function matching score (the cosine similarity mapping value of device capability and user demand). For example, the stable distance value of a certain smart speaker is 2.5 meters (near the user), the basic distance value is 1 meter, the historical use weight score is 85 points (high frequency of use and recent activity), the scene correlation score is 90 points (currently in a home scene), and the function matching score is 88 points (supports high-fidelity audio, matches the user's demand to play music).

[0056] The system performs weighted summation on the four dimensions according to the preset weight coefficients , , , . Assuming that the weight coefficients are automatically optimized by historical successful pairing data to be = 10 (distance), = 0.3 (historical use), = 0.25 (scene), = 0.25 (function), then the comprehensive priority score is calculated as: The smaller the stable distance value, the closer the device, which positively contributes to the priority, so it is directly involved in the calculation (if the negative correlation of distance needs to be enhanced, the reciprocal or normalized processing can be performed on the distance).

[0057] Subsequently, the system normalizes the comprehensive scores of all devices and linearly maps them to the [0, 100] interval. For example, assuming that there are three devices in the current environment, their original comprehensive scores are 70.6 (smart speaker), 65.0 (Bluetooth earphone), and 50.2 (smart bulb), respectively, the maximum score is 70.6, and the minimum score is 50.2. The normalization formula is: Final score = wherein is the comprehensive priority score of the device, is the minimum value of the device score, is the maximum value of the device score.

[0058] Exemplarily, for the smart speaker: the final score = , Bluetooth earphone: Smart light bulb: .

[0059] This processing makes the smart speaker get the highest priority due to high historical activity, scene adaptation and function matching, while the smart light bulb is excluded due to low score. By dynamically adjusting the weight and normalization, the system can accurately reflect the priority of the device in different scenes, for example, in an office scene, the function matching weight may be higher, and in a dense device environment, the distance weight will be strengthened, so as to realize the optimization of intelligent matching decision.

[0060] In step S18, according to the final priority score, the preset device database is intelligently grouped and conflict is resolved to obtain a target device, and a dynamic pairing strategy is generated according to the target device and connection is performed.

[0061] In an embodiment, the obtaining of the target device according to the final priority score and the intelligent grouping and conflict resolution of the preset device database include: obtaining device function description features and unique features from a preset device database; performing cluster analysis on the devices according to the device function description features to obtain device grouping; obtaining priority values of the devices in the group, and when the number of devices with the highest priority in the priority values is greater than one and the priority difference is less than a preset difference threshold, extracting historical pairing records of the devices to obtain a stability analysis result.

[0062] According to the stability analysis result and the unique feature, the unique feature score of the device is obtained by weighted summation.

[0063] When the unique feature score of the device is the highest, the device is determined as the target device.

[0064] The generation of the dynamic pairing strategy and the execution of the connection according to the target device include: obtaining the priority score of the target device from the preset device database; determining a connection timeout threshold and a retry number according to the priority score to obtain a dynamic pairing strategy; According to the dynamic pairing strategy, a connection request is generated by using a Bluetooth protocol stack to initiate an active connection to the target device.

[0065] It should be noted that the intelligent grouping and conflict resolution process realizes accurate target device selection through multi-dimensional analysis of device characteristics and user behavior data. First, the system extracts device function description characteristics (such as device type, supported service protocol) and unique characteristics (such as device naming rules, historical pairing stability) from the preset device database. For example, in the smart home scenario, the device database may contain the function description of devices such as smart speakers and smart light bulbs. The function description of the smart speaker may include tags such as "voice interaction" and "music playback", while the smart light bulb includes attributes such as "brightness adjustment" and "color temperature control". At this time, the system performs clustering analysis on the devices based on the function description characteristics, and divides devices with similar functions into the same group. For example, when there are multiple devices supporting the "audio output" function (such as Bluetooth earphones A, Bluetooth speakers B, and car audio C) around the user, these devices will be clustered into an "audio device group", forming an initial device grouping.

[0066] When there are multiple devices with similar priority in the device group (such as Bluetooth earphones A with a priority of 85 points and Bluetooth speakers B with a priority of 83 points, and a preset difference threshold of 5 points), the system will further extract historical pairing records for stability analysis. For example, Bluetooth earphones A have been successfully paired 12 times in the past week, with an average connection duration of 45 minutes, while Bluetooth speakers B have only been paired 5 times, with an average duration of 20 minutes. The system calculates the weighted value of pairing frequency and connection duration (such as frequency weight 0.6, duration weight 0.4) to obtain the stability score of earphone A as 0.8 and the stability score of speaker B as 0.5. At the same time, the system analyzes the device uniqueness characteristics, such as the device name of Bluetooth earphones A being "user's exclusive earphones", which contains personalized naming (long character length and unique semantics), while the name of speaker B is "living room speaker_01", which is highly standardized. By calculating the uniqueness characteristic score (such as name uniqueness accounting for 60%, stability score accounting for 40%), earphone A may obtain 0.9 points and speaker B may obtain 0.6 points, thereby uniquely determining earphone A as the target device.

