IoT communication platform

Abstract

The present invention discloses an Internet of Things (IoT) communication platform, which relates to the field of IoT communication technology and is used to address the security issue of information leakage, such as the mobile device's audio automatically connecting to the vehicle's Bluetooth when the vehicle is started, as long as the mobile device's Bluetooth is turned on. The platform includes screening out users with high vehicle Bluetooth demand as samples based on their vehicle Bluetooth demand, obtaining in-vehicle mobile device data and user driver's seat adjustment information, and selecting whether the vehicle's Bluetooth connection is automatic or manual based on whether the driver is the user themselves. The platform also classifies users by analyzing their vehicle Bluetooth demand index, and uses users with high vehicle Bluetooth demand as valid data to analyze whether they are the users currently in the vehicle. This ensures that user information security is maintained when users use vehicle Bluetooth, improving safety.

Description

Technical Field

[0001] The present invention relates to the field of Internet of Things communication technology, and more specifically, to an Internet of Things communication platform. Background Art

[0002] The Internet of Things (IoT) connects various physical devices, such as sensors, actuators, and smart appliances, to the internet, enabling them to communicate and collaborate with each other. It primarily consists of sensors, actuators, network connectivity, and communication protocols. HarmonyOS has widespread application in the IoT sector. Its distributed architecture, full-scenario adaptability, and multi-terminal collaboration make it an ideal choice for connecting smart devices and enabling applications such as smart homes and smart cities.

[0003] The existing technology has the following deficiencies:

[0004] HarmonyOS can connect and control mobile devices through various technologies and protocols. When handling Bluetooth vehicle connections, the initial connection of a mobile device requires manual connection. Subsequent mobile device data is stored in a database. As long as the mobile device's Bluetooth is turned on, the mobile device's audio automatically connects to the vehicle's Bluetooth when the vehicle is started. However, in situations such as borrowing a vehicle, when a non-owner starts the vehicle or returns it, the owner's mobile device's audio automatically connects to the vehicle's Bluetooth within a certain range, leading to security risks such as information leakage. This does not meet safety regulations.

[0005] In view of the above problems, the present invention proposes a solution. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an Internet of Things communication platform, through which the in-vehicle Bluetooth is automatically connected or manually connected according to whether the driver is the user himself, so as to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] Internet of Things communication platform, data acquisition module, data processing module and data storage module, the processing steps of each module are as follows;

[0009] Step S1, obtaining user data, including user's car Bluetooth connection data and user's audio playback data;

[0010] Step S2, building a user in-car Bluetooth demand evaluation model based on the user's in-car Bluetooth connection data and the user's audio playback data to evaluate the user's in-car Bluetooth demand level;

[0011] Step S3: Screening out users with high in-car Bluetooth demand as samples based on their in-car Bluetooth demand, obtaining in-car mobile device data and user driver's seat adjustment information, and comprehensively determining whether the current driver is the user based on the in-car mobile device data and the user driver's seat adjustment information;

[0012] Step S4 , selecting whether the in-car Bluetooth is automatically connected or manually connected according to whether the driver is the user himself.

[0013] In a preferred embodiment, in step S1, the user's in-car Bluetooth connection data includes the number of times the user's in-car Bluetooth is connected to the same mobile device, the duration of the mobile device being connected to the in-car Bluetooth, and the number of times audio is played using the mobile device when connected to the in-car Bluetooth.

[0014] In a preferred embodiment, in step S2, if the number of times the current user's in-car Bluetooth is connected to the same mobile device is greater than a preset in-car Bluetooth connection threshold, the current user is marked.

[0015] Get the user's car Bluetooth usage time T, record the specific time point of the car Bluetooth usage time T, mark it and arrange it from small to large to obtain T1, T2, T3 and other data, and merge them into a time series T x , x is the specific time numerical sequence number, x={1,2,3…N}, where N is a positive integer;

[0016] Using the LOF algorithm, we detect abnormal outlier data points based on the density of data points in the time series, calculate the local outlier factor of each data point, classify the collected specific time value points by setting the local outlier factor deviation threshold, and select the duration data points within the deviation threshold to take the average value to obtain the average usage time of the in-vehicle Bluetooth.

