Fire-fighting equipment intelligent management method and system based on RFID and Bluetooth technology

Abstract

The present application belongs to the technical field of vehicle-mounted equipment positioning, and particularly relates to a fire-fighting equipment intelligent management system and method based on RFID and Bluetooth technologies. The system comprises an RFID tag, a Bluetooth beacon, an RFID reader, a Bluetooth receiver, and a central processor. The fire-fighting equipment is provided with the RFID tag, which is used to store and send an RFID tag signal with basic information and an identification code of the fire-fighting equipment. The Bluetooth beacon preset in the area of the fire truck sends a Bluetooth signal. The central processor is used to perform data fusion on the RFID signal received by the RFID reader and the Bluetooth signal received by the Bluetooth receiver, so as to obtain the positioning of the fire-fighting equipment. The system can realize the rapid and accurate positioning of the fire truck equipment.

Description

Technical Field

[0001] This invention belongs to the field of vehicle-mounted equipment positioning technology, specifically relating to an intelligent management system and method for fire-fighting equipment based on RFID and Bluetooth technologies. Background Technology

[0002] Fire trucks are equipped with a large number of tools and equipment to play a corresponding role in disaster relief. Fire trucks often respond to emergencies, and the situation at disaster sites is complex. The tools and equipment on fire trucks are easily forgotten or lost, causing unnecessary losses. Existing technologies have corresponding intelligent management methods for fire truck equipment. For example, a fire truck equipment status monitoring system (patent publication number: CN206103172U) relates to such a system. This monitoring system includes: active RFID tags affixed to various standard equipment on the fire truck, including a CPU controller and an RF module, an audible and visual indicator, and a battery module connected to the CPU controller. The RF module is connected to an antenna and the battery module. An RF reader is wirelessly connected to the RFID tags, receives and processes the RF signals sent by the active RFID tags, and is connected to a display screen and an alarm device, which facilitates effective monitoring of the status of the fire truck equipment.

[0003] The problem with the above solution is that although existing RFID identification technology can achieve wide-range object identification, the positioning accuracy is low, and RFID technology may be limited in complex environments or when identifying objects at long distances, resulting in unstable data transmission.

[0004] Bluetooth technology offers more precise positioning. Combining RFID identification and Bluetooth technologies allows for coarse positioning over a wide area and fine positioning in specific regions, improving overall positioning accuracy. Bluetooth also boasts advantages such as low power consumption, low cost, and good compatibility, providing more stable data transmission, especially in short-range communication scenarios. Summary of the Invention

[0005] To achieve precise positioning of fire truck equipment, this paper proposes an intelligent management system and method for fire equipment based on RFID and Bluetooth technologies, combining the wide-area coverage of RFID with the precise positioning capability of Bluetooth, to enable rapid positioning of fire equipment.

[0006] To achieve the above objectives, the following technical solutions are proposed:

[0007] The intelligent management system for fire-fighting equipment based on RFID and Bluetooth technologies includes RFID tags, Bluetooth beacons, RFID readers, Bluetooth receivers, and a central processing unit.

[0008] Firefighting equipment is equipped with RFID tags, which are used to store and transmit RFID tag signals containing basic information and identification codes of the firefighting equipment.

[0009] Bluetooth beacons positioned in the pre-designated area by the fire truck emit Bluetooth signals;

[0010] The central processing unit is used to fuse the RFID signals received by the RFID reader and the Bluetooth signals received by the Bluetooth receiver to obtain the location of the fire-fighting equipment.

[0011] Furthermore, the central processing unit also identifies, parses, and analyzes the status of fire-fighting equipment based on the received data to obtain real-time information, and updates the real-time information to the database server.

[0012] Furthermore, it also includes a data uploading unit, which transmits the data in the central processing unit to the database server via wireless transmission. The wireless transmission method includes 4G / NB-IoT wireless transmission, CAN bus, RS232, and RS485.

[0013] Based on the same concept, a smart management method for fire-fighting equipment based on RFID and Bluetooth technologies is also proposed. Using any of the above-mentioned smart management systems for fire-fighting equipment based on RFID and Bluetooth technologies, the following steps are performed:

[0014] The location of RFID tags is determined based on the collected RFID tag signals;

[0015] The location of the Bluetooth beacon is obtained based on the collected Bluetooth signals;

[0016] A data fusion algorithm is used to fuse the positioning data of RFID tags and Bluetooth beacons to obtain a fused distance estimate, and the positioning of fire-fighting equipment is calculated based on the distance estimate.

[0017] Preferably, a weighted average method is used to fuse the distance estimates of RFID and Bluetooth. The weights in the weighted average are determined based on the stability of the signal strength, the reliability of the device, and the accuracy of previous data.

[0018] Preferably, the calculation process for locating firefighting equipment includes: obtaining the known locations of the RFID reader and Bluetooth scanning device, and calculating the position coordinates of the equipment using triangulation based on the fused distance estimate and the relative angle information between the firefighting equipment and the RFID reader or Bluetooth receiver.

[0019] Preferably, signal strength correction is included before data fusion. This signal strength correction is implemented using a logarithmic distance path loss model, which is as follows:

[0020] ;

[0021] in, This is the reference signal strength at a distance of 1 meter, where n is the path loss exponent and d is the distance between the device and the reader. It follows a normal distribution Environmental noise.

[0022] Preferably, the method further includes: obtaining the movement trajectory of the fire-fighting equipment based on its location, and predicting the movement trend of the fire-fighting equipment using a dynamic path analysis algorithm.

[0023] Preferably, it further includes: an abnormal behavior recognition model based on machine learning, which determines whether the fire-fighting equipment is in an abnormal state based on the location of the fire-fighting equipment, and issues an early warning based on the abnormal state, wherein the abnormal state includes abnormal movement and prolonged stay.

[0024] Preferably, it further includes: determining the usage frequency of the fire-fighting equipment based on its location, and adjusting the working mode according to the usage frequency and the time period of use, wherein the adjusted working mode includes a sleep mode and a wake-up mode.