[0067] If the uniqueness score of the devices in the group still cannot be uniquely distinguished (for example, earphone A and another earphone D both have a score of 0.9), the system will backtrack the historical pairing records and select the device whose last successful pairing time is closer to the current time. For example, earphone A was last paired 10 minutes ago, while earphone D was last paired 3 days ago, so earphone A is preferred as the target device. This mechanism strengthens the user's recent use preference through time decay logic, ensuring the timeliness of the decision.

[0068] After determining the target device, the system generates a dynamic pairing strategy according to its priority score. For example, the priority score of the target device earphone A is 85 points (belongs to the high priority range), and the system queries the corresponding connection parameters from the preset rule library: the timeout threshold is set to 2 seconds, and the retry number is 5 times. This configuration means that the system will try to connect with higher frequency and shorter waiting time, improving the pairing efficiency. Subsequently, the Bluetooth protocol stack generates a connection request packet according to the strategy, and the MAC address, service UUID, and other information of the target device are encapsulated into the request, and the active connection is initiated through the Bluetooth controller. During the connection process, the system monitors the handshake protocol state in real time, and if the first connection fails due to signal interference, it will initiate the second attempt immediately after 2 seconds of timeout, and continue until the 5th retry is exhausted or the connection is successfully established. After successful pairing, the system writes the connection quality data (such as signal strength -65dBm) and response delay (150ms) into the log for subsequent model optimization. For example, if earphone A successfully connects after 3 retries, the system will record the effectiveness of this connection parameter and preferentially adopt a similar strategy in the next pairing. This closed-loop mechanism continuously improves the pairing robustness in multiple scenarios by dynamically adjusting the strategy parameters.

[0069] For example, in a home scenario, after the user starts the music playing application, the system preferentially connects to sound box A rather than earphone B, because the overall priority is higher; if the user enters the car scenario, the system may automatically switch to the car Bluetooth device due to the change in scenario relevance. This mechanism ensures that the device selection not only meets the functional requirements, but also fits the user's habits, significantly improving the pairing efficiency and experience through multi-dimensional conflict resolution and dynamic strategy.

[0070] In summary, the present application discloses a Bluetooth device fast pairing method and system based on multiple scenarios. Through multi-dimensional data fusion and intelligent analysis, the present application significantly improves the fast pairing efficiency and accuracy of Bluetooth devices in multiple scenarios. The system collects scene characteristics, historical pairing records, user demand features, device performance parameters, and signal strength data in real time, accurately calculates the stable distance value of the device using the multi-factor signal correction model and Kalman filter algorithm, and effectively overcomes environmental noise interference; uses the time decay function to dynamically weight the historical usage behavior, and generates a scene relevance score based on the scene matching model, ensuring that the device selection fits the user's current environment; accurately quantifies the adaptation degree of the device and user demand through the cosine similarity mapping of the functional demand vector and the device capability vector. Finally, based on the weighted priority score and the intelligent grouping conflict resolution mechanism, the system automatically selects the optimal target device and executes efficient connection according to the dynamically generated timeout threshold and retry strategy. This scheme not only realizes accurate pairing in a multi-device dense environment and reduces the probability of misconnection, but also optimizes the weight parameters through continuous learning, making the pairing strategy adaptive to the scene, significantly reducing the user operation time, and significantly improving the device collaboration efficiency and user experience in complex scenarios such as smart home and vehicle systems.

[0071] Referring to Figure 2 The second embodiment of the present application provides a multi-scene-based Bluetooth device fast pairing system, comprising: a data acquisition module, configured to acquire current scene characteristic data, device historical pairing data, demand characteristic data, an efficiency characteristic set, and signal strength data; a stable distance module, configured to sample the signal strength data to obtain a time series data set, and combine a pre-set multi-factor signal strength correction model to obtain a stable distance value of the device; a weight score module, configured to weight the historical pairing data through a pre-set time decay function, and perform weight distribution to obtain a historical weight score; a scene matching module, configured to match a pre-defined scene model according to the scene characteristic data to obtain a scene correlation score; a function vector calculation module, configured to combine a pre-set function mapping table according to the demand characteristic data to obtain a first demand vector through weighted summation, and perform standardization processing on the efficiency characteristic set to obtain a first function vector; a function matching module, configured to calculate a cosine similarity of the first demand vector and the first function vector, and perform mapping to obtain a function matching score; a priority calculation module, configured to calculate a comprehensive priority score of each device through weighted summation according to the stable distance value, the historical use weight score, the scene correlation score, and the function matching score, and perform normalization processing to obtain a final priority score; a pairing connection module, configured to perform intelligent grouping and conflict resolution on a pre-set device database according to the final priority score to obtain a target device, generate a dynamic pairing strategy according to the target device, and perform connection.