[0017] In a preferred embodiment, the average usage time of the vehicle Bluetooth is The specific implementation steps are as follows:

[0018] The specific time value point T x Calculate the standard deviation, standard deviation In the formula is the average value of the time data point; the prediction deviation coefficient T is calculated based on the standard deviation of the time value point z , prediction deviation coefficient T z The calculation expression can be T z =S*exp(S+1) 2 , take the average of the filtered data to get the average usage time of car Bluetooth Then, based on the current user's car Bluetooth usage time and the average car Bluetooth usage time The operator obtains the current user's in-car Bluetooth duration ratio coefficient T b , and T b >0.

[0019] In a preferred embodiment, the number of times a user's in-car Bluetooth audio is played is obtained and organized into an audio data set, and the play data in the audio data set is divided into a training set, a validation set, and a test set;

[0020] The training set is divided into several training samples according to a certain ratio, and iterative training is performed to obtain the squared difference. The squared difference obtained from the validation set is compared with the squared difference obtained from the training set. If the squared difference of the validation set is larger than the squared difference of the training set, the number of iterative training is increased for the training set.

[0021] After the validation set is satisfied, the square difference of the test set data is calculated by the square difference formula. If the square difference of the test set data is greater than the square difference of the training set, the data in the validation set and the test set are randomly swapped and the square difference is recalculated. The operation is repeated until the square difference of the test set is less than or equal to the square difference of the training set. The median of the test set data after the operation is taken as the mean number of car Bluetooth audio playback times and marked as Divide the current user's in-car Bluetooth audio playback times by the average in-car Bluetooth audio playback times Get the audio playback ratio coefficient K C ;

[0022] The obtained vehicle Bluetooth duration ratio coefficient T b and audio playback ratio coefficient K C Construct a car Bluetooth demand model, and compare the results of the logistic regression function calculation with the demand threshold to determine the user's demand for car Bluetooth. Compare the calculated demand evaluation index with the preset demand threshold. When the demand evaluation index is greater than or equal to the preset demand threshold, the user is considered a high-demand car Bluetooth user.

[0023] In a preferred embodiment, in step S3, the in-vehicle Bluetooth high-demand users screened and marked according to the demand assessment index are used as samples to determine whether the current driver is the user himself. The specific steps are as follows:

[0024] Select a period of time as the sample interval, collect the height and length data of the user's seat adjustment while driving, select time points with the same time interval within the sample period and record them as t1, t2, t3, etc., record the seat height H and length L at each point, organize the seat height into a height data set, and organize the seat length L into a length data set. After removing the largest and smallest data in the height data set, arrange the data from small to large to obtain the median H of the height data in the remaining height data. Z , taking the median of height data HZ As the center, take the average of the height data on both sides to get and and After removing the largest and smallest data in the length data, arrange the data in ascending order and obtain the median H of the remaining length data. Z , with the median length data H Z As the center, take the average of the length data on both sides to get and and

[0025] When the user's seat height H and length L are both within the user seat threshold range, it is determined that the current driver is the user himself.

[0026] In a preferred embodiment, in step S4, the in-vehicle Bluetooth high-demand users are first analyzed from the user data, and the in-vehicle Bluetooth high-demand users are marked as marked users. When the marked user's vehicle is started, it is determined whether the driver is the driver himself based on whether a new Bluetooth device appears and whether an abnormal position adjustment occurs. Then, the in-vehicle Bluetooth is automatically connected to the mobile device. Specifically:

[0027] When a new Bluetooth device appears in the car, the car Bluetooth will not be automatically connected, and a prompt will be given to manually connect the Bluetooth;

[0028] When the current driver connects the car's Bluetooth to a mobile device in the user's history, the user is prompted whether to connect the Bluetooth;

[0029] When no new device is connected in the car, the current driver's driving position is recorded and compared with the user seat threshold to determine whether the car's Bluetooth automatically connects to the mobile device:

[0030] If the current driver's driving position is within the user seat threshold range, the current driver is determined to be the user himself, and the car Bluetooth is turned on to automatically connect to the Bluetooth device. If the current driver's driving position is outside the user seat threshold range, the car Bluetooth is not automatically connected, and the current driver is prompted whether to manually connect the mobile device.

