Multi-sensor fusion security monitoring method and system

Abstract

The invention belongs to the technical field of security monitoring, and particularly relates to a multi-sensor fusion security monitoring method and system, which realizes time alignment of data by starting a data synchronization process and dynamically adjusting the data flow rate of each sensor, and effectively solves the real-time challenge caused by the time delay difference of the sensor data. According to the method, the real-time performance of the system and the consistency of data processing are improved, the consistency and reliability of data are enhanced by constructing a unified data framework and executing an exception detection process, and the situations of data loss and inconsistency are reduced; besides, anomaly detection is carried out by utilizing a unified data framework, potential risks can be identified more accurately, and corresponding early warning signals are activated in time, so that the early warning response capability of the system is optimized; in conclusion, the overall performance and reliability of the multi-sensor fusion safety monitoring system are remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of security monitoring, and in particular relates to a multi-sensor fusion security monitoring method and system. Background Art

[0002] In existing multi-sensor fusion security monitoring systems, environmental data is usually collected through various types of sensors (such as cameras, temperature sensors, smoke detectors, etc.). The data from these sensors is transmitted to the central processing unit for analysis and processing to achieve comprehensive coverage and real-time monitoring of the monitoring area. However, due to differences in data acquisition frequency and transmission rate of different sensors, data latency issues arise. This latency difference not only affects the real-time performance of the system, but may also cause data inconsistency or loss, thereby reducing the reliability and accuracy of the overall monitoring system.

[0003] The general scheme of the existing technology: Data collection and transmission: Each sensor collects data independently and transmits the data to the central processing unit through wired or wireless networks.

[0004] Data synchronization processing: Use timestamps to mark the time information of each sensor data, and perform preliminary synchronization processing on the data through software algorithms.

[0005] Data integration and analysis: Integrate the synchronized data into a unified data framework for further analysis and processing.

[0006] Anomaly detection and warning: Based on the set thresholds and rules, the data is detected for anomalies, the corresponding warning signals are triggered, and relevant personnel are notified.

[0007] Although the above methods solve the data synchronization problem to a certain extent, it is difficult to completely eliminate the real-time challenges caused by latency differences due to the lack of a dynamic adjustment mechanism for the sensor data flow rate. This problem is particularly prominent in high-density sensor deployment environments. Summary of the invention

[0008] The purpose of the present invention is to provide a multi-sensor fusion security monitoring method and system, which realizes the time alignment of data by starting the data synchronization process and dynamically adjusting the data flow rate of each sensor. This method can not only effectively solve the real-time challenges caused by the delay difference of sensor data, but also improve the response speed of the system and the consistency of data processing, so as to solve the problems raised in the above background technology.

[0009] To achieve the above object, the present invention proposes a multi-sensor fusion security monitoring method, comprising the following steps: Acquire an initial environmental data set of the sensor, and start a data synchronization process based on the initial environmental data set; adjust the data flow rate of each sensor through the data synchronization process to achieve time alignment, and use the time-aligned data to build a unified data framework; execute an anomaly detection process within the unified data framework, and activate a corresponding warning signal based on the result of the anomaly detection process; use the warning signal to update the safety status display panel, and optimize the collection strategy of the initial environmental data set based on the feedback of the safety status display panel.

[0010] Preferably, the step of obtaining an initial environmental data set of a sensor includes: Define the boundaries of the monitoring area and set data collection points, deploy sensors based on the data collection points and start preliminary data collection; Normalize the preliminary data to make the data format of each sensor consistent. The formula is D_norm=(D-Min_D) / (Max_D-Min_D), where D_norm represents the normalized data, D is the original data, Min_D and Max_D are the minimum and maximum values ​​of the data respectively; Integrate the normalized data into a unified data structure to form the initial environment dataset.

[0011] Preferably, starting the data synchronization process based on the initial environment data set includes: Analyze the timestamp information in the initial environmental data set to determine the time base of each sensor data; Calculate the time difference of each sensor data arrival based on the time reference, the formula is Delta_T=T_sensor-T_ref, where Delta_T represents the time difference, T_sensor is the timestamp of the sensor data, and T_ref is the reference time reference; The data transmission frequency of each sensor is adjusted according to the time difference to align the time of all sensor data, and the time-aligned data streams are integrated into a unified time axis to form a synchronous data stream.