[0025] Compared with the prior art, the beneficial effects of the present invention are: the system and method of the present invention can realize the rapid and accurate positioning of fire truck equipment, which is suitable for scenarios that require high-efficiency and high-precision fire equipment management, and has broad market application prospects and social and economic benefits. Attached Figure Description

[0026] Figure 1 This is a system architecture diagram of the intelligent management system for fire-fighting equipment based on RFID and Bluetooth technologies in Example 1;

[0027] Figure 2 This is a flowchart of the intelligent management method for fire-fighting equipment based on RFID and Bluetooth technologies in Example 1;

[0028] Figure 3 This is a flowchart of data fusion in Example 2;

[0029] Figure 4 This is a flowchart illustrating the use of dynamic path analysis algorithm to predict the movement trend of firefighting equipment in Example 3;

[0030] Figure 5 This is the abnormal state early warning flowchart in Example 4;

[0031] Figure 6 This is a flowchart of adjusting the working mode in Example 5. Detailed Implementation

[0032] The present invention will be further described in detail below with reference to experimental examples and specific embodiments. However, this should not be construed as limiting the scope of the above-mentioned subject matter of the present invention to the following embodiments; all technologies implemented based on the content of the present invention fall within the scope of the present invention.

[0033] Example 1

[0034] The intelligent management system for fire-fighting equipment based on RFID and Bluetooth technologies has the following system architecture diagram: Figure 1 As shown,

[0035] It includes RFID tags, Bluetooth beacons, RFID readers, Bluetooth receivers, central processing units, data upload units, and database servers;

[0036] Each piece of fire-fighting equipment is attached with a unique RFID tag, which is used to store and transmit the basic information and identification code of the fire-fighting equipment.

[0037] Bluetooth beacons, deployed in key areas of fire trucks, work in conjunction with RFID tags to emit Bluetooth signals;

[0038] An RFID reader, installed inside the fire truck, is used to collect RFID tag signals and transmit the collected RFID tag signal data to the central processing unit.

[0039] Bluetooth receiver: Installed inside the fire truck, used to collect Bluetooth signals and transmit the collected Bluetooth signal data to the central processing unit;

[0040] Central Processing Unit: Processes the received data, identifies fire-fighting equipment, analyzes its location and status, obtains real-time information, and updates the real-time information to the database server;

[0041] The database server stores real-time information and historical data of all equipment for retrieval and data analysis.

[0042] It also includes a data uploading unit, which transmits the collected RFID tag signal data and Bluetooth signal data to the database server through several wireless transmission methods, including 4G / NB-IoT wireless transmission, CAN bus, RS232, and RS485.

[0043] In addition, it also includes user terminals, which include mobile devices and PC software, allowing managers to view equipment status, perform searches, generate reports, and perform other operations.

[0044] The system's workflow includes:

[0045] The RFID reader automatically scans RFID tags that enter its reading range to obtain basic equipment information.

[0046] Bluetooth beacons interact with Bluetooth receivers on fire trucks (integrated into some RFID tags or existing independently) to achieve precise positioning.

[0047] The central processing unit integrates RFID and Bluetooth data, and through algorithm processing, performs fire equipment identification, location analysis, and status analysis to achieve real-time tracking and status monitoring of the equipment.

[0048] The database server stores real-time information and historical data of all fire-fighting equipment.

[0049] In addition, user terminals can query equipment information as needed, and the system provides real-time feedback on equipment location, status, and historical movement trajectory. The system supports anomaly alarm functions, automatically notifying management personnel if equipment is moved out of a designated area without authorization.

[0050] Hardware design details:

[0051] RFID Tag Optimization: During the RFID tag selection process, to ensure reliability, it's crucial to choose tags with superior hardware attributes and reserve sufficient space for extended information beyond basic data, such as maintenance records, equipment status, and maintenance personnel information. Therefore, RFID tags are made from anti-metal, waterproof, and high-temperature resistant materials to ensure stable operation even in harsh environments. A low-power chip is embedded within the tag to store basic equipment information such as the equipment's unique ID, model, and production date, with reserved storage space for recording maintenance records and status changes. The reserved storage space is divided into four independent storage blocks: EPC area, TID area, User area, and Reserved area. At least a 512-bit User area is used to store basic equipment information, including: equipment ID, model, production date, last maintenance date, next maintenance date, usage records, user information, and remarks. Each field has 64 bits of reserved storage space to meet data storage needs.

[0052] Bluetooth beacon deployment strategy: Based on the actual environment, Bluetooth beacons with adaptive power control are used to reduce interference and precisely control power consumption. A grid-like layout of beacons ensures that equipment in any location is covered by at least three beacons, thereby achieving three-dimensional spatial positioning.

[0053] RFID reader and Bluetooth receiver integration: The design integrates an RFID high-frequency reader and a Bluetooth 5.0 receiver module, supporting fast switching and simultaneous reading of two signals, improving data acquisition efficiency and system response speed.

[0054] Central Processing Unit (CPU) and Cloud Interconnection: The CPU should not only process data locally, but also have the ability to synchronize efficiently with cloud servers, utilize edge computing technology for preliminary data analysis, reduce the burden on the cloud, and enable rapid decision feedback.

[0055] User-friendly interface: Develop cross-platform user applications that support operating systems such as iOS, Android, and Windows. The interface is intuitive and easy to use, providing features such as map view display of equipment distribution, search and filtering functions, report generation functions, and permission management module.

[0056] Example 2

[0057] The intelligent management system for fire equipment based on RFID and Bluetooth technologies enables intelligent management of fire equipment using these technologies. The flowchart is shown below. Figure 2 As shown, it includes the following steps:

[0058] The location of RFID tags is determined based on the collected RFID tag signals;

[0059] The location of the Bluetooth beacon is obtained based on the collected Bluetooth signals;

[0060] By employing a data fusion algorithm, the positioning of RFID tags and Bluetooth beacons is combined to calculate the accurate location of fire-fighting equipment.

[0061] Furthermore, data fusion algorithms include data fusion and location calculation.

[0062] The data fusion algorithm includes: obtaining RFID tag-based distance estimates and Bluetooth beacon-based distance estimates based on RFID tag positioning and Bluetooth beacon positioning; and then fusing these two estimates using a weighted average method. The weights can be determined based on factors such as signal strength stability, device reliability, and the accuracy of previous data. For example, if the RFID tag signal strength of a device is consistently stable and accurate, while the Bluetooth signal is subject to significant interference in the current environment, then a larger weight can be assigned to the RFID tag-based distance estimate.