[0031] The technical effects and advantages of the digital cultural tourism management system based on multi-source data analysis of the present invention are as follows:

[0032] This invention detects and analyzes user in-car Bluetooth usage information in a domestic operating system to assess whether there is a risk in automatically connecting to the in-car Bluetooth and whether to proceed with the automatic connection. First, user usage information for in-car Bluetooth is collected, and a user in-car Bluetooth demand assessment model is constructed. Users are then classified by analyzing their in-car Bluetooth demand index, and users with high in-car Bluetooth demand are used as valid data to determine whether they are actually in the vehicle. This ensures that user information security is maintained when users use in-car Bluetooth, improving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a logic flow chart of the Internet of Things communication platform processing of the present invention;

[0034] Figure 2 This is a flow chart of the Internet of Things communication platform processing method of the present invention. DETAILED DESCRIPTION

[0035] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0036] This invention detects and analyzes user in-car Bluetooth usage information in a domestic operating system to assess whether there is a risk in automatically connecting to the in-car Bluetooth and whether to proceed with the automatic connection. First, user usage information for in-car Bluetooth is collected, and a user in-car Bluetooth demand assessment model is constructed. Users are then classified by analyzing their in-car Bluetooth demand index, and users with high in-car Bluetooth demand are used as valid data to determine whether they are actually in the vehicle. This ensures that user information security is maintained when users use in-car Bluetooth, improving safety.

[0037] Example 1

[0038] like Figure 1 、 Figure 2 As shown, in step S1, user data is obtained, including the user's in-vehicle Bluetooth connection data and the user's audio playback data. The vehicle manufacturer information can be obtained by accessing the vehicle identification code. The manufacturer's cloud server is accessed to collect the user's in-vehicle Bluetooth connection data and the user's audio playback data.

[0039] Step S2: constructing a user's in-car Bluetooth demand assessment model based on the user's in-car Bluetooth connection data and the user's audio playback data to assess the user's in-car Bluetooth demand level.

[0040] In step S3, users with high in-car Bluetooth demand are selected as samples based on their level of in-car Bluetooth demand. In-car mobile device data is obtained through the in-car entertainment system, and driver's seat adjustment information is obtained by accessing the manufacturer's electronic control system. Based on the in-car mobile device data and the driver's seat adjustment information, it is determined whether the current driver is the user.

[0041] Step S4 determines whether the driver is the user and selects whether to automatically connect or manually connect the in-car Bluetooth. Automatic connection means that the in-car Bluetooth automatically connects to the Bluetooth mobile device in the database. Manual connection prompts the driver whether to manually connect. If the connection is made, a notification and an option to connect to the in-car Bluetooth are sent to the Bluetooth mobile device.

[0042] The specific implementation is as follows:

[0043] In step S1, basic vehicle information, such as the year of manufacture and manufacturer, is retrieved using the vehicle's VIN. Once the manufacturer information is known, the user's in-car Bluetooth usage can be accessed by accessing the manufacturer's cloud server or by accessing the vehicle's in-car entertainment system. This information can be used to determine the number of times the user's in-car Bluetooth has been connected to the same mobile device, the duration of the mobile device's connection, and the number of times the mobile device has played audio while connected to the in-car Bluetooth.

[0044] It should be noted that a VIN query is a vehicle identification number query, which is a unique identifier that can be used to look up a vehicle's registration information. Some online service providers allow detailed vehicle information, including user information, to be queried using the VIN. A vehicle manufacturer's cloud server refers to a cloud computing platform managed and operated by the automaker that collects, stores, and processes vehicle-related data. These servers are typically used to connect the vehicle to the Internet and provide various services and functions, such as vehicle data analysis and safety services. An in-car entertainment system is a multimedia system installed in a car that is designed to provide various entertainment and information functions, including Bluetooth connectivity. In this case, vehicle Bluetooth usage data is typically stored in the in-car entertainment system's internal memory or on a connected device, such as an on-board control unit or media center.

[0045] In step S2, the number of times the current user's in-car Bluetooth is connected to the same mobile device over a random period of time is collected, and a threshold for in-car Bluetooth connection is set based on actual conditions. The random period of time can be one month, and the threshold for in-car Bluetooth connection can be 20 times. The obtained number of in-car Bluetooth connections to the same mobile device is compared with the threshold for in-car Bluetooth connection. If the number of times the current user's in-car Bluetooth is connected to the same mobile device is greater than the preset threshold for in-car Bluetooth connection, the current user is marked. Otherwise, no marking is performed. The preset threshold for in-car Bluetooth connection is determined by actual conditions and is not limited here.

