Intelligent ground lock remote management method and system based on Bluetooth communication

Abstract

The present invention relates to the field of remote management of ground locks, and in particular to a method and system for remote management of smart ground locks based on Bluetooth communication. The method comprises the following steps: obtaining the operation log of the smart ground lock and the real-time Bluetooth connection status parameters of the smart ground lock; evolving the personalized behavior habits of users according to the operation log of the smart ground lock, and modeling the user portrait to construct the unlocking behavior portrait of each user; quantifying the real-time user behavior abnormal deviation and abnormal ground lock usage decision of the real-time Bluetooth connection status parameters of the smart ground lock based on the unlocking behavior portrait of each user, and generating an abnormal ground lock early warning strategy; when the real-time user behavior abnormal deviation value does not exceed the preset behavior deviation threshold, triangulating the real-time Bluetooth connection status parameters of the smart ground lock and performing user precise positioning calculation to obtain the real-time position coordinates of the user. The present invention realizes safe and efficient remote management of ground locks.

Description

Technical Field

[0001] The present invention relates to the field of remote management of ground locks, and in particular to a method and system for remote management of intelligent ground locks based on Bluetooth communication. Background Art

[0002] With the rapid development of Internet of Things (IoT) technology and the popularization of smart homes, smart floor locks, as a core component of smart home systems, have been widely used in residential, commercial, office and other scenarios to provide more efficient, convenient and secure access control management services. Smart floor locks integrate multiple technical means such as Bluetooth communication, fingerprint recognition, password input, and remote control to achieve real-time monitoring and remote operation of door lock status, greatly improving the user experience and security.

[0003] However, with the popularity of smart ground locks and the increase in frequency of use, traditional ground lock management methods have gradually exposed many problems. For example, the maintenance and management of the equipment lack intelligence, the use status cannot be monitored in real time, and the user needs cannot be responded to quickly. As a result, when users face lock failures or safety hazards, it is often difficult to deal with them in a timely manner. In addition, since most smart ground locks rely on Bluetooth communication for operation, they are affected by the external environment (such as Bluetooth signal interference, equipment failure, etc.), which may lead to unstable communication connections, delayed responses, and other problems, thereby affecting the unlocking efficiency and system security.

[0004] Existing smart ground lock management methods usually rely on traditional single device detection and manual intervention, lacking automated and intelligent monitoring methods. The functions of active monitoring and early warning of the operating status of the device and user behavior are not yet perfect, and cannot effectively avoid the occurrence of equipment abnormalities and safety hazards. Therefore, how to achieve remote management of smart ground locks through intelligent means, monitor and optimize Bluetooth communication performance, and timely warn of equipment failures and repair them has become an important topic in the development of smart ground lock technology. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a remote management method and system for smart ground locks based on Bluetooth communication to solve at least one of the above technical problems.

[0006] To achieve the above object, the present invention provides a remote management method for an intelligent ground lock based on Bluetooth communication, comprising the following steps:

[0007] Step S1: Obtain the operation log of the smart ground lock and the real-time Bluetooth connection status parameters of the smart ground lock; evolve the user's personalized behavior habits according to the operation log of the smart ground lock, and perform user portrait modeling to build an unlocking behavior portrait of each user;

[0008] Step S2: Based on the unlocking behavior profile of each user, the real-time Bluetooth connection status parameters of the smart lock are quantified for abnormal user behavior deviations and abnormal lock usage decisions, and abnormal lock warning strategies are generated;

[0009] Step S3: When the real-time user behavior abnormal deviation value does not exceed the preset behavior deviation threshold, the real-time Bluetooth connection state parameters of the smart lock are triangulated and the user is accurately positioned to obtain the user's real-time location coordinates;

[0010] Step S4: predicting the user unlocking time period based on the unlocking behavior profile of each user, and generating a pre-unlocking strategy for the ground lock;

[0011] Step S5: unlocking and identifying the user's real-time location coordinates according to the ground lock pre-unlocking strategy, and performing a quick unlocking process to generate quick unlocking response data;

[0012] Step S6: Bluetooth communication parameters are tuned for the quick unlock response data, and collaborative reinforcement learning is performed based on the abnormal ground lock warning strategy to build an intelligent ground lock management model to perform remote management operations for the intelligent ground lock.

[0013] The present invention can accurately capture the usage habits, behavior patterns and preferences of each user by collecting the operation log and real-time Bluetooth connection status parameters of the smart lock. The daily unlocking time, frequency, unlocking device, etc. of a certain user. Based on these data, a personalized unlocking behavior portrait can be established for each user. Over time, the user's behavior pattern may change. By continuously updating and optimizing the user portrait, it is possible to better adapt to the evolution of the user's behavior habits and ensure that the system can be consistent with user needs. By comparing the user's unlocking behavior portrait with the real-time data, the deviation of the user's behavior from the normal mode can be discovered in time. These abnormal deviations may represent security risks (such as abnormal unlocking time, location or device, etc.). By quantifying the abnormal deviation of user behavior, corresponding abnormal lock usage decisions can be generated. If an abnormal unlocking time of the user is detected, the system can trigger an early warning and take corresponding security measures (such as mandatory identity authentication). Triangulation positioning is performed through real-time Bluetooth connection status parameters to locate the user's position in the actual scene. Through data such as Bluetooth signal strength, the system can provide high-precision user location judgment. Real-time positioning can help determine whether the user is in a normal unlocking scene. If the user's position is in an abnormal area, the system can trigger additional identity authentication or restrict unlocking. By analyzing the user's unlocking time period and behavior pattern, the system can predict the user's unlocking demand time. If the system knows that the user usually unlocks in a certain time period, the corresponding unlocking strategy can be prepared in advance. By predicting the user's needs in advance, the system can shorten the response time, reduce the user's waiting time, and provide a smoother experience. By combining the user's location and pre-unlocking strategy, the system can achieve fast unlocking. When the user approaches the ground lock, the system can automatically identify and start the unlocking process without the user having to do too much operation. Users do not need to perform tedious identity authentication or wait for unlocking operations, which greatly improves the convenience of use. Especially in emergency or high-demand scenarios, fast unlocking is particularly important. Bluetooth communication parameter tuning can improve the connection stability between devices, especially in environments with large signal interference. By adjusting the communication parameters, the smart ground lock can ensure efficient and stable Bluetooth connection. Through collaborative reinforcement learning, the system can continuously adjust and optimize the management strategy of the ground lock based on historical data and real-time feedback. This not only helps to improve security, but also improves the response speed and accuracy of the device. Through the intelligent ground lock management model, the system can achieve remote management of the ground lock, including unlocking history query, device status monitoring, remote control and other functions. Administrators can manage multiple ground locks from anywhere to improve work efficiency and convenience.

[0014] Preferably, step S1 comprises the following steps:

[0015] Step S11: Obtaining the smart ground lock operation log and the real-time Bluetooth connection status parameters of the smart ground lock; Step S12: Identifying the Bluetooth connection address according to the smart ground lock operation log, and generating multiple Bluetooth connection users;

[0016] Step S13: Based on the intelligent ground lock operation log, the ground lock unlocking behavior detection is performed on multiple Bluetooth connected users, and the unlocking behavior data of each user is extracted;

[0017] Step S14: performing personalized behavior habit evolution on the unlocking behavior data of each user to generate unlocking behavior habit features of each user;

[0018] Step S15: performing user unlocking authority formulation processing based on the unlocking behavior habit characteristics of each user to obtain the unlocking authority of each user;

[0019] Step S16: Perform user portrait modeling based on the unlocking authority of each user and the unlocking behavior habit characteristics of each user to construct an unlocking behavior portrait of each user.

[0020] The present invention can grasp the usage status, connection status and operation health of the ground lock in real time by collecting the operation log and real-time Bluetooth connection status parameters of the smart ground lock. These data provide a basis for subsequent user behavior analysis, anomaly detection and authority control. Acquiring real-time Bluetooth connection status parameters enables the system to accurately monitor the device connection status, such as whether the Bluetooth connection is normal, whether there is signal interference, etc., thereby ensuring the stable operation of the smart ground lock. By identifying multiple Bluetooth connection addresses, the system can process unlocking requests from multiple users at the same time, which is very important for scenarios where multiple users (such as homes, offices, etc.) use smart ground locks. By identifying the Bluetooth connection address, the system can accurately distinguish between different users. By analyzing the operation log of the smart lock, the specific unlocking behavior data of each user can be obtained, such as the time, frequency, and device used for unlocking. Such data provides valuable information for further personalized analysis and behavior modeling. The system can analyze the user's unlocking behavior from multiple dimensions, not only including a single time point or location, but also multiple data points such as behavior frequency and abnormal operations, to help understand the user's behavior pattern more accurately. The system conducts personalized behavior habit evolution based on the unlocking behavior data of each user, thereby generating behavioral characteristics for each user, which makes each The behavior of each user can be understood and predicted individually, which improves the intelligence level of the system. As time goes by, the user's behavior habits may change (changing the unlocking time or the device used). The system can capture these changes in time through behavior evolution and automatically update the user's unlocking behavior characteristics. Based on the personalized unlocking behavior characteristics of each user, the system can dynamically formulate unlocking permissions for users. The system can determine whether a user should have higher permissions based on the user's behavior habits, or whether there are unsafe unlocking behaviors that need to be restricted. Different users have different behavior habit characteristics. The system can decide whether to adjust the user's unlocking permissions based on factors such as the security and stability of their behavior. A user who frequently attempts to unlock unsuccessfully may be restricted from further unlocking operations or require stronger identity authentication. By combining the unlocking behavior habit characteristics and unlocking permissions of each user, the system can create a detailed unlocking behavior portrait for each user. These portraits can comprehensively and accurately describe the behavior characteristics, needs and security status of each user. User portraits enable the smart lock system to provide personalized services and management for each user. According to the information in the portrait, the system can optimize the unlocking response time, permission management, etc. in a targeted manner to improve the user experience. As user behavior changes, user portraits can also be dynamically updated, which means that the system can adapt to changes in user needs and behavior in real time to maintain intelligence and efficiency.

[0021] Preferably, the specific steps of step S14 are:

[0022] Calculate the periodic unlocking frequency of each user's unlocking behavior data to generate the user's periodic unlocking frequency;

[0023] Perform unlocking frequency distribution analysis on the user's periodic unlocking frequency to obtain the unlocking frequency distribution characteristics of each user;

[0024] Extract unlocking timestamp based on each user’s unlocking behavior data;

[0025] Perform ground lock unlocking time analysis on the unlocking timestamp to generate the ground lock unlocking time feature of the user;

[0026] The unlocking frequency distribution characteristics of each user and the unlocking time characteristics of the user's ground lock are personalized and evolved to generate the unlocking behavior habit characteristics of each user.