[0063] Furthermore, to ensure greater data accuracy, when effectively fusing RFID and Bluetooth data, methods such as timestamp synchronization and signal strength correction are used to precisely calculate equipment location, while filtering out misreads and redundant information. The data fusion flowchart is as follows: Figure 3 As shown, the specific steps are as follows:

[0064] The processing is divided into several stages: data acquisition, clock tag synchronization, signal strength correction, data fusion and location calculation, and algorithm output and update.

[0065] 1) Data Acquisition Phase

[0066] RFID data acquisition begins with an RFID reader scanning the RFID tags on the equipment at a certain frequency, reading the unique identifier (UID) and Received Signal Strength Indicator (RSSI) information from the tags. Simultaneously, the timestamp of each read is recorded, accurate to the millisecond level. The read RFID data is then temporarily stored in a buffer, with a data format of [UID, RSSI_RFID, timestamp_RFID], where RSSI_RFID represents the RFID signal strength and Timestamp_RFID represents the timestamp of the RFID data.

[0067] Bluetooth data acquisition also begins by enabling the Bluetooth scanning function to search for Bluetooth devices within range and obtain information such as their Bluetooth signal strength (RSSI), device identifiers (e.g., MAC address), and timestamps. Similarly, the timestamps are accurate to the millisecond level. The Bluetooth data is then stored in another buffer, with a data format of [MAC, RSSI_Bluetooth, Timestamp_Bluetooth], where RSSI_Bluetooth represents the Bluetooth signal strength and Timestamp_Bluetooth represents the timestamp of the Bluetooth data.

[0068] 2) Timestamp synchronization phase

[0069] Establish a time reference by selecting the time from either the RFID or Bluetooth data acquisition system. For example, if the RFID system has higher time accuracy and stability, its time should be used as the reference time.

[0070] Timestamp Alignment 1. For each piece of data in the Bluetooth data buffer, calculate the difference between its timestamp and the reference time.

[0071] 2. Based on the difference, insert the Bluetooth data into the appropriate position in the RFID data buffer in chronological order. Specifically, if the difference between the timestamps of two RFID data points is within the range, insert the Bluetooth data between those two RFID data points.

[0072] 3) Signal strength correction stage

[0073] Establishing a signal strength model: 1. Collect a large amount of RFID and Bluetooth signal strength sample data under different distances and environmental conditions. Analyze this sample data to establish a mathematical model between signal strength and distance. For example, a logarithmic distance path loss model can be used: ;

[0074] in, This is the reference signal strength at a distance of 1 meter, where n is the path loss exponent and d is the distance between the device and the reader. It follows a normal distribution Environmental noise.

[0075] 2. Using parameter estimation methods such as least squares, the parameters in the model are estimated and optimized based on the collected sample data to obtain a signal strength model suitable for the current environment.

[0076] Signal strength correction 1. For each record in the synchronized RFID and Bluetooth data, convert the RSSI value into an estimated distance according to its corresponding signal strength model.

[0077] 2. Due to the different operating frequencies and characteristics of RFID and Bluetooth, the converted distance needs to be scaled or adjusted by a certain ratio to ensure better consistency between the two data in terms of distance scale.

[0078] 4) Data fusion and location computing stage

[0079] The data fusion strategy employs a weighted average method to combine the distance estimates from RFID and Bluetooth. The weights can be determined based on factors such as signal strength stability, device reliability, and the accuracy of previous data. For example, if a device's RFID signal strength has consistently been stable and accurate, while the Bluetooth signal is subject to significant interference in the current environment, then a larger weight can be assigned to the RFID distance estimate.

[0080] Let the distance estimate of the RFID be... The weight is The Bluetooth distance estimate is... The weight is Then the distance estimate after fusion is: .

[0081] Location Calculation 1. Assuming the locations of the RFID reader and Bluetooth scanner are known (coordinates respectively), calculate the equipment's location coordinates using triangulation or other suitable positioning algorithms, based on the fused distance estimate and the relative angle information between the device and the reader or scanner (if available).

[0082] 2. For example, in triangulation, if the positions of two devices and their distances to the target device are known, the position of the target device can be calculated by solving triangles. The specific calculation formula can be derived based on geometric relationships.

[0083] 5) Misreading and Redundant Information Filtering Stage

[0084] A signal strength threshold is set for the misread data filtering. RFID and Bluetooth data records with RSSI values ​​below this threshold are considered potentially misread and are removed from the buffer. This is because excessively weak signal strength may indicate inaccurate device readings or unreliable data due to excessive distance.

[0085] Employ outlier detection algorithms, such as statistical-based 3D algorithms. The principle or box plot method is used to detect outliers in distance estimates. If a distance estimate differs from the mean of other data by more than a certain multiple of the standard deviation (e.g., 3 times the standard deviation), it is considered an outlier, possibly due to interference or misreading caused by incorrect measurement, and is deleted.

[0086] Redundant information filtering involves retaining only the most recent record and deleting other redundant data for data read multiple times from the same device within a certain time window (e.g., 1 second) with minimal changes in distance estimates. This reduces the amount of data and improves the efficiency of subsequent processing.

[0087] Data is deduplicated based on the device's UID or MAC address. If multiple consecutive data entries have the same identifier and similar attributes, only one representative entry is retained, removing redundant information.

[0088] 6) Output and Update Phase

[0089] 1. Location information output

[0090] The calculated equipment location coordinates, along with related timestamps and fused signal strength, are output in a suitable data format. For example, they can be output as an array containing [x, y, Timestamp, RSSI_fusion], where x and y are the equipment location coordinates, Timestamp is the timestamp of the final data, and RSSI_fusion is the fused signal strength.

[0091] 2. Data Update and Looping

[0092] The system continuously cycles through data acquisition, processing, and fusion, updating the equipment's location information in real time. Different update frequencies can be set according to the specific application requirements, such as updating location information once per second, to ensure the real-time nature and accuracy of location tracking.

[0093] By following the steps above, RFID and Bluetooth data can be effectively fused to accurately calculate equipment location and filter out misreads and redundant information, providing accurate and reliable location data support. In practical applications, parameters should be adjusted and optimized according to the specific environment and equipment characteristics to achieve the best fusion effect.

[0094] Example 3

[0095] As a preferred embodiment, the method in Example 2 further includes: obtaining the movement trajectory of the fire-fighting equipment based on its location, and predicting the movement trend of the fire-fighting equipment using a dynamic path analysis algorithm. A flowchart of the dynamic path analysis algorithm for predicting the movement trend of the fire-fighting equipment is shown below. Figure 4 The specific steps are as shown:

[0096] Data collection and preprocessing stage

[0097] 1. Bluetooth and RFID data acquisition

[0098] Deploy a network of Bluetooth and RFID sensors to cover the device management area. Ensure the sensor layout is effective in capturing device movement signals.