[0046] Collect enough users' in-car Bluetooth usage time T, record the specific time points of the in-car Bluetooth usage time T, mark them, and arrange them from small to large to get T1, T2, T3 and other data and merge them into a time series T x , x is the specific time value serial number, x = {1, 2, 3... N}, where N is a positive integer. Using the LOF algorithm, we detect abnormal outlier data points based on the density of data points in the time series, calculate the local outlier factor of each data point, and classify the collected specific time value points by setting the local outlier factor deviation threshold. We then select the duration data points within the deviation threshold and take the average value to obtain the average usage time of the in-vehicle Bluetooth. The specific implementation steps are as follows:

[0047] The specific time value point T x Calculate the standard deviation, standard deviation In the formula is the average value of the time data point; the prediction deviation coefficient T is calculated based on the standard deviation of the time value point z , prediction deviation coefficient T z The calculation expression can be T z =S*exp(S+1) 2 From the prediction deviation coefficient expression, we can see that when the prediction deviation coefficient is larger, that is, the standard deviation is larger, the data spacing in the time series is larger. When the prediction deviation coefficient is greater than the preset deviation threshold, that is, the data spacing in the time series is too large, its data value is not used as a reference, which improves the accuracy of the sample data. The average value of the data filtered out in the above steps is taken to obtain the average usage time of the car Bluetooth Then, based on the current user's car Bluetooth usage time and the average car Bluetooth usage time The operator obtains the current user's in-car Bluetooth duration ratio coefficient T b , and T b >0. When the average usage time of car Bluetooth is When determining, the vehicle Bluetooth duration ratio coefficient T bThe larger the value is, the longer the current user's car Bluetooth usage time is, and the greater the current user's demand for car Bluetooth usage is; the car Bluetooth time ratio coefficient T b The smaller it is, the shorter the current user's car Bluetooth usage time is, and the less demand the current user has for using car Bluetooth is.

[0048] It should be noted that the LOF algorithm, or local outlier factor algorithm, is an unsupervised learning algorithm for detecting outliers in a data set. Unlike traditional global density-based methods, the LOF algorithm takes into account the density difference between each data point and its neighboring data points. It determines the degree of abnormality of each data point by comparing the density of each data point with its neighboring data points, making the filtered data more accurate. An abnormal outlier data point refers to an observation value that is significantly different from its data point. The local outlier factor is a commonly used anomaly detection algorithm used to identify local outliers in a time series, where local outliers are determined relative to their surrounding neighboring points. The deviation thresholds mentioned in the above steps are all set by those skilled in the art based on actual conditions.

[0049] Collect enough in-car Bluetooth audio playback times from users and organize them into an audio dataset. Divide the playback data in the audio dataset into a training set, a validation set, and a test set. The training set accounts for 75% of the audio dataset, the validation set accounts for 15% of the audio dataset, and the test set accounts for 10% of the audio dataset. The audio dataset ratio distribution mentioned in the above steps is not unique and can be set according to actual conditions.

[0050] The training set is divided into several training samples according to a certain ratio. A random sampling method is used to select a data from each sample, and the minimum square difference is calculated for the extracted data. Several iterative trainings are performed, and the average of the obtained square differences is taken. The same is true for the validation set. The above method is used for iterative training and the square difference is obtained. The square difference obtained by the validation set is compared with the square difference obtained by the training set. If the square difference of the validation set is smaller than the square difference of the training set, it means that the sample data of the validation set is within the sample data interval of the training set. The above operation is correct. If the square difference of the validation set is larger than the square difference of the training set, then add iterative training times to the training set to enhance the accuracy. Specifically, the formula for calculating the square difference can be Among them, PFC is the mean square error, Z is the total number of training samples, is the predicted value of the i-th training sample, Y iis the actual value of the i-th training sample. After satisfying the validation set, the data of the test set is squared using the above formula for squared difference. If the squared difference of the test set data is less than or equal to the squared difference of the training set, it means that the data accuracy of the test set is high. If the squared difference of the test set data is greater than the squared difference of the training set, the data in the validation set and the test set are randomly swapped and the squared difference is recalculated. Repeat the operation until the squared difference of the test set is less than or equal to the squared difference of the training set. The median of the test set data after the operation is taken as the mean of the number of car Bluetooth audio playback times and marked as Divide the current user's in-car Bluetooth audio playback times by the average in-car Bluetooth audio playback times Get the audio playback ratio coefficient K C .