[0027] By calculating the periodic unlocking frequency of users, the present invention can identify the regularity of users' unlocking behaviors. Some users may unlock at fixed time periods every day, while other users may show more random unlocking behaviors. Periodic unlocking frequency analysis can help the system identify users' unlocking patterns, thereby providing a basis for subsequent personalized services and security strategies. If a user unlocks at a fixed time every night, the system can prepare in advance and optimize the unlocking response. Analyzing the distribution characteristics of unlocking frequency can also help detect abnormal usage patterns. If a user's unlocking frequency is significantly different from past behavior patterns, the system can identify and trigger early warning measures in a timely manner to avoid potential safety hazards. Unlocking frequency distribution analysis can help the system more accurately understand the unlocking needs of each user in different time periods. Extracting unlocking timestamps can accurately record the specific time of each unlocking. The system can analyze users' unlocking time habits based on these timestamps. This data helps to identify users' unlocking needs in different time periods (for example: during the day, at night, or during specific periods of time). ), through timestamps, the system can refine the user's unlocking behavior. For example, some users may only unlock in the morning or at night, while other users may frequently unlock at specific times on weekdays. Through time analysis, the system can provide more accurate services for different users. By analyzing the unlocking timestamp, the system can extract the typical unlocking time period for each user and form the user's unlocking time characteristics with these time periods. A user usually unlocks at 8 o'clock in the evening. The system can use this time period as the user's time characteristic to provide customized services. Understanding the user's unlocking time characteristics can help the system provide more timely and accurate services when the user unlocks. Before the user's common unlocking time, the system can pre-optimize the Bluetooth connection to reduce the waiting time when unlocking. Through the combined analysis of the unlocking frequency distribution and the unlocking time characteristics, the system can establish personalized unlocking behavior habit characteristics for each user. These characteristics include not only the user's unlocking frequency, but also the analysis of multiple dimensions such as unlocking time, unlocking method, and device usage habits.

[0028] Preferably, the specific steps of step S2 are:

[0029] Step S21: Detecting the current user's ground lock usage behavior data according to the real-time Bluetooth connection status parameters of the smart ground lock;

[0030] Step S22: performing real-time user behavior mining on the current user's ground lock usage behavior data to generate real-time user ground lock usage behavior features;

[0031] Step S23: quantifying the abnormal behavior deviation of the real-time user's lock usage behavior characteristics based on the unlocking behavior portrait of each user, and generating a real-time user behavior abnormal deviation value;

[0032] Step S24: When the real-time user behavior abnormal deviation value exceeds the preset behavior deviation threshold, a ground lock abnormality warning signal is generated;

[0033] Step S25: making an abnormal ground lock usage decision based on the abnormal ground lock warning signal, and generating an abnormal ground lock warning strategy.

[0034] The present invention uses the real-time Bluetooth connection status parameters of the smart lock, and the system can timely capture the current user's lock usage data. These data include the user's connection status, unlocking frequency, usage time period and other information, which provides a basis for subsequent behavior analysis and decision-making. The real-time Bluetooth connection status parameters help the system identify whether it is a legitimate user operation, avoiding illegal operations of unauthorized devices or users. This data reflects the device status in real time and enhances the accuracy of monitoring. Through real-time mining of the user's lock usage behavior data, the system can dynamically understand the user's usage habits and behavior changes. This process can timely capture the user's unlocking frequency, time period, device usage and other key features. Real-time behavior mining enables the system to flexibly adapt to the needs of different users, continuously adjust the unlocking strategy and permission control, thereby improving user experience and management efficiency. The quantitative deviation value provides a more accurate metric for anomaly detection. If the user's behavior deviates greatly in a certain time period or condition, the system can immediately identify and further process it. By quantifying the user's behavior deviation, the system can respond in real time to situations that do not conform to normal behavior patterns, thereby effectively improving security. If the user unlocks at an unusual place or time, the system can warn and start an additional verification mechanism. When the real-time user When the user behavior abnormal deviation value exceeds the preset threshold, the system can automatically generate an abnormal warning signal for the ground lock, which can quickly respond to potential security threats, such as password leakage, illegal unlocking and other risks. Real-time monitoring and generating warning signals can help intervene in the early stage of abnormal behavior, which provides administrators with more time to deal with potential security incidents and avoid losses. By setting reasonable thresholds and alarm mechanisms, the system can efficiently handle emergencies and enhance the system's emergency response capabilities. If a user's unlocking behavior shows great abnormality, the system can quickly issue an alarm and require identity verification. When the ground lock generates an abnormal warning signal, the system can automatically make abnormal usage decisions based on these signals. The system can choose to temporarily disable the ground lock, require additional identity verification, or send an alarm message to the administrator. By generating an abnormal warning strategy, the system can quickly take action when abnormal behavior occurs to ensure the security of the smart ground lock system, which includes security measures such as restricting user operations and enhancing identity verification. The system can generate customized warning strategies for different types of abnormal behaviors. If the system detects that an uncommon device is trying to unlock, it may require the user to perform fingerprint verification; if it is a frequent wrong attempt, the system may lock the ground lock and notify the administrator.

[0035] Preferably, step S3 specifically comprises the following steps:

[0036] Step S31: When the real-time user behavior abnormal deviation value does not exceed the preset behavior deviation threshold, the real-time Bluetooth connection state parameter of the smart lock is subjected to abnormal outlier data filtering to obtain outlier optimized Bluetooth connection state parameter;

[0037] Step S32: calculating the current user Bluetooth connection strength for the outlier optimized Bluetooth connection state parameter to generate a user Bluetooth connection strength parameter;

[0038] Step S33: performing triangulation positioning and ranging on the user's Bluetooth connection strength parameter to obtain the distance between the current user and the ground lock;

[0039] Step S34: Perform precise positioning calculation of the user according to the current distance between the user and the ground lock to obtain the user's real-time position coordinates.

[0040] By filtering abnormal outlier data, the system can eliminate erroneous data caused by signal interference, equipment failure or other abnormal reasons, thereby improving the quality and accuracy of the data. This ensures the reliability of subsequent calculations and avoids misjudgments and wrong decisions. The Bluetooth connection of the smart ground lock may be affected by environmental factors (such as building structure, equipment interference, etc.), resulting in an unstable connection state. By filtering outlier data, the system can ensure that only stable and reliable connection status data is used for subsequent analysis to reduce noise interference. Bluetooth connection strength is a key parameter for evaluating the stability of communication between the device and the smart ground lock. By calculating the Bluetooth connection strength, the system can understand the connection quality between the current user and the ground lock in real time. Higher strength usually means a closer distance or better signal quality, while lower strength may indicate an unstable connection or a longer distance. Through accurate connection strength calculation, the system can judge the reliability of the connection and take appropriate measures (such as retrying the connection, optimizing the signal or adjusting the communication strategy) when the user's connection is unstable, thereby improving the user's experience and reducing unlocking failures caused by poor connection. Triangulation positioning ranging technology can calculate the exact distance between the user and the ground lock based on the Bluetooth signal strength of multiple known locations. Through this technology, the system can determine the relative position of the user and the ground lock in real time and improve the accuracy of positioning. By combining Bluetooth strength with triangulation positioning, the system can quickly obtain real-time data on the user's distance from the ground lock, and can respond quickly based on this data (such as automatic unlocking, notification, etc.). This provides users with a convenient user experience and avoids waiting time caused by unclear location. Compared with only based on a single Bluetooth signal strength, the triangulation positioning method reduces errors through multi-point measurement, ensuring that the distance calculation between the user and the ground lock is more accurate. Especially in complex environments (such as multiple obstacles or multipath propagation of signals), triangulation positioning can effectively overcome the limitations of Bluetooth single distance measurement. By combining triangulation positioning and Bluetooth signal strength, the system can calculate the user's real-time location coordinates very accurately. This means that the smart ground lock can accurately know the user's location, thereby achieving accurate unlocking operations and service optimization. The user's real-time location coordinates can help the smart ground lock system prepare for unlocking operations in advance when the user approaches, reducing waiting time, while ensuring that unlocking is only performed when the user reaches the predetermined location. This not only improves security, but also enhances the smoothness of operation.

[0041] Preferably, the specific steps of step S4 are:

[0042] Step S41: Perform multi-time point unlocking time analysis on the unlocking behavior portrait of each user to obtain the multi-time point unlocking features of each user;

[0043] Step S42: predicting the user unlocking time period based on the multi-time point unlocking feature of each user, and generating the user predicted unlocking time period;

[0044] Step S43: performing connection identification for the user predicted unlocking time period according to the real-time Bluetooth connection status parameters of the smart ground lock, and when a user Bluetooth connection is detected within the user predicted unlocking time period, performing pre-unlocking preparation for the smart ground lock, and generating a ground lock pre-unlocking strategy.

[0045] By analyzing the unlocking behavior of users at multiple time points, the present invention can identify the unlocking habits of users in different time periods. A certain user may unlock more frequently in the morning and evening, but less frequently in other time periods. In this way, the system can better understand the personalized behavior patterns of users. The extraction of multi-time point unlocking features enables the system to infer the user's future unlocking needs based on historical behaviors. This prediction is not limited to a certain time point, but covers the user's behavioral patterns in multiple time periods, which helps to make more accurate responses in future unlocking requests. Based on the unlocking features of multiple time points, the system can more accurately predict the unlocking time period of each user. This time period prediction can identify the user's high-frequency unlocking time period and prepare optimized unlocking strategies for these time periods. By identifying the predicted unlocking time period, the system can schedule resources in advance, such as optimizing Bluetooth connections, accelerating unlocking responses, etc., to improve the efficiency of the system. This also helps to improve the response speed of smart locks in high-demand periods (such as morning or evening) to avoid users waiting for too long during peak hours. By predicting the user's unlocking time period, the system can prepare related equipment and services in advance. , reducing the user's waiting time. If the system predicts that the user will unlock at 8 am, the system can establish a Bluetooth connection in advance to ensure that the unlocking process is quick and without delay. The system can detect in real time whether the user is connected to the smart ground lock via Bluetooth within the predicted unlocking time period. When the Bluetooth connection between the user's device and the ground lock is detected, the system can start the pre-unlocking preparation in time to ensure that the user can quickly enter the unlocking process. By performing pre-unlocking preparation in advance (through Bluetooth connection optimization, identity authentication, etc.), the system can quickly complete the unlocking operation when the user arrives at the unlocking area, reducing the user's waiting time. This improvement in response speed can greatly improve the user's experience. Outside the predicted unlocking time period, the system will not perform too much pre-unlocking preparation, avoiding unnecessary calculations and resource waste. The system can intelligently identify when to perform pre-unlocking preparation, thereby optimizing energy use and system resources. Pre-unlocking preparation can not only speed up the unlocking speed, but also improve security by combining the identity authentication of Bluetooth devices. By identifying legitimate devices through Bluetooth, the system can ensure that only authorized users can unlock within the predicted unlocking time period, thereby enhancing security protection.