[0099] Bluetooth devices periodically broadcast signals, including device identifiers (such as MAC addresses) and signal strength information. Simultaneously, RFID readers read the RFID tags on the devices in real time, obtaining data such as tag IDs, RSSI (Received Signal Strength Indicator), and timestamps.

[0100] The collected data is transmitted to the data processing center via wired or wireless means for storage and preprocessing.

[0101] 2. Data Cleaning and Integration

[0102] The raw data is cleaned to remove noise and outliers. For example, data points with weak or unstable signal strength, as well as data records with abnormal timestamps, are removed.

[0103] Based on the device's unique identifier (Bluetooth MAC address or RFID tag ID), Bluetooth and RFID data are integrated to create a comprehensive data record for the device. Each record includes the device ID, timestamp, Bluetooth RSSI, RFID RSSI, and other relevant information (such as sensor location).

[0104] 3. Location estimation

[0105] By utilizing a signal strength versus distance model and combining it with sensor location information, the device's location can be estimated using both Bluetooth and RFID data. A common model is the logarithmic distance path loss model, which calculates the distance between the device and the sensor using known signal strength reference values ​​and the path loss exponent.

[0106] By fusing the location estimates from Bluetooth and RFID, methods such as weighted averaging can be used to assign different weights based on the accuracy and reliability of the two technologies, resulting in a more accurate device location estimate.

[0107] Movement trajectory analysis stage

[0108] 1. Trajectory Construction

[0109] Based on the time sequence, the estimated positions of the equipment at different points in time are connected to form the equipment's movement trajectory. The trajectory diagram is then plotted with time on the horizontal axis and position coordinates on the vertical axis.

[0110] The trajectory is smoothed to remove sharp inflection points and abnormal fluctuations caused by measurement errors or signal fluctuations. Techniques such as moving averages or low-pass filtering can be used to achieve trajectory smoothing.

[0111] 2. Motion Feature Extraction

[0112] Calculate the velocity and acceleration of the equipment's trajectory. Velocity is calculated by observing changes in position at adjacent time points, and acceleration is then calculated based on these velocity changes. Velocity and acceleration reflect the equipment's movement status and trend.

[0113] Analyze the changes in the trajectory's direction to determine the equipment's turning point and turning angle. Calculate the trajectory's direction vector at different time intervals; when the direction vector changes significantly, mark it as a turning point and calculate the turning angle.

[0114] Statistical analysis of device dwell time and access frequency in different areas. The managed area is divided into multiple sub-areas; device trajectories are used to determine the dwell time and access frequency of each sub-area, thus understanding the device's activity patterns.

[0115] Moving trend prediction stage

[0116] 1. Model training based on historical data

[0117] Collect the device's movement trajectory data over a period of time to construct a training dataset. The dataset contains the device's historical trajectory features (speed, acceleration, direction, dwell time, etc.) as well as corresponding subsequent movement direction and distance information.

[0118] Choose an appropriate machine learning algorithm, such as linear regression, decision tree, or neural network, to train the model. Using historical trajectory features as input and the device's future direction and distance of movement as output, train the model to learn the relationship between trajectory features and movement trends.

[0119] The trained model is evaluated and optimized using techniques such as cross-validation to assess its accuracy and generalization ability. Based on the evaluation results, the model's parameters are adjusted or a more suitable algorithm is selected to improve its predictive performance.

[0120] 2. Real-time trend prediction

[0121] When new device location and trajectory data are acquired, the current trajectory features (speed, acceleration, direction, etc.) are extracted.

[0122] The extracted features are input into the trained prediction model to obtain the device's movement trend prediction results, including the direction of movement and the possible range of movement distance in the future.

[0123] Based on the prediction results, the predicted movement path of the equipment is plotted on the map to provide a reference for subsequent path planning and scheduling.

[0124] Path planning and scheduling optimization stage

[0125] 1. Goal setting and constraint definition

[0126] Define clear equipment management objectives, such as minimizing equipment travel time, maximizing asset turnover efficiency, and reducing energy consumption. Set corresponding optimization metrics based on different objectives.

[0127] Define the constraints for path planning, including physical limitations of equipment (such as maximum speed, turning radius, etc.), restrictions of the management area (such as obstacles, no-entry zones, etc.), and other business rules (such as equipment priority, working time limits, etc.).

[0128] 2. Path planning algorithm selection and implementation

[0129] Choose an appropriate path planning algorithm based on the objective and constraints. Common algorithms include A* algorithm, Dijkstra's algorithm, and genetic algorithm.

[0130] Starting from the device's current location and ending at the target location (determined based on task requirements or the device's predetermined destination), a path planning algorithm searches for the optimal path, taking into account map information and constraints. During the search process, the algorithm comprehensively considers factors such as path length, estimated travel time, and the possibility of conflicts with other devices to select the optimal path.

[0131] 3. Dynamic scheduling optimization

[0132] Based on the predicted movement trends of the equipment and real-time location information, the route planning is dynamically adjusted. If it is predicted that the equipment may encounter congestion or other interference factors on the original planned route, the route is replanned in a timely manner to avoid potential problem areas.

[0133] Considering the collaborative scheduling of multiple devices, optimize the scheduling order and time interval between devices based on their task requirements, priorities, and movement trajectories to avoid conflicts and waiting between devices and improve overall asset turnover efficiency.

[0134] A real-time feedback mechanism is introduced. When the equipment actually moves along the planned path, it continuously collects actual operating data (such as actual transit time, actual obstacles encountered, etc.) and optimizes and adjusts the path planning and scheduling strategy in real time based on this feedback information.

[0135] Results Output and Visualization Stage

[0136] 1. Path planning results output

[0137] The final determined equipment movement path is output in a visual way, such as displaying the path lines on an electronic map and marking the start, end, and waypoints of the path, as well as relevant navigation information (such as turn prompts, estimated arrival time, etc.).

[0138] At the same time, the path planning results are provided to the equipment control system or related management software in a data format (such as JSON or XML) so that the equipment can automatically navigate according to the planned path or be manually guided by the operator according to the plan.