[0072] It should be noted that the multi-scene-based Bluetooth device fast pairing system provided by the embodiment of the present application is used to execute all process steps of the multi-scene-based Bluetooth device fast pairing method of the above embodiment, and the working principles and beneficial effects of the two are one-to-one correspondence, thus not being repeated.

[0073] The embodiment of the present application further provides an electronic device. The electronic device comprises a processor, a memory, and a computer program stored in the memory and executable on the processor. The processor implements the steps in the above various multi-scene-based Bluetooth device fast pairing method embodiments when executing the computer program, such as Figure 1 the step S11 shown. Alternatively, the processor implements the functions of each module in the above various device embodiments when executing the computer program, such as the peak value module.

[0074] For example, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present application. The one or more modules can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program in the electronic device.

[0075] The electronic device can be a computing device such as a desktop computer, a notebook computer, a palm computer, and a smart tablet. The electronic device can include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device, and the electronic device can include more or fewer components than the above, or combine certain components, or different components, for example, the electronic device can also include an input / output device, a network access device, a bus, etc.

[0076] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic components, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, which connects all parts of the electronic device through various interfaces and lines.

[0077] The memory can be used to store the computer program and / or modules, and the processor realizes various functions of the electronic device by running or executing the computer program and / or modules stored in the memory, and calling data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required by a function (such as a sound playing function, an image playing function, etc.), and the like; and the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), and the like. In addition, the memory can include a high-speed random access memory, and can also include a nonvolatile memory, for example, a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state memory devices.

[0078] The modules / units integrated in the electronic device, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium. Based on this understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. The computer program can realize the steps of the above-mentioned various method embodiments when executed by a processor. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate forms, etc. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the contents included in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.

[0079] It should be noted that the apparatus embodiments described above are merely illustrative, and the units described as separate units can or can not be physically separate, and the units displayed as units can or can not be physical units, i.e. can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. In addition, the connection relationship between the modules in the apparatus embodiment provided by the present application indicates that there is a communication connection between them, which can be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement it without creative labor.

[0080] The above specific embodiments further illustrate the purpose, technical scheme and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the protection scope of the present application. It is particularly pointed out that any modification, equivalent replacement, improvement, etc. made by those skilled in the art within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A multi-scenario based Bluetooth device fast pairing method, characterized in that, The method comprises the following steps: obtaining current scene characteristic data, historical pairing data of the device, demand characteristic data, performance characteristic set and signal strength data; sampling the signal strength data to obtain a time series data set, and combining a preset multi-factor signal strength correction model to obtain a stable distance value of the device; weighting the historical pairing data by a preset time decay function and performing weight distribution to obtain a historical weight score; matching a predefined scene model according to the scene characteristic data to obtain a scene correlation score; combining a preset function mapping table according to the demand characteristic data to obtain a first demand vector by weighted summation, and performing standardization processing on the performance characteristic set to obtain a first function vector; calculating the cosine similarity of the first demand vector and the first function vector, and performing mapping to obtain a function matching score; calculating the comprehensive priority score of each device by weighted summation according to the stable distance value, the historical use weight score, the scene correlation score and the function matching score, and performing normalization processing to obtain a final priority score; according to the final priority score, intelligently grouping and conflict resolving a preset device database to obtain a target device, and generating a dynamic pairing strategy according to the target device and executing connection.

2. The multi-scenario based Bluetooth device fast pairing method according to claim 1, wherein, The sampling of the signal strength data to obtain a time series data set, and combining a preset multi-factor signal strength correction model to obtain a stable distance value of the device comprises: obtaining environmental noise level data; sampling the signal strength data to generate a time series data set containing signal strength values at multiple time points; if the noise level in the environmental noise data set exceeds a preset threshold, increasing the sampling frequency to regenerate the time series data set; processing the time series data set by a Kalman filtering algorithm to obtain a filtered signal strength data set; inputting the filtered signal strength data set into a preset multi-factor signal strength correction model to obtain a stable distance value of the device. 3.The multi-scenario based Bluetooth device fast pairing method of claim 1, wherein, The weighting of the historical pairing data by a preset time decay function and the weight distribution to obtain a historical weight score comprises: weighting the historical pairing data by a preset time decay function to obtain a space-time characteristic data set, and the preset time decay function calculation formula is: wherein, is a time decay weight, is a current time, is a recording time, is a decay coefficient; linearly weighting the space-time characteristic data set to calculate the historical weight score of each device. 4.The multi-scenario based Bluetooth device fast pairing method of claim 1, wherein, The combination of the demand characteristic data with a preset function mapping table to obtain a first demand vector by weighted summation comprises: preprocessing the demand characteristic data and quantifying the function demand to obtain a function demand degree; mapping the function demand degree with a preset function mapping table to obtain a demand weight; calculating the first demand vector by weighted summation according to the demand weight and the demand characteristic data.