[0046] Preferably, the specific steps of step S5 are:

[0047] Step S51: define the safe unlocking range of the smart ground lock;

[0048] Step S52: performing time series position fitting on the user's real-time position coordinates to generate a user time series position sequence;

[0049] Step S53: unlock and identify the user time sequence position sequence according to the safe unlocking range of the smart ground lock. When the user time sequence position sequence enters the safe unlocking range of the smart ground lock, quickly unlock the smart ground lock according to the ground lock pre-unlocking strategy to generate quick unlocking response data.

[0050] The present invention ensures that the unlocking operation is only performed within a specific, preset safety range by defining the safe unlocking range of the intelligent ground lock. This helps prevent malicious attacks and unauthorized unlocking operations. A ground lock may only be allowed to be unlocked in a specific area (such as the user's home or parking lot), but not in other unsafe places, which enhances the physical security of the system. By clarifying the safe unlocking range, the system can accurately prepare for pre-unlocking when the user approaches the ground lock, improving the response speed and fluency of unlocking. If the user is in an unsafe area, the ground lock will not respond, avoiding unnecessary attempts and waiting. By performing time-series fitting on the user's real-time location coordinates, the system can accurately track the user's movement trajectory. Position fitting can smooth the user's trajectory data, reduce the inaccuracy caused by signal fluctuations or measurement errors, and enable the system to perform position prediction and behavior analysis more stably. The generation of a time-series position sequence can help the system analyze the user's movement pattern and predict the location it may reach next. In this way, the system can prepare for the unlocking operation in advance and make pre-unlocking preparations in advance when the user approaches the safe unlocking range, thereby improving efficiency. By tracking the user's time-series position sequence in real time, the system can instantly identify whether the user has entered the safe unlocking range. Once the user's current location enters the set unlocking area, the system can quickly identify and start the unlocking process. This real-time performance greatly shortens the user's waiting time and improves the user experience. According to the user's time-series position sequence and the safe unlocking range, the system can start the pre-unlocking strategy in advance, optimize Bluetooth connection, strengthen signal reception, etc. This ensures that the user can quickly complete the unlocking operation when they reach the unlocking range, avoiding wasting time waiting for connection or verification. Through the intelligent ground lock pre-unlocking strategy, the system can adjust the unlocking strategy according to the user's approaching position. When the user approaches, the system can unlock in advance to ensure that it is immediately available when the user enters the unlocking range. This not only improves efficiency, but also ensures a smooth user experience. Only when the user enters the safe unlocking range will the system perform a quick unlocking process to prevent misoperation and security vulnerabilities. By combining the time-series position sequence and the unlocking range, the system can more accurately determine the user's true location, thereby enhancing security.

[0051] Preferably, step S6 comprises the following steps:

[0052] Step S61: performing multiple unlock response time calculations on the quick unlock response data to obtain a time step for each quick unlock response;

[0053] Step S62: analyzing the response delay of each quick unlocking response time step at different locations based on the user's real-time location coordinates, thereby obtaining unlocking response delay data at different user locations;

[0054] Step S63: optimizing the Bluetooth communication parameters of the smart lock based on the unlocking response delay data of different user locations to generate delay-optimized Bluetooth parameters;

[0055] Step S64: Perform collaborative reinforcement learning on the delayed tuning Bluetooth parameters and abnormal ground lock warning strategy to build an intelligent ground lock management model to perform intelligent ground lock remote management operations.

[0056] By calculating multiple unlocking response times, the present invention can quantify the performance of the intelligent ground lock unlocking operation and obtain the response time step of each unlocking, which is the basis for evaluating the unlocking efficiency of the intelligent ground lock and providing data support for subsequent optimization. By calculating the response time multiple times, the system can find bottlenecks or delay factors that may exist in the unlocking process. Some operations may be slower than other operations, or the unlocking response time may be longer in some cases. Identifying these bottlenecks can help to perform targeted optimization in subsequent steps. By combining the user's real-time location coordinates and analyzing the unlocking response delays at different locations, the system can identify the spatial distribution characteristics of the user's unlocking response. The user may have a shorter response time when approaching the ground lock, while Delays may occur when you are far away from the ground lock, which enables the system to optimize the unlock response according to the user's location. Through this analysis, the system can accurately calculate the unlock response time in different locations, thereby identifying and optimizing the response delay in a specific location. This is especially important for applications in diverse environments (such as homes, offices, large commercial areas, etc.), and can provide differentiated unlock optimization strategies. By analyzing the unlock response delay data, the system can identify the weaknesses of the Bluetooth signal in different locations, and then tune the Bluetooth communication parameters. If a high delay is found in a certain location, the system can adjust the Bluetooth signal strength, frequency, power consumption and other parameters to improve signal quality and communication stability, reduce delays, and optimize By optimizing Bluetooth parameters, the system can reduce communication delays caused by unstable signals or interference, thereby improving the unlocking response speed, which is crucial for efficient and secure smart ground lock management, especially in environments with a large number of devices and users. By combining delay-tuned Bluetooth parameters with abnormal ground lock early warning strategies, the system can perform reinforcement learning. The goal of reinforcement learning is to continuously adjust and optimize system parameters based on real-time feedback, making smart ground lock management more intelligent and automated. The system can adjust parameters based on the results of each unlocking operation to continuously improve the performance of the system. Through collaborative reinforcement learning, the system can gradually learn and optimize the unlocking process in different environments and usage scenarios. If Bluetooth communication often occurs at a certain location, Signal delay, the system will gradually optimize the Bluetooth connection strategy through the reinforcement learning mechanism, reduce the response delay at that location, and improve the unlocking efficiency. Combined with the abnormal ground lock warning strategy, the system can predict and identify potential equipment failures, communication problems or security risks in advance. Through reinforcement learning, the system can continuously improve the warning strategy to make it more sensitive and accurate, so as to take countermeasures in advance and avoid unnecessary failures and security incidents. By establishing a self-learning and optimized management model, the system can realize the automation of remote management operations, which means that the smart ground lock can not only adjust the unlocking strategy in real time according to user behavior, but also optimize the management strategy through learning, reduce manual intervention, and improve the efficiency and accuracy of remote management.

[0057] In this specification, a smart ground lock remote management system based on Bluetooth communication is provided, which is used to execute the smart ground lock remote management method based on Bluetooth communication as described above, including:

[0058] The user portrait module is used to obtain the operation log of the smart ground lock and the real-time Bluetooth connection status parameters of the smart ground lock; the user's personalized behavior habits are evolved according to the operation log of the smart ground lock, and user portrait modeling is performed to build the unlocking behavior portrait of each user;

[0059] The abnormal warning module is used to quantify the abnormal deviation of user behavior and abnormal lock usage decisions based on the unlocking behavior profile of each user, and generate abnormal lock warning strategies;

[0060] The user positioning module is used to perform triangulation and distance measurement on the real-time Bluetooth connection status parameters of the smart lock and user precise positioning calculation to obtain the user's real-time location coordinates when the real-time user behavior abnormal deviation value does not exceed the preset behavior deviation threshold;

[0061] The unlocking time prediction module is used to predict the user unlocking time period based on the unlocking behavior profile of each user and generate a pre-unlocking strategy for the ground lock;

[0062] The quick unlocking module is used to unlock and identify the user's real-time location coordinates according to the ground lock pre-unlocking strategy, and perform quick unlocking processing to generate quick unlocking response data;

[0063] The collaborative reinforcement module is used to tune the Bluetooth communication parameters for the quick unlock response data, and to perform collaborative reinforcement learning based on the abnormal ground lock early warning strategy to build an intelligent ground lock management model to perform remote management operations of the intelligent ground lock.

[0064] The present invention collects and analyzes the operation log and Bluetooth connection data of the smart lock. The user portrait module can establish the unlocking behavior habit characteristics of each user, and perform personalized processing based on the historical data of each user to ensure a more accurate and customized unlocking experience. The unlocking behavior portrait of each user not only reflects their unlocking habits, but also helps the system identify the user's behavioral deviations. Through the evolution analysis of the behavioral data, the system can intelligently predict the user's needs, thereby providing personalized unlocking strategies. Through real-time monitoring of the user's unlocking behavior, the system can timely identify potential abnormal behaviors, such as unauthorized unlocking attempts or location anomalies. The abnormal warning module can generate a warning signal before the abnormality occurs by quantifying the user's behavioral deviations, thereby enhancing the security of the smart lock. Through the combination of Bluetooth signal strength, distance and location data, the user positioning module can accurately determine the user's real-time location. Accurate positioning ensures that the smart lock can perform unlocking operations based on the user's exact location to avoid misoperation or misidentification. Through accurate user positioning, the system can adjust the unlocking strategy in real time, such as preparing for unlocking in advance when the user approaches the door lock, reducing waiting time and improving user experience. By predicting the user's unlocking time period, the smart lock can start the unlocking process in advance to reduce user waiting time. Time, when the user approaches to unlock, the ground lock will prepare in advance to improve the response speed of unlocking. The prediction of the unlocking time period can not only speed up the unlocking response, but also optimize the energy use of the ground lock. The ground lock can reduce power consumption during non-unlocking peak periods and reduce unnecessary resource consumption. According to the user's unlocking time period, the system can dynamically adjust the pre-unlocking strategy according to different user behaviors to ensure that the unlocking operation is performed at the right time and enhance the adaptive ability of the smart ground lock. Through the fast unlocking strategy, when the user approaches the smart ground lock, the ground lock can immediately respond to the unlocking request, greatly reducing the waiting time and improving the user's operating experience. Fast unlocking not only improves the user experience, but also ensures that only at the right time and location can the user successfully unlock, avoiding improper unlocking operations. By optimizing the unlocking response process, the system can maintain an efficient response speed in different usage environments and work stably even in complex signal environments. By analyzing the fast unlocking response data, the collaborative reinforcement module can perform real-time tuning of Bluetooth communication parameters, reduce communication delays and instability, and ensure a smooth fast unlocking process. Using reinforcement learning, the system can continuously adjust and optimize Bluetooth communication parameters and early warning strategies. Each feedback and optimization enables the performance of the smart ground lock to be continuously improved, achieving higher management and unlocking efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 A schematic flow chart of the steps of a remote management method of an intelligent ground lock based on Bluetooth communication according to the present invention;

[0066] Figure 2Detailed implementation flow chart of step S1;

[0067] Figure 3 Detailed implementation flow chart of step S2;

[0068] Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0069] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0070] The present application example provides a method and system for remote management of smart ground locks based on Bluetooth communication. The execution subject of the method and system for remote management of smart ground locks based on Bluetooth communication includes but is not limited to: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio and image management system, an information management system, and a cloud data management system.