[0139] 2. Scheduling suggestion generation and display

[0140] Based on the route planning and equipment task assignments, detailed scheduling suggestions are generated, including equipment departure time, estimated task completion time, and operation prompts at different nodes.

[0141] Dispatch recommendations can be presented to equipment managers in the form of reports or notifications, facilitating their monitoring and scheduling decisions regarding equipment operation. Dispatch recommendation notifications can be sent via email, SMS, or pop-up windows within the management system interface.

[0142] 3. Visual monitoring and analysis interface

[0143] Develop a visual monitoring and analysis interface to display the real-time location, movement trajectory, predicted path, and related scheduling information of the equipment. Managers can use this interface to intuitively understand the equipment's operating status and trends, promptly identify anomalies, and take appropriate measures.

[0144] The interface provides data analysis tools, such as trajectory playback and statistical chart generation, to facilitate managers in analyzing and evaluating the historical operating data of the equipment, so as to further optimize equipment management strategies and path planning algorithms.

[0145] By following the steps above, dynamic path planning for equipment management using Bluetooth and RFID technologies can be achieved. This effectively predicts equipment movement trends, optimizes scheduling suggestions, improves asset turnover efficiency, and provides intelligent decision support for equipment management. In practical applications, appropriate adjustments and optimizations need to be made according to specific equipment management scenarios and requirements to ensure the system's stability and usability.

[0146] Example 4

[0147] The machine learning-based abnormal behavior recognition model determines whether the fire-fighting equipment is in an abnormal state based on its location, and issues an early warning based on the abnormal state. The abnormal state early warning flowchart is as follows. Figure 5 As shown, the specific steps include:

[0148] Data collection and annotation phase

[0149] 1. Data Collection

[0150] Utilizing Bluetooth and RFID devices deployed in the device management area, relevant device data is continuously collected. This data includes device identification information (such as Bluetooth MAC address, RFID tag ID), timestamps, location information (estimated through signal strength), and other possible relevant parameters (such as device operating status, connection status, etc., if available).

[0151] Ensure a sufficiently high data acquisition frequency to accurately capture the dynamic behavior of the device. Simultaneously, guarantee the stability and reliability of the data acquisition system to prevent data loss or errors.

[0152] 2. Data labeling

[0153] Professional staff will annotate the collected historical data to identify normal and abnormal behavior samples. Abnormal behavior includes, but is not limited to, abnormal movement of equipment (such as rapid deviation from the predetermined path, sudden acceleration or deceleration, etc.), excessive dwell time (staying in a specific area for more than a preset threshold), and other behaviors that do not conform to the normal operating mode.

[0154] Assign a corresponding label to each labeled data sample, such as "normal" or "abnormal_abnormal movement", "abnormal_timeout", etc., for subsequent use in model training and validation.

[0155] Feature engineering stage

[0156] 1. Feature Extraction

[0157] Extract variables that reflect the behavioral characteristics of the equipment from the raw data. For position data, features such as the equipment's speed, acceleration, and rate of change of direction can be calculated. Speed ​​and acceleration can be calculated by dividing the position difference between adjacent time points by the time interval, while the rate of change of direction can be obtained by analyzing the changes in the direction vector at consecutive time points.

[0158] The study analyzes the distribution of device dwell time and access frequency across different areas. For example, it calculates the average dwell time, longest dwell time, and standard deviation of dwell time for each device in each area, as well as the number of times the device visits each area within a certain time period.

[0159] Considering time characteristics, such as the different behavior patterns of devices at different times (daytime, nighttime, weekdays, weekends, etc.), we can extract the corresponding time features, such as the number of hours corresponding to the timestamp, the day of the week, etc., as input variables for the model.

[0160] 2. Feature Selection and Transformation

[0161] Feature selection algorithms, such as correlation analysis and principal component analysis (PCA), are used to screen out the most representative and discriminative features for identifying abnormal behavior. Redundant or weakly correlated features are removed to reduce model complexity and improve training efficiency and prediction accuracy.

[0162] Performing necessary transformations on certain features, such as standardizing or normalizing numerical features, can make different features have similar scales and distribution ranges, which is beneficial for model training and optimization. For some categorical features (such as the day of the week in time features), one-hot encoding can be performed to convert them into numerical form for model processing.

[0163] Model selection and training phase

[0164] 1. Model Selection

[0165] Based on the characteristics of the task of identifying abnormal behavior in equipment management, a suitable machine learning model should be selected. Common models include decision trees, random forests, support vector machines (SVMs), and neural networks. For complex nonlinear problems, neural networks may have better performance; while for cases with small amounts of data and relatively simple features, decision trees or random forests may be more suitable.

[0166] Consider the interpretability and real-time requirements of the model. If interpretation and analysis of the early warning results are required, models with better interpretability, such as decision trees or linear models, can be selected. Simultaneously, ensure that the model can quickly make predictions in real-time monitoring scenarios to meet the timeliness requirements of the early warning system.

[0167] 2. Model Training

[0168] The labeled dataset is divided into training, validation, and test sets. Typically, this division is done in a certain proportion, for example, 70% for training, 20% for validation, and 10% for test.

[0169] The selected machine learning model is trained using the training set data. During training, model parameters (such as the number of layers, nodes, learning rate, and decision tree depth) are adjusted to optimize model performance. The model is then validated using the validation set data to monitor performance changes during training and prevent overfitting. Training is stopped when the model's performance on the validation set reaches its optimal or stable state, and the model parameters at this point are saved.

[0170] Model evaluation and optimization phase

[0171] 1. Model Evaluation

[0172] The trained model is comprehensively evaluated using test set data. Evaluation metrics include accuracy, recall, F1 score, and precision. Accuracy measures the proportion of samples correctly predicted by the model out of the total samples; recall represents the proportion of samples that are actually anomalous that were correctly predicted by the model; the F1 score is the harmonic mean of accuracy and recall; and precision reflects the proportion of samples predicted as anomalous that were actually anomalous.

[0173] Analyze the model's performance across different types of abnormal behavior to identify any areas where it performs poorly in identifying certain anomalies. Simultaneously, observe the model's false positive and false negative rates. The false positive rate is the proportion of normal samples incorrectly predicted as anomalous, while the false negative rate is the proportion of anomalous samples not correctly identified. These metrics are crucial for evaluating the performance and usability of the early warning system.