[0012] Preferably, adjusting the data flow rate of each sensor through the data synchronization process to achieve time alignment includes: Measure the initial rate of each sensor data stream and record it as the base rate value Base_Rate; Calculate the difference between the actual rate of each sensor and the base rate based on the base rate value. The formula is Diff_Rate=Rate_sensor-Base_Rate, where Diff_Rate represents the rate difference and Rate_sensor is the actual rate of the sensor. According to the rate difference, the data transmission interval of each sensor is adjusted to make the data rate of all sensors consistent. The adjustment formula is Interval_T=Ref_T / (1+Diff_Rate), where Interval_T represents the new data transmission interval and Ref_T is the reference time interval. Apply the adjusted data sending interval to each sensor and verify whether the data stream is time-aligned to ensure that all sensor data are synchronized on the same timeline.

[0013] Preferably, constructing a unified data framework using the time-aligned data includes: Define unified time base and data format standards to ensure that all sensor data follow the same rules; Based on the time base and data format standards, the time-aligned data is converted into unified format data Unified_D; Sort and integrate the data in a unified format, and arrange the data of each sensor in chronological order. The formula is Sorted_D=Sort(Unified_D,Time_Key), where Sorted_D represents the sorted data set and Time_Key is the timestamp key value; Create a global data structure and insert the sorted data set into the structure to form a complete unified data framework.

[0014] Preferably, the performing of the anomaly detection process within the unified data framework includes: Set the threshold range for anomaly detection and determine the upper limit Upper_Limit and lower limit Lower_Limit based on historical data and system requirements; Based on the time series data in the unified data framework, the data deviation at each time point is calculated using the formula Deviation = (Unified_D-Avg_D) / Std_D, where Deviation represents the deviation, Unified_D is the data point in the unified data framework, Avg_D is the average value, and Std_D is the standard deviation; Compare the deviation with the threshold range and mark the data points exceeding the threshold as potential anomaly points Anomaly_Points; Verify and classify potential anomalies, determine whether they are real abnormal events based on continuity rules, and update the anomaly record table.

[0015] Preferably, activating a corresponding warning signal according to the result of the abnormality detection process includes: Define different types of abnormal levels and their corresponding warning signals, and set the abnormal level threshold range Level_Thresholds based on the results of the abnormal detection process; Based on the deviation and type of potential anomaly points Anomaly_Points, determine the anomaly level of each anomaly point. The formula is: Alert_Level=Match(Deviation,Level_Thresholds), where Alert_Level indicates the abnormal level and Match is the matching operation; According to the result of the abnormal level, the corresponding warning signal mode is selected, and the abnormal level is mapped to the predefined warning signal set Signal_Set; Activate selected warning signals and send them to the monitoring center to ensure timely response and processing, and update the warning record table to track the processing progress.

[0016] Preferably, the updating of the safety status display panel using the warning signal comprises: Define the layout and elements of the safety status display panel, and determine the layout Panel_Layout of the safety status display panel for each warning signal on the panel; Based on the selected warning signal set Signal_Set, each warning signal is mapped to the corresponding position of the safety status display panel. The formula is: Display_Pos=Map(Signal_Set,Panel_Layout), where Display_Pos indicates the specific position of the warning signal on the panel, and Map is the mapping operation; According to the mapping results, the status of each warning signal is updated in real time on the safety status display panel, including color, icon and text description; Records and archives a snapshot of the updated security status dashboard and generates a log file to track all update activity.

[0017] Preferably, the step of optimizing the collection strategy of the initial environment data set based on the feedback from the safety status display panel includes: Analyze the feedback information on the safety status display panel and identify the areas High_Areas and time periods High_Times that frequently trigger warning signals; Based on the high-frequency warning areas High_Areas and time periods High_Times, the effectiveness of the current data collection strategy is evaluated, and the coverage of each sensor in these areas and time periods is calculated. The formula is: , where Coverage represents coverage, Collected_D is the amount of data collected, and Required_D is the amount of data required; According to the coverage results, adjust the deployment density and sampling frequency of sensors in high-frequency warning areas and time periods to ensure more comprehensive and timely data collection, and update sensor configuration parameters, recorded as Config_Params; Apply the new sensor configuration parameters Config_Params to the data acquisition system and verify its effect. Confirm the optimization effect by comparing the quality of the new and old data sets and the frequency of warning signals. Record the optimization results and update the acquisition strategy document.