[0071] See also Figures 1 to 4 The present invention provides a remote management method for a smart lock based on Bluetooth communication, and the remote management method for a smart lock based on Bluetooth communication comprises the following steps:

[0072] Step S1: Obtain the operation log of the smart ground lock and the real-time Bluetooth connection status parameters of the smart ground lock; evolve the user's personalized behavior habits according to the operation log of the smart ground lock, and perform user portrait modeling to build an unlocking behavior portrait of each user;

[0073] Step S2: Based on the unlocking behavior profile of each user, the real-time Bluetooth connection status parameters of the smart lock are quantified for abnormal user behavior deviations and abnormal lock usage decisions, and abnormal lock warning strategies are generated;

[0074] Step S3: When the real-time user behavior abnormal deviation value does not exceed the preset behavior deviation threshold, the real-time Bluetooth connection state parameters of the smart lock are triangulated and the user is accurately positioned to obtain the user's real-time location coordinates;

[0075] Step S4: predicting the user unlocking time period based on the unlocking behavior profile of each user, and generating a pre-unlocking strategy for the ground lock;

[0076] Step S5: unlocking and identifying the user's real-time location coordinates according to the ground lock pre-unlocking strategy, and performing a quick unlocking process to generate quick unlocking response data;

[0077] Step S6: Bluetooth communication parameters are tuned for the quick unlock response data, and collaborative reinforcement learning is performed based on the abnormal ground lock warning strategy to build an intelligent ground lock management model to perform remote management operations for the intelligent ground lock.

[0078] The present invention can accurately capture the usage habits, behavior patterns and preferences of each user by collecting the operation log and real-time Bluetooth connection status parameters of the smart lock. The daily unlocking time, frequency, unlocking device, etc. of a certain user. Based on these data, a personalized unlocking behavior portrait can be established for each user. Over time, the user's behavior pattern may change. By continuously updating and optimizing the user portrait, it is possible to better adapt to the evolution of the user's behavior habits and ensure that the system can be consistent with user needs. By comparing the user's unlocking behavior portrait with the real-time data, the deviation of the user's behavior from the normal mode can be discovered in time. These abnormal deviations may represent security risks (such as abnormal unlocking time, location or device, etc.). By quantifying the abnormal deviation of user behavior, corresponding abnormal lock usage decisions can be generated. If an abnormal unlocking time of the user is detected, the system can trigger an early warning and take corresponding security measures (such as mandatory identity authentication). Triangulation positioning is performed through real-time Bluetooth connection status parameters to locate the user's position in the actual scene. Through data such as Bluetooth signal strength, the system can provide high-precision user location judgment. Real-time positioning can help determine whether the user is in a normal unlocking scene. If the user's position is in an abnormal area, the system can trigger additional identity authentication or restrict unlocking. By analyzing the user's unlocking time period and behavior pattern, the system can predict the user's unlocking demand time. If the system knows that the user usually unlocks in a certain time period, the corresponding unlocking strategy can be prepared in advance. By predicting the user's needs in advance, the system can shorten the response time, reduce the user's waiting time, and provide a smoother experience. By combining the user's location and pre-unlocking strategy, the system can achieve fast unlocking. When the user approaches the ground lock, the system can automatically identify and start the unlocking process without the user having to do too much operation. Users do not need to perform tedious identity authentication or wait for unlocking operations, which greatly improves the convenience of use. Especially in emergency or high-demand scenarios, fast unlocking is particularly important. Bluetooth communication parameter tuning can improve the connection stability between devices, especially in environments with large signal interference. By adjusting the communication parameters, the smart ground lock can ensure efficient and stable Bluetooth connection. Through collaborative reinforcement learning, the system can continuously adjust and optimize the management strategy of the ground lock based on historical data and real-time feedback. This not only helps to improve security, but also improves the response speed and accuracy of the device. Through the intelligent ground lock management model, the system can achieve remote management of the ground lock, including unlocking history query, device status monitoring, remote control and other functions. Administrators can manage multiple ground locks from anywhere to improve work efficiency and convenience.

[0079] In the embodiment of the present invention, refer to Figure 1 , is a flowchart of a method for remote management of a smart lock based on Bluetooth communication of the present invention. In this example, the steps of the method for remote management of a smart lock based on Bluetooth communication include:

[0080] Step S1: Obtain the operation log of the smart ground lock and the real-time Bluetooth connection status parameters of the smart ground lock; evolve the user's personalized behavior habits according to the operation log of the smart ground lock, and perform user portrait modeling to build an unlocking behavior portrait of each user;

[0081] In this embodiment, the operation log source of the smart lock is determined, including unlocking events, connection status, fault records, etc. The log recording frequency is set to automatically record each time the lock is unlocked and each time the connection status changes to ensure the real-time and accuracy of the data. The operation log should contain key fields such as user ID, unlocking time, unlocking method (such as fingerprint, password), connection status (success / failure), signal strength, etc. The connection status parameters are obtained in real time through the Bluetooth module of the smart lock. The collection frequency is set to once every 5 seconds to record whether the connection is successful or not and signal strength and other information. The collected log data and Bluetooth connection status parameters are stored in the data In the database, a structured storage method is used to ensure the convenience of subsequent analysis. The acquired log data is cleaned, redundant and invalid data is removed, and data with abnormal signal strength (such as less than -100dBm) or repeated records are excluded to ensure data quality. The data is formatted into a unified structure for subsequent analysis. The timestamp is converted to a standard format, and the unlocking method and other fields are encoded to facilitate the use of machine learning models. Based on the user's unlocking log, the user's behavioral characteristics are extracted, such as daily unlocking frequency, unlocking time preference, common unlocking methods, etc. Statistical analysis methods (such as mean, standard deviation) can be used to identify the user's behavior. The user's unlocking behavior is grouped into patterns using clustering algorithms (such as K-Means or DBSCAN), and different user behavior groups are identified. Users are divided into high-frequency unlocking users and low-frequency unlocking users to facilitate subsequent personalized management. According to the extracted behavioral features, appropriate features are selected for user portrait modeling. The set features include unlocking frequency, unlocking time, unlocking method, device connection strength, etc., and a personalized unlocking behavior portrait is generated for each user. Multi-dimensional features are used to describe user behavior. User portraits can include features such as "users unlocking during peak hours" and "tending to use fingerprint unlocking". The user portrait database table structure is designed to ensure that the multi-dimensional portrait features of each user can be stored. The set fields include user ID, unlocking frequency, unlocking preference, and the last unlocking time. The generated user portrait data is inserted into the database to ensure that the user portrait can be updated after each user behavior evolution to maintain the real-time nature of the data. The user portrait is used to recommend personalized services, such as providing quick services or safety tips when users unlock, to improve user experience. Combined with user portraits, the user's unlocking behavior is monitored, abnormal behavior is identified in a timely manner, and optimized. If the unlocking frequency of a user has dropped significantly recently, the system can actively push safety tips or check the device status.

[0082] Step S2: Based on the unlocking behavior profile of each user, the real-time Bluetooth connection status parameters of the smart lock are quantified for abnormal user behavior deviations and abnormal lock usage decisions, and abnormal lock warning strategies are generated;

[0083] In this embodiment, the user's unlocking behavior portrait data is collected, including unlocking frequency, unlocking method, common unlocking time period, etc. At the same time, the real-time Bluetooth connection status parameters of the smart lock, such as signal strength, connection status, etc., are obtained. The data collection frequency is set to once per minute to ensure real-time performance. It is assumed that the user portrait data contains the following features: User ID: 1, Unlocking frequency: 3 times a day Common unlocking method: fingerprint Last unlocking time: 2024-12-29 12:00:00 The collected data is cleaned to remove redundant and invalid data, such as missing values ​​and outliers. For the Bluetooth connection status, records with signal strength lower than -100dBm are removed to ensure the validity of the data. The cleaned data should be formatted into a unified structure for subsequent analysis and processing. The baseline of the user's normal behavior is defined, usually based on the unlocking behavior features extracted from the user portrait. The user's normal unlocking frequency is set to 3 times a day, and the normal signal strength is -70dBm to -50dBm. An abnormal deviation threshold is set. If the user If the actual unlocking frequency is less than 2 times or the signal strength is less than -85dBm, it is considered a potential abnormality. The user's unlocking behavior and Bluetooth connection status are monitored in real time, and the deviation from the normal behavior baseline is calculated. When the calculated deviation value exceeds the set threshold, the behavior is marked as abnormal. If the user's real-time unlocking frequency is 1 time and the signal strength is -90dBm, an abnormal alarm will be triggered. Based on the deviation quantification results, abnormal usage decision rules are formulated. If the user's unlocking behavior is abnormal and the signal strength is low, the system can choose to take the following measures: notify the user through APP or SMS to remind him to pay attention to the status of the ground lock. If the abnormality persists, limit the user's unlocking authority and require re-verification of identity (such as entering a password or using an alternative unlocking method). According to the decision rules, a specific abnormal ground lock early warning strategy is generated. The early warning strategy should include: defining different warning levels (such as low, medium, and high), classifying according to the severity of the deviation, and formulating different response measures for different levels of warnings, such as automatically notifying users and limiting unlocking permissions.

[0084] Step S3: When the real-time user behavior abnormal deviation value does not exceed the preset behavior deviation threshold, the real-time Bluetooth connection state parameters of the smart lock are triangulated and the user is accurately positioned to obtain the user's real-time location coordinates;

[0085] In this embodiment, the user's real-time Bluetooth connection status parameters, especially the RSSI (received signal strength indication) value, are collected. The collection frequency is set to once every 2 seconds to ensure real-time performance. This data should include the user ID, timestamp and corresponding RSSI value. The known position coordinates of the smart ground lock are obtained. Assume that the coordinates of the ground lock are (34.0522, -118.2437). At the same time, considering environmental factors, the coordinates and signal strength information of other reference signal sources (such as other Bluetooth devices) required for triangulation positioning are recorded. The RSSI value is used to calculate the distance between the user and the ground lock. The distance between the user and the ground lock is calculated by the RSSI value obtained in real time. In order to improve the positioning accuracy, it is recommended to use multiple Bluetooth signal sources for triangulation positioning. Assume that there are two other Bluetooth devices with known positions in the scene (Bluetooth base stations A and B), and their coordinates are (34.0525, -118. 2440) and (34.0519, -118.2435). The distance between the user and each base station is calculated by calculating the RSSI value of each base station. The user's precise position is calculated based on the three known coordinates and the corresponding distance values ​​using the triangulation positioning algorithm. The least squares method (LeastSquaresMethod) can be used to solve the user's coordinates. The calculated user's real-time position coordinates ((34.0523, -118.2441)) are stored in the database for subsequent use and analysis. At the same time, real-time position updates are pushed to the user's device to ensure that the user can understand its location status in a timely manner. Multiple experiments are conducted in different environments (such as indoors and outdoors), and the user's RSSI value and the calculated distance and position coordinates are recorded to verify the accuracy and stability of the algorithm. Different signal source positions are set to test their impact on positioning accuracy.