[0051] Audio playback ratio coefficient K C The average of the current user's in-car Bluetooth playback times and in-car Bluetooth audio playback times When the calculated average number of car Bluetooth audio playback times is When determined, the audio playback ratio coefficient K C The larger the value is, the more times the user plays the in-car Bluetooth audio, and the greater the user's demand for in-car Bluetooth; the audio playback ratio coefficient K C The smaller it is, the fewer times the user's in-car Bluetooth audio is played, and the less demand the user has for in-car Bluetooth.

[0052] The obtained vehicle Bluetooth duration ratio coefficient T b and audio playback ratio coefficient K C By building a demand model for in-car Bluetooth and comparing the results of the logistic regression function with the demand threshold, we can determine the current user's demand for in-car Bluetooth. The formula can be: Among them, a and b are weight indexes and are both greater than 0, Q is the demand assessment index of in-vehicle Bluetooth, and the calculated demand assessment index Q is compared with the preset demand threshold. When Q is greater than or equal to the preset demand threshold, it means that the user has a high demand for in-vehicle Bluetooth. It is more necessary to pay attention to the information leakage problem that may be caused by the automatic connection of in-vehicle Bluetooth to mobile devices. The user is marked as a high-demand user for in-vehicle Bluetooth; when the demand assessment index Q is less than the preset demand threshold, it means that the user has a low demand for in-vehicle Bluetooth. Because the risk of information leakage caused by the automatic connection of in-vehicle Bluetooth to mobile devices is low, such users are not marked.

[0053] In step S3, the high-demand in-vehicle Bluetooth users, identified and screened based on the demand assessment index, are used as samples to determine whether the current driver is the user. The following factors are analyzed: The presence of a new Bluetooth mobile device in the vehicle at vehicle startup is used as one factor in determining whether the current driver is the user. The degree of driver seat adjustment before the vehicle starts is used as a second factor in determining whether the current driver is the user.

[0054] The user seat threshold is calculated by obtaining the user's driving seat adjustment information from the manufacturer's electronic control system. When the vehicle is started, the in-car entertainment system will automatically detect if a new device appears on the Bluetooth panel and prompt the user whether to manually connect to Bluetooth. If the in-car entertainment system does not detect the new device in the car, it will record the driver's driving position and compare it with the user seat threshold. If the driver's driving position is within the user seat threshold range, it is determined that the current driver is the user himself. Here is a method for setting the user seat threshold. The specific steps are as follows:

[0055] Select a period of time as a sample interval to collect data on the height and length of the seat when the user adjusts it while driving. The height here refers to the vertical height of the top of the backrest, and the length is the distance the seat cushion is pushed forward and backward. Select time points with the same time interval within the sample period and record them as t1, t2, t3, etc., and record the seat height H and length L at each point. Organize the seat height into a height data set, and organize the seat length L into a length data set. After removing the largest and smallest data in the height data set, sort the data from small to large and obtain the median H of the remaining data. Z , with the median H Z As the center, take the average of the data on both sides to get and and The length data set is operated according to the above method to obtain the average length and and The upper threshold of the seat height can be set to The lower seat height threshold can be set to The upper threshold of seat length can be set to The lower seat length threshold can be set to

[0056] When the user's seat height H and length L are both within the user's seat threshold range, that is, and The current driver is then determined to be the user. In particular, the more precisely the selected sample time period is divided, the more accurate the collected user driving location data. When the accuracy of the user driving location data falls within a preset accuracy threshold, such as a few centimeters or millimeters, the likelihood that the current driver is the user is greater. This accuracy threshold is not unique and is determined by professionals in this field based on actual circumstances. It will not be analyzed here.

[0057] It should be noted that the vehicle manufacturer's electronic control system refers to the computer system used to manage and control various electronic devices and systems in the vehicle. These systems are usually composed of multiple electronic control units, each of which is responsible for monitoring and controlling specific functions and subsystems. By marking users with high demand for in-vehicle Bluetooth as samples, the accuracy of the system in responding to risk issues can be improved. When a new Bluetooth mobile device is detected in the car, it means that a stranger has entered the car or the user has obtained a new Bluetooth device. In order to reduce the risk, the automatic connection is not turned on and a manual Bluetooth connection prompt is issued. When a stranger uses the in-vehicle Bluetooth to manually connect to the user's mobile Bluetooth device, the user's mobile device will receive pairing information on whether to connect to the in-vehicle Bluetooth, so that the user can choose whether to connect, thereby avoiding risks. The above method of setting seat thresholds is not unique.