[0018] On the other hand, the present invention provides a multi-sensor fusion security monitoring system, including a data acquisition module for acquiring an initial environmental data set of a sensor; A data synchronization module, used to start a data synchronization process based on the initial environment data set; A rate adjustment and time alignment module, used to adjust the data flow rate of each sensor through the data synchronization process to achieve time alignment; A data framework construction module, used to construct a unified data framework using the time-aligned data; An anomaly detection module, used to perform an anomaly detection process within the unified data framework; An early warning signal activation module is used to activate the corresponding early warning signal according to the result of the abnormality detection process; A status display panel update module, used to update the safety status display panel using the warning signal; A strategy optimization module is used to optimize the collection strategy of the initial environment data set according to the feedback of the security status display panel.

[0019] Technical effects and advantages of the present invention: Compared with the prior art, the multi-sensor fusion security monitoring method and system proposed in the present invention have the following advantages: The present invention realizes the time alignment of data by starting the data synchronization process and dynamically adjusting the data flow rate of each sensor, effectively solving the real-time challenge caused by the delay difference of sensor data. This method not only improves the real-time performance of the system and the consistency of data processing, but also enhances the consistency and reliability of data and reduces data loss and inconsistency by building a unified data framework and executing anomaly detection process. In addition, the use of a unified data framework for anomaly detection can more accurately identify potential risks and activate corresponding early warning signals in a timely manner, thereby optimizing the early warning response capability of the system. In summary, the present invention significantly improves the overall performance and reliability of the multi-sensor fusion security monitoring system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flow chart of a multi-sensor fusion security monitoring method of the present invention; Figure 2 The block diagram of a multi-sensor fusion security monitoring system of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0022] The present invention provides Figure 1 A multi-sensor fusion safety monitoring method shown includes the following steps: Step 1: Obtain the initial environmental data set of the sensor; specifically including: Defining the boundary range of the monitoring area and setting data collection points ensures that the sensor can cover the entire monitoring area, avoids monitoring blind spots, and improves the comprehensiveness and accuracy of data collection.

[0023] Deploying sensors based on data collection points and starting preliminary data collection ensures that each sensor can start working according to the predetermined plan, providing basic data support for subsequent data processing and analysis.

[0024] Normalize the preliminary data to make the data format of each sensor consistent. The formula is D_norm=(D-Min_D) / (Max_D-Min_D), where D_norm represents the normalized data, D is the original data, Min_D and Max_D are the minimum and maximum values ​​of the data respectively; The normalized data is integrated into a unified data structure to form the initial environmental data set. By integrating the normalized data, a unified data structure is formed, which is convenient for subsequent data processing, analysis and storage, and ensures the consistency and readability of the data.

[0025] Example 1 Assume that there is a monitoring area with three sensors A, B, and C, which are used to monitor temperature, humidity, and light intensity respectively. Each sensor collects the following data in a period of time: Sensor A (temperature): [25, 27, 30]; Sensor B (humidity): [45,50,60]; Sensor C (light intensity): [100, 200, 300]; Implementation steps: Define the monitoring area boundary and set data collection points: The monitoring area is set as a 10mx10m room, and five data collection points are set at the four corners and the center of the room.

[0026] Deploy sensors based on data collection points and start initial data collection: A temperature sensor A, a humidity sensor B, and a light intensity sensor C are installed at each data collection point, and data collection is started to obtain the above data.