[0086] Step S4: predicting the user unlocking time period based on the unlocking behavior profile of each user, and generating a pre-unlocking strategy for the ground lock;

[0087] In this embodiment, the user's past unlocking records are collected, including unlocking time, unlocking method, unlocking frequency, etc. The data range is set to unlocking events within the past 30 days to obtain sufficient sample data for analysis. Feature extraction is performed on the unlocking time to extract specific attributes of the unlocking time, such as hours, days of the week, whether it is a holiday, etc. These features will be used in subsequent time period prediction models. Select a suitable time series prediction model or classification model to predict the unlocking time period. The models that can be selected include linear regression, decision tree, random forest or deep learning model (such as LSTM), and the specific selection depends on the complexity of the data and the prediction accuracy requirements. Assume that the random forest model is selected because it performs better in dealing with nonlinear relationships. The extracted feature data is divided into a training set and a test set (for example, 80% training and 20% testing), and the model is trained using the training set. During the training process, hyperparameters are set and model performance is optimized through cross-validation. The predictive ability of the model is monitored by evaluation indicators (such as mean square error MSE). Using the trained model, the user's future unlocking time period is predicted based on his historical unlocking behavior. The model may predict that the user has a higher probability of unlocking at 8 am and 6 pm every day. According to the predicted unlocking time period, formulate a corresponding pre-unlocking strategy. The pre-unlocking strategy should include automatically preparing the device to unlock when the user approaches the unlocking time, such as opening the Bluetooth connection in advance, preloading user data, etc. Set a pre-unlocking time window, such as automatically preparing the device 5 minutes before the user's predicted unlocking time.

[0088] When the pre-unlock strategy is implemented, the system will activate the Bluetooth module in advance according to the predicted time and reduce the unlocking delay when the user arrives. If the user is predicted to unlock at 8 am, the system will automatically open the Bluetooth connection at 7:55. In addition, the system can send reminders to users based on their historical behavior, such as "Your ground lock will be ready in 5 minutes, please arrive as soon as possible." Monitor the user unlocking experience after implementation, collect user feedback and record the difference between the actual unlocking time and the predicted time to evaluate the effectiveness of the pre-unlock strategy. Based on feedback and data, regularly adjust the prediction model and pre-unlock strategy to improve prediction accuracy and user satisfaction.

[0089] Step S5: unlocking and identifying the user's real-time location coordinates according to the ground lock pre-unlocking strategy, and performing a quick unlocking process to generate quick unlocking response data;

[0090] In this embodiment, the user's location information is monitored in real time, and the user's latitude and longitude coordinates are obtained using a Bluetooth positioning system or a GPS module. The collection frequency is set to once every 2 seconds to ensure the real-time nature of the location data. According to the user's historical unlocking behavior and pre-unlocking strategy, a "safe unlocking area" is set, and a geographical fence with a radius of 5 meters is set to ensure that the user is within a safe range. Check whether the user's real-time location is within the safe area. If the user's location is within the set range, prepare to unlock and identify. Unlock and identify based on the user's real-time location coordinates and the known coordinates of the ground lock, calculate the distance between the user and the ground lock, and once it is confirmed that the user is in the safe area, the system will automatically perform a quick unlocking operation. This process includes: when the user approaches the ground lock, the system automatically activates the Bluetooth connection to ensure a quick response, and performs identity authentication according to the preset unlocking method (such as fingerprint, password). If the user uses a fingerprint to unlock, the fingerprint is automatically read for matching. After the verification is passed, the ground lock immediately performs the unlocking operation to reduce the user's waiting time.

[0091] Step S6: Bluetooth communication parameters are tuned for the quick unlock response data, and collaborative reinforcement learning is performed based on the abnormal ground lock warning strategy to build an intelligent ground lock management model to perform remote management operations for the intelligent ground lock.

[0092] In this embodiment, Bluetooth communication parameters in the quick unlock response data, such as signal strength (RSSI), connection success rate, response time, etc., are collected and analyzed, and indicators are set. For example, the signal strength should be between -50dBm and -70dBm to ensure the stability of the connection. According to the analysis results, a tuning strategy for the Bluetooth communication parameters is formulated. If it is found that the signal strength of some users is low in a specific environment, it is possible to consider adjusting the Bluetooth transmission power or modifying the connection frequency. The tuning target is set to reduce the average response time to less than 4 seconds, and Bluetooth parameter tuning is implemented. Multiple experiments are conducted to verify the tuning effect. By setting different parameter combinations (such as Transmit power, connection interval, etc.), evaluate their impact on the quick unlock response time, record the results of each experiment, form a data set for subsequent analysis, and design a reinforcement learning model, including state space (user behavior, Bluetooth status), action space (adjust Bluetooth parameters, send warnings) and reward mechanism (unlock success rate, response time). The state space can include the user's real-time location and unlock behavior deviation, and the action space can be "increase transmit power" or "send warnings". Use historical quick unlock response data and abnormal warning strategies for model training, and use Q-Learning or deep reinforcement learning algorithms to optimize Strategy, set training parameters, such as learning rate, discount factor, etc. Through multiple rounds of training, the model can self-learn and optimize the decision-making process. Deploy the trained smart ground lock management model and perform real-time management operations. The model will automatically adjust Bluetooth communication parameters according to user behavior and environmental changes, optimize the unlocking process, record the results of each decision, and continuously optimize the model through the feedback mechanism to ensure that it adapts to different environments and user needs. Build a remote management system for smart ground locks and integrate the decision output of the model. The system should support real-time monitoring of user behavior, Bluetooth status and device performance. Set the interface of the management platform to display real-time data, such as current connection status, user location and abnormal warning. Through the remote management system, the status of the smart ground lock is monitored in real time. Once abnormal behavior or connection problems are found, the system immediately responds according to the warning strategy, such as adjusting Bluetooth parameters or issuing safety prompts to users. Record all management operations for subsequent analysis and optimization. Regularly evaluate the effect of the smart ground lock management model, analyze indicators such as unlocking success rate, response time, and user feedback, and continuously optimize the model and management strategy based on the evaluation results. Through methods such as A / B testing, verify the impact of different parameter settings and management strategies on user experience, and continuously improve the management intelligence level of smart ground locks.

[0093] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include:

[0094] Step S11: Obtaining the smart ground lock operation log and the real-time Bluetooth connection status parameters of the smart ground lock;

[0095] Step S12: Identify the Bluetooth connection address according to the smart lock operation log and generate multiple Bluetooth connection users;

[0096] Step S13: Based on the intelligent ground lock operation log, the ground lock unlocking behavior detection is performed on multiple Bluetooth connected users, and the unlocking behavior data of each user is extracted;

[0097] Step S14: performing personalized behavior habit evolution on the unlocking behavior data of each user to generate unlocking behavior habit features of each user;

[0098] Step S15: performing user unlocking authority formulation processing based on the unlocking behavior habit characteristics of each user to obtain the unlocking authority of each user;

[0099] Step S16: Perform user portrait modeling based on the unlocking authority of each user and the unlocking behavior habit characteristics of each user to construct an unlocking behavior portrait of each user.

[0100] In this embodiment, a log collection module is designed in the smart ground lock to regularly record the operation status and connection information. This module should be independent of the main control logic to ensure that it does not affect the normal operation of the ground lock. Use Bluetooth Low Energy (BLE) technology to monitor devices connected to the smart ground lock. Through the Bluetooth module of the smart ground lock, the address, connection status, signal strength and other parameters of the connected device are obtained in real time. Define a unique identifier (such as MAC address) for each Bluetooth connected device, and record the connection time and disconnection time. In the main controller of the ground lock, set a timer task to record the current status (such as battery power, unlocking status, alarm status, etc.) and connected device information every minute. Store this information in a local database or cloud to ensure data security and accessibility. Parse the operation log to extract the MAC address and corresponding connection time of each connected device. Build a dictionary structure of connected users to record the number of connections and the last connection time of each device. According to the connection time and frequency, determine whether the device is a common user. Set a threshold, and devices with more than 5 connections are marked as "active users". Store the identified device information in the user database to generate records containing user information. Each user record should contain information such as the device MAC address, number of connections, and last connection time. When a Bluetooth connection request is received and an unlock operation is performed, the unlock time, the MAC address of the unlocking user, and the unlocking method (such as through APP, physical key, etc.) are recorded. All unlocking events are stored in the unlocking behavior log for subsequent analysis. The unlocking event data of each user is extracted from the unlocking behavior log to construct a user unlocking behavior dataset. The dataset should include fields such as user ID, unlocking time, and unlocking method. Cluster analysis is performed on the user's unlocking behavior to identify the user's behavior pattern. The K-Means algorithm is used to classify users according to the unlocking time and method. The behavioral evolution trend of each user is recorded. If the user's unlocking frequency increases within a specific time period, it is marked as "high activity period". User behavior feature records are created, including unlocking habits (such as peak unlocking time, common unlocking methods, etc.). These features are stored in the user portrait database for subsequent access and analysis. Peak unlocking time: 09:00-10:00. Common unlocking method: APP unlocking. K-Means cluster number: set to 3, respectively "low active users", "medium active users", and "high active users". Design rules for unlocking permissions, considering the user's activity level, unlocking method, and behavioral habits. Highly active users can enjoy more flexible unlocking permissions. Permissions are evaluated based on the user's behavioral characteristics. If the user's unlocking behavior in the past 30 days is good (such as no abnormal unlocking), their unlocking permissions can be increased. If the user's unlocking behavior is abnormal, consider reducing their permissions and restricting certain unlocking methods (such as prohibiting APP unlocking). Generate an unlocking permission record for each user, including user ID, permission level, permission effective time and other information. Store the permission record in the permission management database for subsequent verification and auditing.Integrate the user's unlocking behavior and habit characteristics with the unlocking permission information to build a user portrait. Each portrait should include the user's identification information, behavior characteristics, permission level, etc. Store the user portrait in a centralized database to ensure fast retrieval and update. Regularly evaluate the accuracy of the user portrait to ensure that it reflects the current user's behavior and permission status. Update the user portrait in a timely manner based on the user's new behavior data and permission adjustments.

[0101] {

[0102] "user_id":"user123",

[0103] "unlock_habits":{

[0104] "peak_unlock_time":"09:00-10:00",

[0105] "preferred_unlock_method":"APP"

[0106] },

[0107] "access_level":"trustworthy"

[0108] }

[0109] Update user profiles monthly, incorporating newly collected behavioral data.