[0174] 2. Model Optimization

[0175] Based on the model evaluation results, optimize the model. If overfitting is found, measures such as increasing the amount of data, performing data augmentation, and adding regularization terms can be taken. If the overall performance of the model is not ideal, you can try adjusting the model structure, choosing different algorithms, or further optimizing feature engineering.

[0176] The model's threshold is adjusted to balance the false positive rate and the false negative rate. A suitable threshold is found through experimentation on the validation set or an independent tuning dataset, ensuring the early warning system maintains a high recall rate while minimizing the false positive rate to meet the needs of practical applications.

[0177] Real-time monitoring and early warning stage

[0178] 1. Real-time data acquisition and processing

[0179] Establish real-time connections with Bluetooth and RFID data acquisition systems to obtain the latest data from the devices in real time. Preprocess the real-time data, including data cleaning (removing noise and outliers), feature extraction, and transformation, to make it meet the requirements of the model input.

[0180] 2. Abnormal Behavior Prediction

[0181] The processed real-time data is input into a trained abnormal behavior recognition model. Based on the learned patterns and features, the model predicts whether the device's current behavior is abnormal. If the model predicts an anomaly, it further determines the type of anomaly (such as abnormal movement, excessive timeout, etc.).

[0182] 3. Sending early warning notifications

[0183] When abnormal behavior is detected, a warning notification should be sent to the administrator promptly. Notification methods can include email, SMS, system pop-ups, etc., to ensure that the administrator receives and pays attention to the warning information in a timely manner. The warning notification should include relevant equipment information (such as equipment identification, location, anomaly type, etc.), the time of occurrence, and possible recommended actions (such as immediately checking the equipment status, verifying equipment operation, etc.) so that the administrator can quickly take appropriate action.

[0184] System maintenance and update phase

[0185] 1. Data Update and Model Retraining

[0186] Over time, device behavior patterns may change, and new anomalies may emerge. Therefore, it is essential to periodically collect new data and add it to the existing dataset. The updated dataset should then be used to retrain or fine-tune the model to ensure it adapts to changes in the device management environment and maintains good performance and accuracy.

[0187] 2. System monitoring and performance evaluation

[0188] Establish a system monitoring mechanism to monitor the operational status of the early warning system in real time, including the accuracy of data collection, the timeliness and accuracy of model predictions, and the success rate of early warning notifications. Regularly evaluate and analyze the system's performance, generating performance reports to promptly identify and resolve any potential problems.

[0189] 3. Rule Adjustment and Optimization

[0190] Based on feedback and experience from practical applications, the rules and parameters of the early warning system are adjusted and optimized. For example, the definition and thresholds of abnormal behavior are adjusted according to the importance and risk level of different devices; the content and delivery method of early warning notifications are optimized according to changes in device management processes. Simultaneously, cases and handling experiences related to abnormal behavior are continuously summarized and accumulated to provide a basis and reference for system improvement.

[0191] By following the steps above, a machine learning-based intelligent early warning system for equipment management can be established. This system effectively monitors abnormal equipment conditions and promptly sends alerts to administrators, improving the security and efficiency of equipment management. During implementation, continuous optimization and improvement of each step are necessary to adapt to different equipment management scenarios and needs.

[0192] Example 5

[0193] The frequency of use of the fire-fighting equipment is determined based on its location. The operating mode is then adjusted according to the frequency and duration of use. This adjustment includes a sleep mode and a wake-up mode. A flowchart for adjusting the operating mode is shown below. Figure 6 As shown, the specific implementation steps are as follows:

[0194] Data monitoring and analysis phase

[0195] 1. Data Collection

[0196] Monitoring points are set up in the equipment deployment area to collect relevant data from Bluetooth beacons and RFID tags in real time using dedicated monitoring equipment or by integrating with existing systems. This data includes signal strength, number of reads, timestamps, and environmental parameters (such as temperature, humidity, and light intensity, if available).

[0197] Ensure that the frequency of data collection is appropriate, accurately reflecting equipment usage without excessively consuming resources. At the same time, guarantee the accuracy and integrity of the data to avoid data loss or errors.

[0198] 2. Using frequency analysis

[0199] Based on the collected timestamps and read count data, statistical analysis is performed on the usage frequency of Bluetooth beacons and RFID tags. Time can be divided into different time periods (such as hours, days, weeks, etc.), and the average number of reads or usage frequency distribution of the device within each time period can be calculated.

[0200] Identify peak and off-peak usage times for devices. For example, data analysis might reveal that a particular Bluetooth beacon is used more frequently between 9 AM and 11 AM and 2 PM and 5 PM on weekdays, and less frequently in the evenings and on weekends.

[0201] 3. Correlation analysis of environmental factors

[0202] This study investigates the correlation between environmental parameters (temperature, humidity, light intensity, etc.) and equipment usage frequency. Data analysis methods, such as correlation analysis and regression analysis, can be used to identify the patterns of influence of environmental factors on equipment use.

[0203] For example, it was found that the number of reads of a certain RFID tag increased significantly when the ambient temperature and humidity were high. This may be because the operating frequency of related equipment increased in such an environment, or environmental factors had a certain impact on the signal transmission of the tag, resulting in changes in the number of reads.

[0204] Hibernation and Wake-up Strategy Formulation Phase

[0205] 1. Set threshold

[0206] Based on the results of usage frequency analysis and environmental factor correlation analysis, sleep and wake-up thresholds are set for Bluetooth beacons and RFID tags, respectively. These thresholds may include usage frequency thresholds, environmental parameter thresholds, and comprehensive thresholds.

[0207] For example, for a Bluetooth beacon, it can be set to enter sleep mode when the number of reads is less than 10 per hour and the ambient temperature is between 15°C and 25°C (normal room temperature range). When the number of reads increases to more than 20 within a short period of time (e.g., within 10 minutes), or when the ambient temperature changes significantly (more than 5°C beyond the normal range), the wake-up mechanism is triggered.

[0208] 2. Establish priority rules

[0209] When multiple devices simultaneously meet the conditions for sleep or wake-up, priority rules should be established to avoid conflicts and ensure reasonable resource allocation. Priorities can be determined based on the importance of the device, the criticality of its location within the region, or other business needs.

[0210] For example, equipment located in core production areas has a higher priority and may not easily enter deep sleep mode even if its usage frequency is relatively low, in order to ensure a rapid response when needed. For equipment in auxiliary areas, however, sleep and wake-up strategies can be adjusted more flexibly according to actual conditions, prioritizing energy efficiency.