[0027] Normalize the preliminary data to make the data format of each sensor consistent: Normalize the data of each sensor: Sensor A (Temperature): Min_D=25,Max_D=30; D_norm_A=[(25-25) / (30-25),(27-25) / (30-25),(30-25) / (30-25)]=[0,0.4,1]; Sensor B (humidity): Min_D=45,Max_D=60; D_norm_B=[(45-45) / (60-45),(50-45) / (60-45),(60-45) / (60-45)]=[0,0.333,1]; Sensor C (light intensity): Min_D=100,Max_D=300; D_norm_C=[(100-100) / (300-100),(200-100) / (300-100),(300-100) / (300-100)]=[0,0.5,1]; Integrate the normalized data into a unified data structure to form the initial environment data set: Integrate the normalized data into a unified data structure as follows: { "sensor_A":[0,0.4,1], "sensor_B":[0,0.333,1], "sensor_C":[0,0.5,1] }; Through the above steps, the initial environmental data set of the sensor was successfully obtained and normalized to ensure the consistency and comparability of the data, laying a solid foundation for subsequent data processing and analysis.

[0028] Step 2: Based on the initial environment data set, start the data synchronization process; specifically including: By analyzing the timestamp information in the initial environmental data set and determining the time base of each sensor data, a unified time base can be determined. This step ensures that the data of all sensors can be compared and processed in the same time frame, providing a basis for subsequent time alignment.

[0029] The time difference of the arrival of each sensor data is calculated based on the time base. The formula is Delta_T=T_sensor-T_ref, where Delta_T represents the time difference, T_sensor is the timestamp of the sensor data, and T_ref is the reference time base. By calculating the time difference of each sensor data, the deviation between each sensor data and the reference time base can be quantified, thereby providing a basis for adjusting the frequency of data sending.

[0030] The data transmission frequency of each sensor is adjusted according to the time difference to align the time of all sensor data; by adjusting the data transmission frequency of each sensor, the data of all sensors can be kept consistent on a unified time axis, solving the real-time problem caused by sensor delay differences and improving the system's response speed and data consistency.

[0031] Integrate time-aligned data streams onto a unified timeline to form synchronized data streams, which not only improves data consistency and readability, but also provides reliable data support for subsequent anomaly detection and early warning signal activation.

[0032] Example 2 Assume that there are three sensors A, B, and C, which are used to monitor temperature, humidity, and light intensity respectively, and each sensor collects the following data at different time points: Sensor A (Temperature): Data: [25, 27, 30]; Timestamp: [10:00:00,10:00:05,10:00:10]; Sensor B (humidity): Data: [45,50,60]; Timestamps: [10:00:02, 10:00:07, 10:00:12]; Sensor C (light intensity): data: [100, 200, 300]; Timestamps: [10:00:01, 10:00:06, 10:00:11]; Implementation steps: Analyze the timestamp information in the initial environmental data set to determine the time base of each sensor data: Set the reference time base T_ref to 10:00:00.

[0033] Calculate the time difference of each sensor data arrival based on the time base: Calculate the timestamp of each sensor to get the time difference Delta_T: Sensor A (Temperature): Delta_T_A1=10:00:00-10:00:00=0 seconds; Delta_T_A2 = 10:00:05-10:00:00 = 5 seconds; Delta_T_A3=10:00:10-10:00:00=10 seconds; Sensor B (humidity): Delta_T_B1=10:00:02-10:00:00=2 seconds; Delta_T_B2 = 10:00:07-10:00:00 = 7 seconds; Delta_T_B3=10:00:12-10:00:00=12 seconds; Sensor C (light intensity): Delta_T_C1 = 10:00:01-10:00:00 = 1 second; Delta_T_C2 = 10:00:06-10:00:00 = 6 seconds; Delta_T_C3 = 10:00:11-10:00:00 = 11 seconds; Adjust the data transmission frequency of each sensor according to the time difference to align the time of all sensor data: According to the time difference, the data transmission frequency of each sensor is adjusted so that its data can be aligned on a unified time axis. For example, the alignment can be achieved by increasing or decreasing the sampling interval.

[0034] Sensor A (Temperature): No adjustment is required because its timestamps are already aligned to the reference time base.

[0035] Sensor B (humidity): You can delay sending data for 2 seconds after each acquisition to align timestamps.

[0036] Sensor C (light intensity): You can delay sending data for 1 second after each acquisition to align timestamps.