[0110] In this embodiment, the specific steps of step S14 are:

[0111] Calculate the periodic unlocking frequency of each user's unlocking behavior data to generate the user's periodic unlocking frequency;

[0112] Perform unlocking frequency distribution analysis on the user's periodic unlocking frequency to obtain the unlocking frequency distribution characteristics of each user;

[0113] Extract unlocking timestamp based on each user’s unlocking behavior data;

[0114] Perform ground lock unlocking time analysis on the unlocking timestamp to generate the ground lock unlocking time feature of the user;

[0115] The unlocking frequency distribution characteristics of each user and the unlocking time characteristics of the user's ground lock are personalized and evolved to generate the unlocking behavior habit characteristics of each user.

[0116] In this embodiment, relevant data including user ID, unlocking timestamp and unlocking method are extracted from the user's unlocking behavior log. Ensure that the data is complete and accurate for subsequent analysis. Determine the calculation period, such as daily, weekly or monthly. When selecting a period, the natural law of user behavior should be considered. If the user has a high unlocking frequency on certain specific days (such as weekends), it is more appropriate to select a weekly period. Set the period to "weekly", that is, Monday to Sunday is a statistical period. Store the calculation results in a database or CSV file for subsequent retrieval and analysis. Extract the unlocking frequency data from the periodic unlocking frequency table generated in the first step. Perform statistical analysis on the unlocking frequency of each user, including calculating the mean, standard deviation, maximum and minimum values, etc. Record these statistical features for subsequent analysis of user behavior patterns. Use visualization tools (such as Matplotlib, Tableau, etc.) to make histograms or box plots to show the distribution of users' unlocking frequencies. This helps to intuitively understand the central tendency and discreteness of unlocking behavior. By analyzing the chart, identify the distribution type of unlocking frequency (such as normal distribution, skewed distribution, etc.), and evaluate different types of users (such as high-frequency users, medium-frequency users, and low-frequency users). Store the results of the frequency distribution analysis and visualization charts in a report for decision makers to refer to. The results should include a detailed description of each user's unlocking frequency and its distribution characteristics. Extract the timestamps of all unlocking events from the unlocking behavior data. Ensure that the extracted timestamp information is complete and in a consistent format (such as unified UTC time).

[0117] {

[0118] "user_id":"user1",

[0119] "unlock_timestamps":[

[0120] "2024-01-01T08:00:00Z",

[0121] "2024-01-02T09:00:00Z",

[0122] "2024-01-05T11:00:00Z" ]

[0124] }

[0125] Ensure that all timestamps are in the same format for subsequent analysis. Convert all timestamps to ISO8601 format to ensure data uniformity. Store the extracted unlock timestamps in the database to form a timestamp record table for subsequent use. The record should contain the user ID and the corresponding timestamp. From the extracted unlock timestamps, analyze the unlock time distribution of each user. Record the hours and dates of the unlock times to identify the peak unlocking hours. Count the number of unlocks per hour to identify the peak unlocking time of the user in a day. Record the number of unlocks by users between 08:00–09:00. Determine the peak unlocking time for each user and find the time with the most unlocks. For each user, record multiple peak time periods (such as morning, noon, evening, etc.) to form a comprehensive unlocking time feature. You can record something like "user A's peak unlocking time is 08:00-09:00 and 17:00-18:00". Record the unlocking time features of each user in the user portrait database for subsequent analysis and personalized services. Combine the user's unlocking frequency and unlocking time features to identify the behavior pattern of each user. If the user unlocks frequently in the morning, they can be marked as "morning active users". Record behavioral habit characteristics, including peak unlocking time, common unlocking methods, and frequent unlocking days. Track the user's unlocking behavior over a long period of time to identify the evolution trend of the behavior. If the user's unlocking behavior gradually increases during certain periods, mark it as "habit evolution". Compare the user's behavioral habit characteristics with previous records, update the user portrait, and ensure the timeliness and accuracy of the data. Store the generated user unlocking behavior habit characteristics in the user database to facilitate subsequent permission management and personalized services. These features can be used for security risk assessment and user behavior prediction to provide decision support for the smart ground lock system. {

[0126] "user_id":"user1",

[0127] "unlock_habit_features":{

[0128] "peak_unlock_time":"08:00-09:00",

[0129] "frequent_unlock_days":["Monday","Tuesday"]

[0130] }

[0131] }

[0132] In this embodiment, refer to Figure 3 , is a schematic flow chart of detailed implementation steps of step S2, wherein the detailed implementation steps of step S2 include:

[0133] Step S21: Detecting the current user's ground lock usage behavior data according to the real-time Bluetooth connection status parameters of the smart ground lock;

[0134] Step S22: performing real-time user behavior mining on the current user's ground lock usage behavior data to generate real-time user ground lock usage behavior features;

[0135] Step S23: quantifying the abnormal behavior deviation of the real-time user's lock usage behavior characteristics based on the unlocking behavior portrait of each user, and generating a real-time user behavior abnormal deviation value;

[0136] Step S24: When the real-time user behavior abnormal deviation value exceeds the preset behavior deviation threshold, a ground lock abnormality warning signal is generated;

[0137] Step S25: making an abnormal ground lock usage decision based on the abnormal ground lock warning signal, and generating an abnormal ground lock warning strategy.

[0138] In this embodiment, the Bluetooth module of the smart lock is used to monitor the connection status with the user's smart device (such as a mobile phone) in real time. Assuming that each user's device will send a connection confirmation signal when the lock is turned on, the connection status parameters can be obtained through the BLE (Bluetooth Low Energy) protocol. The data collection frequency is set to once every 5 seconds, and the connection status (such as successful connection, lost connection, signal strength, etc.) and timestamp are recorded. Use Python's bluepy library to implement Bluetooth connection monitoring.

[0139] frombluepy.btleimportScanner

[0140] scanner=Scanner()

[0141] devices = scanner.scan(5.0)

[0142] Analyze the collected ground lock usage behavior data in real time to identify the user's usage pattern. Use data mining technology, especially cluster analysis (such as K-Means) to identify common usage behaviors and extract key behavior features, such as average connection time, number of connections, number of disconnections, frequency of use, etc. Set the time window to every hour and calculate the behavior features of each user.

[0143] user_behavior=df.groupby('user_id').agg({'connection_time':'mean','disconnects':'count'})

[0144] Based on historical unlocking behavior data, build an unlocking behavior profile for each user, including unlocking frequency, usage time preference and other features. Set up a deviation quantification model to calculate the deviation between real-time user behavior features and user profiles. Use Euclidean distance or Manhattan distance to quantify the deviation: import numpy asnp

[0145] deviation=np.linalg.norm(user_behavior-user_profile)

[0146] According to historical data analysis, set the threshold for abnormal behavior deviation. Set the threshold to 3 standard deviations to ensure that abnormal behavior can be effectively identified. Monitor the user's deviation value in real time. When the deviation value exceeds the set threshold, the system automatically generates an abnormal warning signal. The warning signal contains information such as user ID, deviation value, abnormal type (such as frequency abnormality, abnormal connection time), etc., and is recorded in the warning log. When receiving the abnormal warning signal, the system automatically starts the warning response mechanism, including notifying the user, recording the event, and analyzing the cause of the abnormality. Generate corresponding warning strategies based on the abnormality type. If the user frequently disconnects, it is recommended that the user check the device power or distance; if the frequency of use is abnormal, the user is prompted that there may be security risks. The generated warning strategy may include: push messages through the APP to inform the user of abnormal behavior. After the abnormal behavior continues for more than a certain period of time, the user's access to the ground lock is restricted, and identity authentication is required before unlocking.

[0147] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include:

[0148] Step S31: When the real-time user behavior abnormal deviation value does not exceed the preset behavior deviation threshold, the real-time Bluetooth connection state parameter of the smart lock is subjected to abnormal outlier data filtering to obtain outlier optimized Bluetooth connection state parameter;

[0149] Step S32: calculating the current user Bluetooth connection strength for the outlier optimized Bluetooth connection state parameter to generate a user Bluetooth connection strength parameter;

[0150] Step S33: performing triangulation positioning and ranging on the user's Bluetooth connection strength parameter to obtain the distance between the current user and the ground lock;

[0151] Step S34: Perform precise positioning calculation of the user according to the current distance between the user and the ground lock to obtain the user's real-time position coordinates.

[0152] In this embodiment, a suitable behavior deviation threshold is set according to historical user behavior data. Assuming that the user's normal connection status parameters (such as connection delay and signal strength) fluctuate within a certain range through analysis of historical data, the deviation threshold is set to ±10%. Use the Bluetooth module to collect the user's connection status parameters in real time. These parameters include signal strength (RSSI), connection delay, data transmission rate, etc. Assume that data is collected once per second and recorded in the data buffer. Implement statistical methods (such as Z-score or IQR) to detect outliers on the real-time collected data. Calculate the mean and standard deviation of each parameter, and for each latest data point, determine whether its deviation from the mean exceeds the preset threshold. If the signal strength of a certain connection is -80dBm, the average value of the current signal is calculated to be -75dBm, the standard deviation is 5dBm, and the Z-score is (−80-(−75)) / 5=−1, which does not exceed the threshold, it is normal. The data points that do not exceed the deviation threshold are retained, and the data points that exceed the threshold are filtered. The final outlier optimized Bluetooth connection status parameters are used for subsequent calculations. Use the Bluetooth connection status parameters after outlier data filtering to prepare input data for strength calculation. Including valid signal strength (RSSI) values. For the filtered signal strength data, calculate its mean and standard deviation. Assuming that the filtered signal strength data is [-70, -72, -68, -75], the mean value is calculated as follows:

[0153] Average signal strength RSSI = (-70 + -72 + -68 + -75) / 4 = -71.25dBm,

[0154] The calculated average signal strength is used as the user's Bluetooth connection strength parameter. If necessary, it can be converted into a connection quality score (such as a score from 0 to 100) to facilitate subsequent positioning calculations.

[0155] Converting RSSI values ​​to a score from 0 to 100 can use a linear mapping, for example:

[0156] Score = 100-(RSSI+100) / 2, valid signal strength data sample: [-70, -72, -68, -75].

[0157] Average signal strength: -71.25dBm.

[0158] Example of score conversion: -71.25dBm score = 100-(−71.25+100) / 2 = 85.375. Triangulation is based on the relationship between signal strength (RSSI) and distance. The distance between the user and the ground lock is calculated based on the known geometric position and signal strength. Based on the experimental data, a relationship model between signal strength and distance is established. In general, the RSSI value is negatively correlated with the distance. The formula can be used:

[0159] Distance (d) = 10^((A-RSSI) / (10*n))

[0160] Where A is the RSSI value of the signal at 1 meter, and n is the path loss factor (usually 2 to 4). Use the calculated Bluetooth connection strength (RSSI) parameter to calculate the distance. Assuming the RSSI at 1 meter is -60dBm and the path loss factor n is 2, then: Distance = 10^((-60-(-71.25)) / (10*2))≈10^(0.5625)≈3.65 meters RSSI value at 1 meter: -60dBm.