[0058] In step S4, a determination is made as to whether automatic Bluetooth connection is enabled. First, user data is analyzed to identify high-demand users for in-vehicle Bluetooth. These users are then marked as marked users. When the marked user's vehicle is started, the driver is determined to be the user based on the presence of a new Bluetooth device and any abnormal position adjustment. The driver is then prompted to automatically connect to the mobile device using the in-vehicle Bluetooth. If a new Bluetooth device appears in the vehicle, automatic Bluetooth connection is not initiated, and a manual Bluetooth connection prompt is issued. If the current driver attempts to connect the in-vehicle Bluetooth to a mobile device in the user's history, the user is prompted to initiate a Bluetooth connection. If no new device is connected in the vehicle, the current driver's driving position is recorded and compared with the user's seat threshold to determine whether the in-vehicle Bluetooth automatically connects to the mobile device. If the current driver's driving position is within the user's seat threshold, the driver is determined to be the user and automatic Bluetooth connection to the Bluetooth device is enabled. If the current driver's driving position is outside the user's seat threshold, automatic Bluetooth connection is not initiated, and the driver is prompted to manually connect to the mobile device.

[0059] The present invention calculates various factors, compares actual values ​​with the size and range of various thresholds, and analyzes them to screen out and mark users with high in-car Bluetooth demand. This system then addresses the information leakage issues that may occur when in-car Bluetooth automatically connects to mobile devices while driving, posing a greater security risk to these marked users. This system can assess risk while driving, confirm whether the current driver is the user, and then choose whether to automatically connect to mobile devices with in-car Bluetooth. This system provides a reasonable solution, eliminating the security risks of information leakage and other issues previously associated with automatic in-car Bluetooth connection issues. It also mitigates the potential leaks of company secrets and theft of research results caused by these issues. This system significantly improves user safety and convenience, while also recording each user's demand for in-car Bluetooth. This system provides vehicle manufacturers with better user classification and rectification guidance, facilitating the implementation of operational strategies. The aforementioned formulas are all dimensionless and numerically calculated. The formulas are based on recent real-world scenarios obtained through software simulation using a large amount of collected data. The preset parameters in the formulas are set by those skilled in the art based on actual conditions.

[0060] Example 2

[0061] The Internet of Things communication platform of the present invention further includes a data acquisition module, a data processing module and a data storage module to implement the method described in Example 1.

[0062] The data acquisition module is used to collect data such as the user's car Bluetooth connection and user audio playback, and send it to the data processing module to ensure the operation of the subsequent data processing module.

[0063] The data processing module is used to analyze and process the Bluetooth connection status according to the data collected by the data collection module.

[0064] The data storage module is used to store all data generated during the processing of the Internet of Things communication platform.

[0065] The above embodiments may be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0066] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0067] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0068] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0069] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. The Internet of Things communication platform includes a data acquisition module, a data processing module, and a data storage module, and is characterized by: The processing steps of each module are as follows; Step S1, obtaining user data, including user's car Bluetooth connection data and user's audio playback data; Step S2, building a user in-car Bluetooth demand evaluation model based on the user's in-car Bluetooth connection data and the user's audio playback data to evaluate the user's in-car Bluetooth demand level; Step S3: Screening out users with high in-car Bluetooth demand as samples based on their in-car Bluetooth demand, obtaining in-car mobile device data and user driver's seat adjustment information, and comprehensively determining whether the current driver is the user based on the in-car mobile device data and the user driver's seat adjustment information; Step S4 , selecting whether the in-car Bluetooth is automatically connected or manually connected according to whether the driver is the user himself.

2. The Internet of Things communication platform according to claim 1, characterized in that: In step S1 , the user's in-car Bluetooth connection data includes the number of times the user's in-car Bluetooth is connected to the same mobile device, the duration of the mobile device being connected to the in-car Bluetooth, and the number of times the mobile device is used to play audio when connected to the in-car Bluetooth.

3. The Internet of Things communication platform according to claim 2, characterized in that: In step S2, if the number of times the current user's car Bluetooth is connected to the same mobile device is greater than the preset car Bluetooth connection threshold, the current user is marked. Get the user's car Bluetooth usage time T, record the specific time point of the car Bluetooth usage time T, mark it and arrange it from small to large to obtain T1, T2, T3 and other data, and merge them into a time series T x , x is the specific time numerical sequence number, x={1,2,3…N}, where N is a positive integer; Using the LOF algorithm, we detect abnormal outlier data points based on the density of data points in the time series, calculate the local outlier factor of each data point, classify the collected specific time value points by setting the local outlier factor deviation threshold, and select the duration data points within the deviation threshold to take the average value to obtain the average usage time of the in-vehicle Bluetooth.