[0037] Integrate the time-aligned data streams onto a unified timeline to form a synchronized data stream: The adjusted data is as follows: Sensor A (Temperature): Data: [25, 27, 30]; Timestamp: [10:00:00,10:00:05,10:00:10]; Sensor B (humidity): Data: [45,50,60]; Timestamp: [10:00:00,10:00:05,10:00:10]; Sensor C (light intensity): data: [100, 200, 300]; Timestamp: [10:00:00,10:00:05,10:00:10]; The resulting synchronous data flow is as follows: { "time":[10:00:00,10:00:05,10:00:10], "sensor_A":[25,27,30], "sensor_B":[45,50,60], "sensor_C":[100,200,300] }; Through the above steps, the data synchronization process is successfully started, and by adjusting the data sending frequency of each sensor, the time alignment of the data is achieved to form a synchronous data stream.

[0038] Step 3: adjusting the data flow rate of each sensor through the data synchronization process to achieve time alignment; specifically including: Measuring the initial rate of each sensor data stream and recording it as the base rate value Base_Rate can provide a reference standard for subsequent adjustment of the data transmission frequency of each sensor. This step ensures that all sensors have a unified rate benchmark, which is convenient for subsequent calculation and adjustment.

[0039] The difference between the actual rate of each sensor and the base rate is calculated based on the base rate value. The formula is Diff_Rate=Rate_sensor-Base_Rate, where Diff_Rate represents the rate difference and Rate_sensor is the actual rate of the sensor. By calculating this difference, the deviation between the rate of each sensor and the base rate can be quantified, thus providing a basis for adjusting the data sending interval.

[0040] The data transmission interval of each sensor is adjusted according to the rate difference to make the data rate of all sensors consistent. The adjustment formula is Interval_T=Ref_T / (1+Diff_Rate), where Interval_T represents the new data transmission interval and Ref_T is the reference time interval. This formula dynamically adjusts the transmission interval according to the rate difference to ensure that the data of all sensors are synchronized on the same time axis.

[0041] Apply the adjusted data transmission interval to each sensor and verify whether the data stream is time-aligned to ensure that all sensor data is synchronized on a unified time axis. This step improves the real-time performance and data consistency of the system and reduces problems caused by latency differences.

[0042] Example 3 Assume that there are three sensors A, B, and C, which are used to monitor temperature, humidity, and light intensity respectively, and the initial rate of each sensor is as follows: Sensor A (temperature): The initial rate is to collect data once every 5 seconds, that is, Rate_sensor_A=0.2 times / second; Sensor B (humidity): The initial rate is to collect data once every 7 seconds, that is, Rate_sensor_B = 0.143 times / second; Sensor C (light intensity): The initial rate is to collect data once every 10 seconds, that is, Rate_sensor_C=0.1 times / second; The reference rate Base_Rate is set to collect data once every 6 seconds, that is, Base_Rate = 0.167 times / second.

[0043] Implementation steps: Measure the initial rate of each sensor data stream and record it as the base rate value Base_Rate: Base_Rate=0.167 times / second; Calculate the difference between the actual rate and the reference rate for each sensor based on the reference rate value: Sensor A (Temperature): Diff_Rate_A=Rate_sensor_A-Base_Rate=0.2-0.167=0.033 times / second; Sensor B (humidity): Diff_Rate_B=Rate_sensor_B-Base_Rate=0.143-0.167=-0.024 times / second; Sensor C (light intensity): Diff_Rate_C=Rate_sensor_C-Base_Rate=0.1-0.167=-0.067 times / second; Adjust the data transmission interval of each sensor according to the rate difference to make the data rate of all sensors consistent: Set the reference time interval Ref_T to 6 seconds and adjust the data sending interval according to the formula: Sensor A (Temperature): Interval_T_A=Ref_T / (1+Diff_Rate_A)=6 / (1+0.033)=5.81 seconds; Sensor B (humidity): Interval_T_B=Ref_T / (1+Diff_Rate_B)=6 / (1-0.024)=6.15 seconds; Sensor C (light intensity): Interval_T_C=Ref_T / (1+Diff_Rate_C)=6 / (1-0.067)=6.43 seconds; Apply the adjusted data sending interval to each sensor and verify whether the data stream is time-aligned to ensure that all sensor data are synchronized on a unified time axis: The adjusted data sending interval is as follows: Sensor A (temperature): sends data approximately every 5.81 seconds; Sensor B (humidity): sends data every approximately 6.15 seconds; Sensor C (light intensity): sends data approximately every 6.43 seconds; Verify that the data streams are time aligned: On a unified timeline, data from all sensors are sent at intervals close to 6 seconds, ensuring time alignment of the data.