[0211] 3. Dynamic adjustment mechanism design

[0212] Considering the dynamic changes in environment and usage, a mechanism is designed to dynamically adjust sleep and wake-up thresholds and strategies. Based on real-time data feedback and machine learning algorithms, the system can automatically learn and adapt to different scenario changes.

[0213] For example, by using reinforcement learning algorithms, the system rewards or punishes based on the actual effects of each sleep and wake-up decision (such as battery life extension, device responsiveness, etc.), and continuously adjusts strategy parameters to gradually optimize energy efficiency management.

[0214] Algorithm implementation and integration phase

[0215] 1. Implement an intelligent sleep and wake-up algorithm via programming.

[0216] Use a suitable programming language (such as C, Python, etc.) and development tools to write algorithm code according to the established sleep and wake-up strategy. The algorithm code should include a data acquisition module, a threshold judgment module, a sleep and wake-up control module, and a communication module for interfacing with the device hardware.

[0217] The data acquisition module connects to monitoring devices or data sources and reads data, ensuring real-time access to the latest device status and environmental data. The threshold judgment module analyzes and compares the collected data based on set thresholds and rules to determine whether sleep or wake-up conditions are met. The sleep and wake-up control module sends corresponding control commands to Bluetooth beacons and RFID tags to switch device modes.

[0218] For the communication module, it is necessary to ensure compatibility and stability with the device hardware, and to accurately transmit control commands to the device and receive feedback information from the device.

[0219] 2. Integration with hardware devices

[0220] Integrate the written algorithm into the control system for Bluetooth beacons and RFID tags. This may require upgrading or modifying the device's firmware to add support for intelligent sleep and wake-up algorithms.

[0221] During integration, thorough testing and debugging are essential to ensure the algorithm and hardware can work together effectively. Verify the device's functional integrity and performance stability under different operating modes, including signal transmission strength, reading accuracy, and battery power consumption.

[0222] At the same time, the compatibility and scalability of the equipment should be considered to ensure that upgrades and expansions can be easily carried out if new equipment models or functional requirements are needed in the future.

[0223] 3. System Testing and Optimization

[0224] The system, which integrates intelligent sleep and wake-up mechanisms, underwent comprehensive testing in a real-world environment. The testing included functional testing, performance testing, energy efficiency testing, and stability testing.

[0225] Functional testing primarily verifies whether the algorithm can accurately implement the device's sleep and wake-up functions according to the set strategy, and its correctness and reliability under various conditions (such as different usage frequencies, environmental changes, and simultaneous operation of multiple devices). Performance testing focuses on indicators such as system response time and processing power during operation to ensure that it does not significantly affect the normal use of the device. Energy efficiency testing evaluates the effect of the intelligent sleep and wake-up mechanism on extending battery life by measuring the device's battery power consumption and battery life under different operating modes. Stability testing involves running the system for an extended period of time to observe for any abnormalities or malfunctions, ensuring that the system can operate stably and reliably.

[0226] Based on the test results, the algorithm and system will be optimized and adjusted. This may require further adjustments to threshold parameters, optimization of algorithm logic, and improvements to the coordination between hardware and software to enhance system performance and energy efficiency management.

[0227] Operation and monitoring phase

[0228] 1. System Deployment and Startup

[0229] Deploy the tested and optimized energy efficiency management system into the actual equipment management environment. Ensure that all components of the system (including data acquisition devices, servers, Bluetooth beacons, and RFID tags) are correctly installed and configured, and can communicate and operate normally.

[0230] Start the system to begin real-time monitoring and management of the Bluetooth beacons and RFID tags. During startup, closely monitor the system's operation to ensure no errors or anomalies occur.

[0231] 2. Real-time monitoring and data logging

[0232] Establish a real-time monitoring mechanism to view information such as the working status, battery level, usage frequency, and environmental parameters of Bluetooth beacons and RFID tags in real time through a monitoring interface or management tools. Simultaneously, the system should record and store this data in real time for subsequent analysis and evaluation.

[0233] Monitoring personnel can monitor the equipment's operating status at any time through the monitoring interface, promptly identify abnormal issues, and take appropriate measures. For example, if a device's battery level drops too quickly or it frequently enters abnormal operating modes, monitoring personnel can conduct further investigation and handling.

[0234] 3. Regular evaluation and strategy adjustment

[0235] The operational effectiveness of the energy efficiency management system should be evaluated regularly (e.g., weekly, monthly). Based on recorded equipment usage data, battery life extension, and other relevant indicators, the effectiveness and rationality of the intelligent sleep and wake-up mechanisms should be analyzed.

[0236] If the actual operating results differ from the expected goals, or if there are significant changes in the environment and usage, the hibernation and wake-up strategies need to be adjusted and optimized. This can be achieved by re-analyzing the data, adjusting threshold parameters, priority rules, or dynamic adjustment mechanisms to adapt to the new circumstances and further improve energy efficiency management and equipment operational stability.

[0237] By following the steps above, an intelligent sleep and wake-up mechanism for Bluetooth beacons and RFID tags can be effectively developed and implemented. This mechanism dynamically adjusts the operating mode based on usage frequency and environmental changes, thereby extending battery life and improving the energy efficiency management of the equipment. In practical applications, it is essential to continuously monitor the system's operation and optimize and improve it based on actual needs and feedback to ensure that the system can consistently and stably provide efficient energy support for equipment management.

[0238] The system also has other optimization strategies:

[0239] • Enhanced security: Encrypted communication protocols are used to protect data transmission security, and RFID tag data is stored in encrypted form to prevent information from being illegally read or tampered with.

[0240] • Scalable design: The system architecture design follows the principle of modularity, which facilitates the addition of new hardware devices or software functions in the future, such as integrating artificial intelligence recognition modules or introducing Internet of Things platforms.

[0241] • Ease of maintenance and upgrades: Provides remote firmware update capabilities for easy system maintenance and performance optimization, while also incorporating a self-diagnostic function to reduce on-site troubleshooting time.

[0242] This system is suitable for fire equipment management that requires high efficiency and high precision. It can also be used in other scenarios, such as manufacturing, medical, and logistics warehousing industries, and has broad market application prospects and social and economic benefits.

[0243] 1. Improve recognition accuracy

[0244] Precise positioning: The integrated reading device combines an RFID high-frequency reader with a Bluetooth 5.0 receiver module, supporting rapid switching and simultaneous reading of two signals, thereby improving data acquisition efficiency and system response speed.