[0161] Path loss factor (n): 2.

[0162] The distance between the user and the ground lock is about 3.65 meters. Determine the fixed coordinate position of the ground lock, assuming that the coordinates of the ground lock are (X_lock, Y_lock). The coordinates of the ground lock are (100, 200) meters. Based on the distance between the user and the ground lock, use simple geometric calculations to determine the possible location of the user. The user's location can form a circle around the coordinates of the ground lock. Assuming that the direction of the user relative to the ground lock is known, the real-time location of the user is calculated by converting polar coordinates to rectangular coordinates. Use the following formula:

[0163] X_user=X_lock+distance*cos(θ)

[0164] Y_user=Y_lock+distance*sin(θ)

[0165] Where θ is the angle between the user and the ground lock. Ground lock coordinates: (100, 200) meters.

[0166] The distance between the user and the ground lock: 3.65 meters.

[0167] User position calculation example: Assume θ = 30°, then:

[0168] X_user=100+3.65*cos(30°)≈100+3.65*0.866≈103.16.

[0169] Y_user=200+3.65*sin(30°)≈200+3.65*0.5≈201.83.

[0170] The final user coordinates are approximately (103.16, 201.83).

[0171] In this embodiment, step S4 includes the following steps:

[0172] Step S41: Perform multi-time point unlocking time analysis on the unlocking behavior portrait of each user to obtain the multi-time point unlocking features of each user;

[0173] Step S42: predicting the user unlocking time period based on the multi-time point unlocking feature of each user, and generating the user predicted unlocking time period;

[0174] Step S43: performing connection identification for the user predicted unlocking time period according to the real-time Bluetooth connection status parameters of the smart ground lock, and when a user Bluetooth connection is detected within the user predicted unlocking time period, performing pre-unlocking preparation for the smart ground lock, and generating a ground lock pre-unlocking strategy.

[0175] In this embodiment, the user's unlocking behavior data is collected, including unlocking time, unlocking location, device information used, etc. The data can be obtained in real time through the log recording system of the smart lock. Assuming that the data record includes information such as user ID and unlocking timestamp, [(user_id, timestamp), (user_id, timestamp), ...], the unlocking time is classified, and the user's unlocking behavior in different time periods of the day is analyzed. A day can be divided into morning (6:00-9:00), morning (9:00-12:00), afternoon (12:00-18:00), and afternoon (12:00-18:00). Count the number of times and frequency of unlocking for each user in each time period, count the number of times a user unlocks in the morning, and calculate its proportion of all unlocking times. Use statistical analysis methods (such as mean and standard deviation) to analyze the distribution of time points and identify the peak unlocking time of users. Unlocking time data samples: [(1, '2024-01-1507:30:00'), (1, '2024-01-1508:15:00'), ...], morning (6:00-9:00), morning (9:00-12:00), afternoon (12:0 0-18:00), night (18:00-24:00), user 1 unlocks 5 times in the morning, accounting for 25% of the total unlocking times. Select a suitable time series prediction model (such as ARIMA, LSTM, etc.) to model the user's unlocking behavior. Assume that the LSTM model is used to process nonlinear time series data. Convert the user's unlocking time data into a time series format and prepare a training data set. Assume that the data format is a mapping of timestamps and unlocking times, [(timestamp1, count1), (timestamp2, count2), ...]. Use historical unlocking time data to train the model. According to the user's unlocking characteristics, input the corresponding time series data. The model learns the user's unlocking mode. During the training process, cross-validation can be used to evaluate the performance of the model to ensure the generalization ability of the model. Use the trained model to predict the future unlocking time period and generate the user's predicted unlocking time period. The prediction result can be output in the form of a time period, indicating the probability that the user may unlock in a certain time period. Prediction model selection: LSTM, [(timestamp, unlock_count), (timestamp, unlock_count), ...], the user may unlock the device between "2024-01-15 08:00-09:00" with a probability of 80%. The Bluetooth connection status of the user's device is monitored in real time to determine whether the user is connected to the smart ground lock within the predicted unlocking time period. The Bluetooth module is used to obtain connection status data and record the user's Bluetooth signal strength (RSSI) and connection status. When the user is detected to be connected to the ground lock within the predicted unlocking time period (RSSI value is higher than the set threshold), the system will recognize that the user is about to unlock the device. The threshold for connection recognition is set. RSSI value > -75dBm indicates a good connection. Once the user is identified, After Bluetooth connection, the smart lock will perform pre-unlock preparation operations, including activation of the unlocking mechanism and security checks, which can include checking the user's authentication information to ensure that only legitimate users can unlock. According to the result of connection identification, the corresponding pre-unlocking strategy is generated. If the user connects within the predicted time period, it will be unlocked in advance; if the connection is not detected, it will remain locked. The Bluetooth connection status monitoring frequency: once every 500 milliseconds, RSSI value connection identification threshold: >-75dBm, if the connection is detected in the "2024-01-1508:00-09:00" time period, the smart lock will be unlocked in advance. .

[0176] In this embodiment, step S5 includes the following steps:

[0177] Step S51: define the safe unlocking range of the smart ground lock;

[0178] Step S52: performing time series position fitting on the user's real-time position coordinates to generate a user time series position sequence;

[0179] Step S53: unlock and identify the user time sequence position sequence according to the safe unlocking range of the smart ground lock. When the user time sequence position sequence enters the safe unlocking range of the smart ground lock, quickly unlock the smart ground lock according to the ground lock pre-unlocking strategy to generate quick unlocking response data.

[0180] In this embodiment, the safe unlocking range is set according to the actual installation location of the ground lock and the surrounding environment. Generally speaking, the safe unlocking range should be a circular area with the center point being the ground lock position and the radius set according to actual needs. Set the radius to 5 meters. Obtain the precise coordinates of the ground lock through GPS or other positioning technology. Assume that the coordinates of the ground lock are (X_lock, Y_lock), for example (100, 200) meters. Use a geometric model to represent the safe unlocking range. The following formula can be used to determine whether the user is within the safe range, . Where R is the radius of the safe unlocking range (e.g. 5 meters). Ground lock coordinates: (100, 200) meters. Safe unlocking range radius: 5 meters. If the user coordinates are (102, 202), the distance calculation is: 2.83 meters, using GPS or Bluetooth positioning technology to obtain the user's location coordinates in real time. Assume that the user's coordinates are recorded once a second, and the data format is (timestamp, X_user, Y_user). [(t1, 98, 199), (t2, 99, 200), (t3, 101, 201), ...]. Organize the real-time location data into a time series to ensure data continuity and accuracy. Interpolate the user's location at each time point and fill in possible missing values ​​to ensure the integrity of the time series data. Select a suitable fitting model (such as linear regression, polynomial regression, or spline interpolation) to fit the user's time series position. Use the model to smooth the location data to eliminate noise.

[0181] A linear regression model is used to fit the user's movement path. The fitted user time-series position sequence is stored in the database to facilitate the subsequent unlocking and recognition process.

[0182] In this embodiment, step S6 includes the following steps:

[0183] Step S61: performing multiple unlock response time calculations on the quick unlock response data to obtain a time step for each quick unlock response;

[0184] Step S62: analyzing the response delay of each quick unlocking response time step at different locations based on the user's real-time location coordinates, thereby obtaining unlocking response delay data at different user locations;

[0185] Step S63: optimizing the Bluetooth communication parameters of the smart lock based on the unlocking response delay data of different user locations to generate delay-optimized Bluetooth parameters;

[0186] Step S64: Perform collaborative reinforcement learning on the delayed tuning Bluetooth parameters and abnormal ground lock warning strategy to build an intelligent ground lock management model to perform intelligent ground lock remote management operations.

[0187] In this embodiment, the response data of the user's quick unlocking is collected, including the start time and end time of each unlocking. These data can be automatically generated by the log recording system. The collection frequency is set to each unlocking instance. The duration of each quick unlocking is calculated to obtain the response time step of each unlocking. Response time = unlocking end time - unlocking start time. The average response time, minimum response time and maximum response time of each user are calculated, and the distribution characteristics of the response time are identified. These statistical results will provide basic data for subsequent delay analysis. Based on the user's real-time location coordinate data and combined with the response time step, the location evaluation is performed for each quick unlocking, and the user's location information (such as longitude and latitude) and the corresponding unlocking response time are recorded. The unlocking response delay of different user locations is analyzed, and the user location is divided into multiple areas (such as city center, suburbs, etc.), and the average response time of each area is calculated. The response delay of different areas is compared by using statistical methods (such as ANOVA analysis), and the influence of location on unlocking response time is identified. Unlocking response delay data of different user locations are generated and recorded in the database for subsequent Bluetooth communication parameter tuning. Based on the analysis results, Bluetooth is set. The goal of Bluetooth communication parameter tuning is to reduce the response time to less than 1 second, especially in high-latency areas. According to the response delay data of different locations, the Bluetooth communication parameters are adjusted, such as signal strength, connection frequency and data packet size. The optimal Bluetooth communication configuration is determined through experiments to ensure stability and reliability in different environments. The tuned Bluetooth parameters are recorded to form a delay-tuned Bluetooth parameter set, including the optimal signal strength, connection interval, data packet size, etc. These parameters will provide support for the subsequent remote management of smart ground locks. According to the delay-tuned Bluetooth parameters and abnormal ground lock warning strategy, a reinforcement learning model is constructed, and the state space, action space and reward mechanism of the model are set to ensure that the model can self-learn and optimize in different environments. The collected user response data and Bluetooth parameters are used for model training. Through multiple rounds of training (such as Q-learning or deep reinforcement learning), the model can learn the optimal decision-making strategy in different situations. The trained smart ground lock management model is deployed to perform remote management operations of smart ground locks. The model will automatically monitor the user's unlocking behavior and response time, and dynamically adjust the Bluetooth communication parameters according to real-time data to ensure the best user experience and security.