[0044] The resulting synchronous data flow is as follows: { "time":[10:00:00,10:00:06,10:00:12,...], "sensor_A":[25,27,30,...], "sensor_B":[45,50,60,...], "sensor_C":[100,200,300,...] }; Through the above steps, the data flow rate of each sensor is successfully adjusted through the data synchronization process to achieve time alignment.

[0045] Step 4: Using the time-aligned data to construct a unified data framework; specifically including: Define a unified time base and data format standard to ensure that all sensor data follows the same rules, that all sensor data is processed on the same timeline, and that the data format is consistent. This step provides a basis for subsequent data integration and analysis, and reduces errors caused by inconsistent data formats.

[0046] Based on the time reference and data format standards, the time-aligned data is converted into unified format data Unified_D, so that data from different sensors can be compared and processed in the same framework. This step improves the consistency and readability of the data and facilitates subsequent data processing and analysis.

[0047] Sort and integrate the data in a unified format, and arrange the data of each sensor in chronological order. The formula is Sorted_D=Sort(Unified_D,Time_Key), where Sorted_D represents the sorted data set and Time_Key is the timestamp key value. By using the timestamp as the sorting basis, the data of all sensors can be arranged in chronological order to ensure the time consistency of the data.

[0048] Create a global data structure and insert the sorted data set into the structure to form a complete unified data framework. This step not only improves the organization and readability of the data, but also provides reliable data support for subsequent data analysis, anomaly detection and other operations.

[0049] Example 4 Assume that there are three sensors A, B, and C, which are used to monitor temperature, humidity, and light intensity respectively. After time alignment, the following data is obtained: Sensor A (Temperature): Data: [25, 27, 30]; Timestamps: [10:00:00,10:00:06,10:00:12]; Sensor B (humidity): Data: [45,50,60]; Timestamps: [10:00:00,10:00:06,10:00:12]; Sensor C (light intensity): data: [100, 200, 300]; Timestamps: [10:00:00,10:00:06,10:00:12]; Implementation steps: Define a unified time base and data format standard to ensure that all sensor data follows the same rules: Set the unified time base as the system time, and specify the timestamp format of all sensor data as HH:MM:SS. At the same time, specify the data format as floating point type.

[0050] Based on the time base and data format standards, the time-aligned data is converted into unified format data Unified_D: Convert the time-aligned data into a unified format: Sensor A (Temperature): Unified_D_A=[(10:00:00,25.0),(10:00:06,27.0),(10:00:12,30.0)]; Sensor B (humidity): Unified_D_B=[(10:00:00,45.0),(10:00:06,50.0),(10:00:12,60.0)]; Sensor C (light intensity): Unified_D_C=[(10:00:00,100.0),(10:00:06,200.0),(10:00:12,300.0)]; Sort and integrate the data in a unified format, and arrange the data of each sensor in chronological order: Use timestamp as sorting basis to arrange the data of all sensors in chronological order: Merge data: Combined_Unified_D=[(10:00:00,"sensor_A",25.0),(10:00:00,"sensor_B",45.0),(10:00:00,"sensor_C",100.0),(10:00:06,"sensor_A",27.0),(10: 00:06,"sensor_B",50.0),(10:00:06,"sensor_C",200.0),(10:00:12,"sensor_A",30.0),(10:00:12,"sensor_B",60.0),(10:00:12,"sensor_C",300.0)]; Sorting formula: Sorted_D=Sort(Combined_Unified_D,Time_Key); The sorted dataset: Sorted_D=[ (10:00:00,"sensor_A",25.0), (10:00:00,"sensor_B",45.0), (10:00:00,"sensor_C",100.0), (10:00:06,"sensor_A",27.0), (10:00:06,"sensor_B",50.0), (10:00:06,"sensor_C",200.0), (10:00:12,"sensor_A",30.0), (10:00:12,"sensor_B",60.0), (10:00:12,"sensor_C",300.0) ]; Create a global data structure and insert the sorted dataset into the structure to form a complete unified data framework: Create a global data structure, such as a dictionary or list, and insert the sorted dataset into it: { "time":["10:00:00","10:00:00","10:00:00","10:00:06","10:00:06","10:00:06","10:00:12","10:00:12","10:00:12"], "sensor":["sensor_A","sensor_B","sensor_C","sensor_A","sensor_B","sensor_C","sensor_A","sensor_B","sensor_C"], "data":[25.0,45.0,100.0,27.0,50.0,200.0,30.0,60.0,300.0] }; Through the above steps, a unified data framework was successfully constructed using time-aligned data. This not only improved the consistency and readability of the data, but also provided a reliable foundation for subsequent data processing and analysis.