[0245] 2. Improve management efficiency

[0246] Automated management: The system automatically collects, processes, and analyzes data, reducing manual intervention and improving work efficiency.

[0247] Real-time monitoring: Tracks the location and status of equipment in real time, promptly detects abnormalities, and reduces losses caused by information lag.

[0248] User-friendly: Develop cross-platform user applications that support operating systems such as iOS, Android, and Windows, with an intuitive and easy-to-use interface.

[0249] 3. Reduce costs

[0250] Reduce labor costs: Automatic equipment information reading and automated data management reduce reliance on manual labor and lower labor costs.

[0251] 4. Enhance security protection

[0252] Abnormal alarm: The system can promptly detect and alarm abnormal situations, such as unauthorized equipment movement or equipment failure, effectively ensuring equipment safety.

[0253] Access control: By setting access permissions, unauthorized personnel can be effectively prevented from accessing the system, thus protecting data security.

[0254] Finally, it should be noted that the embodiments described in the above description are merely preferred practices of the invention and should not be construed as limiting the scope of the invention. Equivalent substitutions for the technical solutions described in the foregoing embodiments do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the invention, and all such substitutions should be covered within the scope of the claims and specification of the invention.

Claims

1. A fire equipment intelligent management system based on RFID and Bluetooth technologies, characterized in that, Includes RFID tags, Bluetooth beacons, RFID readers, Bluetooth receivers, and a central processing unit. Firefighting equipment is equipped with RFID tags, which are used to store and transmit RFID tag signals containing basic information and identification codes of the firefighting equipment. Bluetooth beacons positioned in the pre-designated area by the fire truck emit Bluetooth signals; The central processing unit is used to fuse the RFID signals received by the RFID reader and the Bluetooth signals received by the Bluetooth receiver to obtain the location of the fire-fighting equipment. The data fusion process includes a timestamp synchronization stage, a signal strength correction stage, and a data fusion and location calculation stage. The timestamp synchronization phase includes the following steps: The unique identifier, signal strength indication information, and timestamp of the RFID tag are read using an RFID reader. The Bluetooth beacon's Bluetooth signal strength, device identifier, and timestamp of reading the Bluetooth beacon are obtained through the Bluetooth receiver. The RFID data acquisition time is selected as the reference time. For each piece of data in the Bluetooth data buffer, the difference between the Bluetooth beacon timestamp and the reference time is calculated. Bluetooth data is inserted into the RFID data buffer in chronological order. If the difference between the Bluetooth beacon's timestamp and the reference time is within the range of the timestamp difference between two RFID data, the Bluetooth data is inserted between these two RFID data. The signal strength correction stage includes the following steps: for each record in the synchronized RFID and Bluetooth data, the RSSI value is converted into an estimated distance according to its corresponding signal strength model; In the data fusion and location calculation stage, the estimated distance after fusion is: ； The distance estimate of RFID is The weight is The Bluetooth distance estimate is... The weight is ; The equipment's position coordinates are calculated based on the merged distance estimate.

2. The intelligent management system for fire-fighting equipment based on RFID and Bluetooth technologies as described in claim 1, characterized in that, The central processing unit also identifies, analyzes, and processes the received data to identify fire-fighting equipment, obtains real-time information, and updates the real-time information to the database server.

3. The intelligent management system for fire-fighting equipment based on RFID and Bluetooth technologies as described in claim 2, characterized in that, It also includes a data uploading unit, which transmits data from the central processing unit to the database server wirelessly, including 4G / NB-IoT. IoT wireless transmission methods, CAN bus, RS232, RS485.

4. A method for intelligent management of fire-fighting equipment based on RFID and Bluetooth technologies, characterized in that, Using as claimed in claim 1 The intelligent management system for fire-fighting equipment based on RFID and Bluetooth technologies described in any one of the three claims shall perform the following steps: The location of RFID tags is determined based on the collected RFID tag signals; The location of the Bluetooth beacon is obtained based on the collected Bluetooth signals; A data fusion algorithm is used to fuse the positioning data of RFID tags and Bluetooth beacons to obtain a fused distance estimate, and the positioning of fire-fighting equipment is calculated based on the distance estimate.

5. The intelligent management method for fire-fighting equipment based on RFID and Bluetooth technologies as described in claim 4, characterized in that, A weighted average method is used to fuse the distance estimates of RFID and Bluetooth. The weights in the weighted average are determined based on the stability of signal strength, the reliability of the device, and the accuracy of historical data.

6. The intelligent management method for fire-fighting equipment based on RFID and Bluetooth technologies as described in claim 5, characterized in that, The calculation process for locating firefighting equipment includes: obtaining the known locations of RFID readers and Bluetooth scanning devices; and using triangulation to calculate the equipment's position coordinates based on the fused distance estimates and the relative angle information between the firefighting equipment and the RFID reader or Bluetooth receiver.

7. The intelligent management method for fire-fighting equipment based on RFID and Bluetooth technologies as described in claim 4, characterized in that, Prior to data fusion, signal strength correction is also included, which is implemented using a logarithmic distance path loss model. The logarithmic distance path loss model is as follows: ； Where P0 is the reference signal strength at a distance of 1 meter, n is the path loss exponent, and d is the distance between the device and the reader. It follows a normal distribution Environmental noise.

8. The intelligent management method for fire-fighting equipment based on RFID and Bluetooth technologies as described in claim 4, characterized in that, Also includes: Based on the location of the fire-fighting equipment, the movement trajectory of the fire-fighting equipment is obtained, and the movement trend of the fire-fighting equipment is predicted by a dynamic path analysis algorithm.

9. The intelligent management method for fire-fighting equipment based on RFID and Bluetooth technologies as described in claim 4, characterized in that, Also includes: The abnormal behavior recognition model based on machine learning determines whether the fire-fighting equipment is in an abnormal state based on its location, and issues an early warning based on the abnormal state, which includes abnormal movement and prolonged stay.

10. The intelligent management method for fire-fighting equipment based on RFID and Bluetooth technologies as described in claim 4, characterized in that, Also includes: The frequency of use of the fire-fighting equipment is determined based on its location, and the working mode is adjusted according to the frequency of use and the time period of use. The adjusted working mode includes a sleep mode and a wake-up mode.

Images