5. The multi-scenario based Bluetooth device fast pairing method of claim 1, wherein, The calculation of the cosine similarity of the first demand vector and the first function vector, and the mapping to obtain a function matching score comprises: calculating the cosine similarity of the first demand vector and the first function vector according to the following formula: wherein, is a cosine similarity of the first demand vector and the first function vector, is a first demand vector, is a first function vector; The cosine similarity is mapped into the interval [0, 100] by a linear mapping method to obtain a function matching score.

6. The multi-scenario based Bluetooth device fast pairing method of claim 1, wherein, The comprehensive priority score of each device is calculated by weighted summation according to the stable distance value, the historical use weight score, the scene correlation score and the function matching score, comprising: The comprehensive priority score is calculated by the following formula: wherein, is a comprehensive priority score of the device, is a stable distance value, is a base distance value, is a historical usage weight score, is a scene relevance score, is a function matching score, , , , is a preset weight coefficient, satisfying .

7. The multi-scenario based Bluetooth device fast pairing method of claim 1, wherein, The target device is obtained by intelligent grouping and conflict resolution of a preset device database according to the final priority score, comprising: The device function description features and unique features are obtained from the preset device database; The device is clustered and analyzed according to the device function description features to obtain device grouping; The priority value of the device in the group is obtained, and when the number of devices with the highest priority in the priority value is greater than one and the priority difference is less than a preset difference threshold, the historical pairing record of the device is extracted to calculate a stability analysis result; The unique feature score of the device is obtained by weighted summation according to the stability analysis result and the unique features; When the unique feature score of the device is the highest, the device is determined as the target device.

8. The multi-scenario based Bluetooth device fast pairing method of claim 1, wherein, The dynamic pairing strategy is generated and the connection is performed according to the target device, comprising: The priority score of the target device is obtained from the preset device database; The connection timeout threshold and the number of retries are determined according to the priority score to obtain the dynamic pairing strategy; The connection request is generated by using the Bluetooth protocol stack according to the dynamic pairing strategy to initiate an active connection to the target device.

9. The multi-scenario based Bluetooth device fast pairing method of claim 1, wherein, After the dynamic pairing strategy is generated and the connection is performed according to the target device, comprising: The pairing result, connection quality and use time data of the connection are recorded; The time series database is used to index the device ID and timestamp to form a structured historical record table; The total amount of data of the historical use record is monitored in real time; When the data amount of the historical use record table exceeds a preset data threshold, the device feature vector in the preset device database is updated by incremental learning, the weight coefficient is adjusted, the direction of weight update is judged, and the optimized pairing decision parameter is obtained.

10. A multi-scenario based Bluetooth device fast pairing system, characterized in that, Comprising: A data acquisition module acquires current scene feature data, historical pairing data of the device, demand feature data, an efficiency feature set and signal strength data; A stable distance module samples the signal strength data to obtain a time series data set, and combines a preset multi-factor signal strength correction model to obtain a stable distance value of the device; A weight score module weights the historical pairing data by a preset time decay function and performs weight distribution to obtain a historical weight score; A scene matching module matches a predefined scene model according to the scene feature data to obtain a scene correlation score; A function vector calculation module combines a preset function mapping table according to the demand feature data to obtain a first demand vector by weighted summation, and normalizes the efficiency feature set to obtain a first function vector; A function matching module calculates the cosine similarity of the first demand vector and the first function vector, and maps to obtain a function matching score; A function matching module calculates the cosine similarity of the first demand vector and the first function vector, and maps to obtain a function matching score; A priority calculation module calculates a comprehensive priority score of each device by weighted summation according to the stable distance value, the historical usage weight score, the scene correlation score and the function matching score, and performs normalization processing to obtain a final priority score; A pairing connection module performs intelligent grouping and conflict resolution on a preset device database according to the final priority score to obtain a target device, generates a dynamic pairing strategy according to the target device, and performs connection.

Images