[0188] In this embodiment, a smart ground lock remote management system based on Bluetooth communication is provided, which is used to execute the smart ground lock remote management method based on Bluetooth communication as described above, including:

[0189] The user portrait module is used to obtain the operation log of the smart ground lock and the real-time Bluetooth connection status parameters of the smart ground lock; the user's personalized behavior habits are evolved according to the operation log of the smart ground lock, and user portrait modeling is performed to build the unlocking behavior portrait of each user;

[0190] The abnormal warning module is used to quantify the abnormal deviation of user behavior and abnormal lock usage decisions based on the unlocking behavior profile of each user, and generate abnormal lock warning strategies;

[0191] The user positioning module is used to perform triangulation and distance measurement on the real-time Bluetooth connection status parameters of the smart lock and user precise positioning calculation to obtain the user's real-time location coordinates when the real-time user behavior abnormal deviation value does not exceed the preset behavior deviation threshold;

[0192] The unlocking time prediction module is used to predict the user unlocking time period based on the unlocking behavior profile of each user and generate a pre-unlocking strategy for the ground lock;

[0193] The quick unlocking module is used to unlock and identify the user's real-time location coordinates according to the ground lock pre-unlocking strategy, and perform quick unlocking processing to generate quick unlocking response data;

[0194] The collaborative reinforcement module is used to tune the Bluetooth communication parameters for the quick unlock response data, and to perform collaborative reinforcement learning based on the abnormal ground lock early warning strategy to build an intelligent ground lock management model to perform remote management operations of the intelligent ground lock.

[0195] The present invention collects and analyzes the operation log and Bluetooth connection data of the smart lock. The user portrait module can establish the unlocking behavior habit characteristics of each user, and perform personalized processing based on the historical data of each user to ensure a more accurate and customized unlocking experience. The unlocking behavior portrait of each user not only reflects their unlocking habits, but also helps the system identify the user's behavioral deviations. Through the evolution analysis of the behavioral data, the system can intelligently predict the user's needs, thereby providing personalized unlocking strategies. Through real-time monitoring of the user's unlocking behavior, the system can timely identify potential abnormal behaviors, such as unauthorized unlocking attempts or location anomalies. The abnormal warning module can generate a warning signal before the abnormality occurs by quantifying the user's behavioral deviations, thereby enhancing the security of the smart lock. Through the combination of Bluetooth signal strength, distance and location data, the user positioning module can accurately determine the user's real-time location. Accurate positioning ensures that the smart lock can perform unlocking operations based on the user's exact location to avoid misoperation or misidentification. Through accurate user positioning, the system can adjust the unlocking strategy in real time, such as preparing for unlocking in advance when the user approaches the door lock, reducing waiting time and improving user experience. By predicting the user's unlocking time period, the smart lock can start the unlocking process in advance to reduce user waiting time. Time, when the user approaches to unlock, the ground lock will prepare in advance to improve the response speed of unlocking. The prediction of the unlocking time period can not only speed up the unlocking response, but also optimize the energy use of the ground lock. The ground lock can reduce power consumption during non-unlocking peak periods and reduce unnecessary resource consumption. According to the user's unlocking time period, the system can dynamically adjust the pre-unlocking strategy according to different user behaviors to ensure that the unlocking operation is performed at the right time and enhance the adaptive ability of the smart ground lock. Through the fast unlocking strategy, when the user approaches the smart ground lock, the ground lock can immediately respond to the unlocking request, greatly reducing the waiting time and improving the user's operating experience. Fast unlocking not only improves the user experience, but also ensures that only at the right time and location can the user successfully unlock, avoiding improper unlocking operations. By optimizing the unlocking response process, the system can maintain an efficient response speed in different usage environments and work stably even in complex signal environments. By analyzing the fast unlocking response data, the collaborative reinforcement module can perform real-time tuning of Bluetooth communication parameters, reduce communication delays and instability, and ensure a smooth fast unlocking process. Using reinforcement learning, the system can continuously adjust and optimize Bluetooth communication parameters and early warning strategies. Each feedback and optimization enables the performance of the smart ground lock to be continuously improved, achieving higher management and unlocking efficiency.

[0196] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0197] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A remote management method for smart ground locks based on Bluetooth communication, characterized in that: The following steps are involved: Step S1: Obtain the smart ground lock operation log and the real-time Bluetooth connection status parameters of the smart ground lock; Evolve the user's personalized behavior habits based on the operation log of the smart lock, and perform user portrait modeling to build the unlocking behavior portrait of each user; Step S2: Based on the unlocking behavior profile of each user, the real-time Bluetooth connection status parameters of the smart lock are quantified for abnormal user behavior deviations and abnormal lock usage decisions, and abnormal lock warning strategies are generated; Step S3: When the real-time user behavior abnormal deviation value does not exceed the preset behavior deviation threshold, the real-time Bluetooth connection state parameters of the smart lock are triangulated and the user is accurately positioned to obtain the user's real-time location coordinates; Step S4: predicting the user unlocking time period based on the unlocking behavior profile of each user, and generating a pre-unlocking strategy for the ground lock; Step S5: unlocking and identifying the user's real-time location coordinates according to the ground lock pre-unlocking strategy, and performing a quick unlocking process to generate quick unlocking response data; Step S6: tuning the Bluetooth communication parameters for the quick unlock response data, and performing collaborative reinforcement learning based on the abnormal ground lock warning strategy to build an intelligent ground lock management model to perform remote management operations for the intelligent ground lock; Among them, the specific steps of step S1 are: Step S11: Obtaining the smart ground lock operation log and the real-time Bluetooth connection status parameters of the smart ground lock; Step S12: Identify the Bluetooth connection address according to the smart lock operation log and generate multiple Bluetooth connection users; Step S13: Based on the intelligent ground lock operation log, the ground lock unlocking behavior detection is performed on multiple Bluetooth connected users, and the unlocking behavior data of each user is extracted; Step S14: performing personalized behavior habit evolution on the unlocking behavior data of each user to generate unlocking behavior habit features of each user; Step S15: performing user unlocking authority formulation processing based on the unlocking behavior habit characteristics of each user to obtain the unlocking authority of each user; Step S16: Perform user portrait modeling based on the unlocking authority of each user and the unlocking behavior habit characteristics of each user, and construct an unlocking behavior portrait of each user; Among them, the specific steps of step S14 are: Calculate the periodic unlocking frequency of each user's unlocking behavior data to generate the user's periodic unlocking frequency; Perform unlocking frequency distribution analysis on the user's periodic unlocking frequency to obtain the unlocking frequency distribution characteristics of each user; Extract unlocking timestamp based on each user’s unlocking behavior data; Perform ground lock unlocking time analysis on the unlocking timestamp to generate the ground lock unlocking time feature of the user; The unlocking frequency distribution characteristics of each user and the unlocking time characteristics of the user's ground lock are personalized and evolved to generate the unlocking behavior habit characteristics of each user.

2. The method for remote management of smart ground locks based on Bluetooth communication according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: Detecting the current user's ground lock usage behavior data according to the real-time Bluetooth connection status parameters of the smart ground lock; Step S22: performing real-time user behavior mining on the current user's ground lock usage behavior data to generate real-time user ground lock usage behavior features; Step S23: quantifying the abnormal behavior deviation of the real-time user's lock usage behavior characteristics based on the unlocking behavior portrait of each user, and generating a real-time user behavior abnormal deviation value; Step S24: When the real-time user behavior abnormal deviation value exceeds the preset behavior deviation threshold, a ground lock abnormality warning signal is generated; Step S25: making an abnormal ground lock usage decision based on the abnormal ground lock warning signal, and generating an abnormal ground lock warning strategy.

3. The method for remote management of smart ground locks based on Bluetooth communication according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: When the real-time user behavior abnormal deviation value does not exceed the preset behavior deviation threshold, the real-time Bluetooth connection state parameter of the smart lock is subjected to abnormal outlier data filtering to obtain outlier optimized Bluetooth connection state parameter; Step S32: calculating the current user Bluetooth connection strength for the outlier optimized Bluetooth connection state parameter to generate a user Bluetooth connection strength parameter; Step S33: performing triangulation positioning and ranging on the user's Bluetooth connection strength parameter to obtain the distance between the current user and the ground lock; Step S34: Perform precise positioning calculation of the user according to the current distance between the user and the ground lock to obtain the user's real-time position coordinates.

4. The method for remote management of smart ground locks based on Bluetooth communication according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: Perform multi-time point unlocking time analysis on the unlocking behavior portrait of each user to obtain the multi-time point unlocking features of each user; Step S42: predicting the user unlocking time period based on the multi-time point unlocking feature of each user, and generating the user predicted unlocking time period; Step S43: performing connection identification for the user predicted unlocking time period according to the real-time Bluetooth connection status parameters of the smart ground lock, and when a user Bluetooth connection is detected within the user predicted unlocking time period, performing pre-unlocking preparation for the smart ground lock, and generating a ground lock pre-unlocking strategy.

5. The method for remote management of smart ground locks based on Bluetooth communication according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: define the safe unlocking range of the smart ground lock; Step S52: performing time series position fitting on the user's real-time position coordinates to generate a user time series position sequence; Step S53: unlock and identify the user time sequence position sequence according to the safe unlocking range of the smart ground lock. When the user time sequence position sequence enters the safe unlocking range of the smart ground lock, quickly unlock the smart ground lock according to the ground lock pre-unlocking strategy to generate quick unlocking response data.

6. The method for remote management of smart ground locks based on Bluetooth communication according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: performing multiple unlock response time calculations on the quick unlock response data to obtain a time step for each quick unlock response; Step S62: analyzing the response delay of each quick unlocking response time step at different locations based on the user's real-time location coordinates, thereby obtaining unlocking response delay data at different user locations; Step S63: optimizing the Bluetooth communication parameters of the smart lock based on the unlocking response delay data of different user locations to generate delay-optimized Bluetooth parameters; Step S64: Perform collaborative reinforcement learning on the delayed tuning Bluetooth parameters and abnormal ground lock warning strategy to build an intelligent ground lock management model to perform intelligent ground lock remote management operations.

7. A smart ground lock remote management system based on Bluetooth communication, characterized in that: The method for remotely managing a smart ground lock based on Bluetooth communication as claimed in claim 1 comprises: The user portrait module is used to obtain the operation log of the smart ground lock and the real-time Bluetooth connection status parameters of the smart ground lock; the user's personalized behavior habits are evolved according to the operation log of the smart ground lock, and user portrait modeling is performed to build the unlocking behavior portrait of each user; The abnormal warning module is used to quantify the abnormal deviation of user behavior and abnormal lock usage decisions based on the unlocking behavior profile of each user, and generate abnormal lock warning strategies; The user positioning module is used to perform triangulation and distance measurement on the real-time Bluetooth connection status parameters of the smart lock and user precise positioning calculation to obtain the user's real-time location coordinates when the real-time user behavior abnormal deviation value does not exceed the preset behavior deviation threshold; The unlocking time prediction module is used to predict the user unlocking time period based on the unlocking behavior profile of each user and generate a pre-unlocking strategy for the ground lock; The quick unlocking module is used to unlock and identify the user's real-time location coordinates according to the ground lock pre-unlocking strategy, and perform quick unlocking processing to generate quick unlocking response data; The collaborative reinforcement module is used to tune the Bluetooth communication parameters for the quick unlock response data, and to perform collaborative reinforcement learning based on the abnormal ground lock early warning strategy to build an intelligent ground lock management model to perform remote management operations of the intelligent ground lock.

Images