4. The Internet of Things communication platform according to claim 3, characterized in that: Average usage time of car Bluetooth The specific implementation steps are as follows; The specific time value point T x Calculate the standard deviation, standard deviation In the formula is the average value of the time data point; the prediction deviation coefficient T is calculated based on the standard deviation of the time value point z , prediction deviation coefficient T z The calculation expression can be T z =S*exp(S+1) 2 , take the average of the filtered data to get the average usage time of car Bluetooth Then, based on the current user's car Bluetooth usage time and the average car Bluetooth usage time The operator obtains the current user's in-car Bluetooth duration ratio coefficient T b , and T b >0.

5. The Internet of Things communication platform according to claim 4, characterized in that: Obtain the user's in-car Bluetooth audio playback times and organize them into an audio dataset. The playback data in the audio dataset is divided into a training set, a validation set, and a test set. The training set is divided into several training samples according to a certain ratio, and iterative training is performed to obtain the squared difference. The squared difference obtained from the validation set is compared with the squared difference obtained from the training set. If the squared difference of the validation set is larger than the squared difference of the training set, the number of iterative training is increased for the training set. After the validation set is satisfied, the square difference of the test set data is calculated by the square difference formula. If the square difference of the test set data is greater than the square difference of the training set, the data in the validation set and the test set are randomly swapped and the square difference is recalculated. The operation is repeated until the square difference of the test set is less than or equal to the square difference of the training set. The median of the test set data after the operation is taken as the mean number of car Bluetooth audio playback times and marked as Divide the current user's in-car Bluetooth audio playback times by the average in-car Bluetooth audio playback times Get the audio playback ratio coefficient K C ; The obtained vehicle Bluetooth duration ratio coefficient T b and audio playback ratio coefficient K C A demand model for in-vehicle Bluetooth is constructed, and the calculation results of the logistic regression function are compared with the demand threshold to determine the user's demand for in-vehicle Bluetooth. The calculated demand assessment index is compared with the preset demand threshold. When the demand assessment index is greater than or equal to the preset demand threshold, the user is classified as a high-demand user for in-vehicle Bluetooth.

6. The Internet of Things communication platform according to claim 1, characterized in that: In step S3, the in-vehicle Bluetooth high-demand users screened and marked according to the demand assessment index are used as samples to determine whether the current driver is the user himself. The specific steps are as follows: Select a period of time as the sample interval, collect the height and length data of the user's seat adjustment while driving, select time points with the same time interval within the sample period and record them as t1, t2, t3, etc., record the seat height H and length L at each point, organize the seat height into a height data set, and organize the seat length L into a length data set. After removing the largest and smallest data in the height data set, arrange the data from small to large to obtain the median H of the height data in the remaining height data. Z , taking the median of height data H Z As the center, take the average of the height data on both sides to get and and After removing the largest and smallest data in the length data, arrange the data in ascending order and obtain the median H of the remaining length data. Z , with the median length data H Z As the center, take the average of the length data on both sides to get and and When the user's seat height H and length L are both within the user seat threshold range, it is determined that the current driver is the user himself.

7. The Internet of Things communication platform according to claim 1, characterized in that: In step S4, the in-vehicle Bluetooth high-demand users are first analyzed from the user data, and the in-vehicle Bluetooth high-demand users are marked as marked users. When the marked user's vehicle is started, the driver is judged to be the driver based on whether a new Bluetooth device appears and whether an abnormal position adjustment occurs. Then, the in-vehicle Bluetooth is automatically connected to the mobile device. Specifically: When a new Bluetooth device appears in the car, the car Bluetooth will not be automatically connected, and a prompt will be given to manually connect the Bluetooth; When the current driver connects the car's Bluetooth to a mobile device in the user's history, the user is prompted whether to connect the Bluetooth; When no new device is connected in the car, the current driver's driving position is recorded and compared with the user seat threshold to determine whether the car's Bluetooth automatically connects to the mobile device: If the current driver's driving position is within the user seat threshold range, the current driver is determined to be the user himself, and the car Bluetooth is turned on to automatically connect to the Bluetooth device. If the current driver's driving position is outside the user seat threshold range, the car Bluetooth is not automatically connected, and the current driver is prompted whether to manually connect the mobile device.

Images