[0051] Step 5: Execute the anomaly detection process within the unified data framework; specifically including: Set the threshold range for anomaly detection, determine the upper limit Upper_Limit and the lower limit Lower_Limit based on historical data and system requirements; this can provide judgment criteria for subsequent deviation calculations. This step ensures that the anomaly detection process can accurately identify data points that are beyond the normal range, thereby improving the reliability of the system.

[0052] Based on the time series data in the unified data framework, the data deviation at each time point is calculated. The formula is Deviation=(Unified_D-Avg_D) / Std_D, where Deviation represents the deviation, Unified_D is the data point in the unified data framework, Avg_D is the average, and Std_D is the standard deviation. By calculating the deviation, the degree of difference between each data point and the average can be quantified, so as to determine whether it is abnormal data.

[0053] Compare the deviation with the threshold range and mark the data points that exceed the threshold as potential anomaly points. This step helps to quickly locate data points that may have problems and provide a basis for further analysis.

[0054] Verify and classify potential anomalies, determine whether they are real anomalies based on continuity rules, and update the anomaly record table. By verifying and classifying potential anomalies, you can further confirm whether these data points are indeed anomalies. Judging based on continuity rules (such as data trends at adjacent time points) can reduce the false alarm rate and update the anomaly record table for subsequent tracking and processing.

[0055] Example 5 Assume that there is a monitoring area with three sensors A, B, and C, which are used to monitor temperature, humidity, and light intensity, respectively. After time alignment and building a unified data framework, the following data is obtained: Timestamps: [10:00:00,10:00:06,10:00:12]; data: Sensor A (temperature): [25.0, 27.0, 30.0]; Sensor B (humidity): [45.0, 50.0, 60.0]; Sensor C (light intensity): [100.0, 200.0, 300.0]; Implementation steps: Set the threshold range for anomaly detection and determine the upper limit Upper_Limit and lower limit Lower_Limit based on historical data and system requirements: Set the threshold range for each sensor as follows: Sensor A (temperature): Lower_Limit=20.0, Upper_Limit=35.0; Sensor B (humidity): Lower_Limit=40.0, Upper_Limit=65.0; Sensor C (light intensity): Lower_Limit=80.0, Upper_Limit=350.0; Based on the time series data in the unified data framework, calculate the data deviation at each time point: Calculate the mean and standard deviation of each sensor at each time point, and then calculate the deviation: Sensor A (Temperature): Data: [25.0, 27.0, 30.0]; Average value Avg_D_A=(25.0+27.0+30.0) / 3=27.33; Standard deviation Std_D_A=2.08; Deviation: Deviation_A=[(25.0-27.33) / 2.08,(27.0-27.33) / 2.08,(30.0-27.33) / 2.08]=[-1.12,-0.16,1.29]; Sensor B (humidity): Data: [45.0, 50.0, 60.0]; Average value Avg_D_B=(45.0+50.0+60.0) / 3=51.67; Standard deviation Std_D_B=6.11; Deviation: Deviation_B=[(45.0-51.67) / 6.11,(50.0-51.67) / 6.11,(60.0-51.67) / 6.11]=[-1.09,-0.27,1.36]; Sensor C (light intensity): data: [100.0, 200.0, 300.0]; Average value Avg_D_C=(100.0+200.0+300.0) / 3=200.0; Standard deviation Std_D_C=100.0; Deviation: Deviation_C=[(100.0-200.0) / 100.0,(200.0-200.0) / 100.0,(300.0-200.0) / 100.0]=[-1.0,0.0,1.0]; Compare the deviation with the threshold range and mark the data points that exceed the threshold as potential anomaly points Anomaly_Points: Set the deviation threshold range to [-2, 2] (can be adjusted according to actual conditions) and mark the data points that exceed the threshold: Sensor A (Temperature): Deviation: [-1.12, -0.16, 1.29], all within the threshold range, no abnormal points.

[0056] Sensor B (humidity): Deviation: [-1.09, -0.27, 1.36], all within the threshold range, no abnormal points.

[0057] Sensor C (light intensity): Deviation: [-1.0, 0.0, 1.0], all within the threshold range, no abnormal points.

[0058] In this example, none of the data points are outside the threshold range, so there are no potential outliers.

[0059] Verify and classify potential anomalies, determine whether they are real abnormal events based on continuity rules, and update the abnormal record table: In this example, since there are no potential outliers, no further verification and classification is required. However, in actual applications, if there are potential outliers, verification and classification can be performed in the following ways: Continuity rule: Check the data trend at adjacent time points to determine whether there are sudden changes or continuous anomalies.

[0060] Verification and classification: Based on the verification results, potential anomalies are classified as real abnormal events or normal fluctuations, and the anomaly record table is updated.

[0061] Example abnormal record table (assuming there are abnormal points): { "time":["10:00:00","10:00:06"], "sensor":["sensor_A","sensor_B"], "data":[25.0,65.0], "deviation":[-1.12,2.18], "status":["normal","abnormal"] }; Through the above steps, the anomaly detection process was successfully implemented within a unified data framework. This not only improved the reliability and accuracy of the system, but also provided a solid foundation for subsequent data analysis and early warning signal activation.

[0062] Step 6: Activate the corresponding warning signal according to the result of the anomaly detection process; specifically including: Defining different types of abnormal levels and their corresponding warning signals, and setting the abnormal level threshold range Level_Thresholds based on the results of the abnormal detection process can provide clear standards for subsequent abnormal level judgments. This step ensures that the system can automatically identify and classify abnormal events based on the degree of deviation, thereby improving the system's response efficiency.

[0063] Based on the deviation and type of potential anomaly points Anomaly_Points, determine the anomaly level of each anomaly point. The formula is: Alert_Level=Match(Deviation,Level_Thresholds), where Alert_Level represents the abnormal level and Match is the matching operation; through this formula, the severity of each abnormal point can be quantified.

[0064] According to the results of the abnormal level, the corresponding warning signal mode is selected and the abnormal level is mapped to the predefined warning signal set Signal_Set, which can ensure that the system can issue appropriate warning signals in time when an abnormality is detected. This step improves the automation level and response speed of the system.

[0065] Activate the selected warning signals and send them to the monitoring center to ensure timely response and processing, and update the warning record table to track the processing progress for subsequent analysis and improvement.

[0066] Example 6 Assume that there are the following specific deviation data: Sensor A (temperature): [25.0, 27.0, 30.0]; Average value Avg_D_A=27.33; Standard deviation Std_D_A=2.08; Deviation: Deviation_A=[(25.0-27.33) / 2.08=-1.12,(27.0-27.33) / 2.08=-0.16,(30.0-27.33) / 2.08=1.29]; Sensor C (light intensity): [100.0, 200.0, 300.0]; Average value Avg_D_C=200.0; Standard deviation Std_D_C=100.0; Deviation: Deviation_C=[(100.0-200.0) / 100.0=-1.0,(200.0-200.0) / 100.0=0.0,(300.0-200.0) / 100.0=1.0]; According to the above deviation data: Sensor A (Temperature): The deviation of timestamp 10:00:12 is 1.29; Using the formula Alert_Level_A=Match(1.29,Level_Thresholds), we get Alert_Level_A="Moderate"; The corresponding warning signal is "OrangeLight"; Sensor C (light intensity): The deviation of timestamp 10:00:12 is 1.0; Use the formula Alert_Level_C=Match(1.0,Level_Thresholds) to get Alert_Level_C="Minor"; The corresponding warning signal is "YellowLight"; The final activated warning signals are as follows: { "time":["10:00:12","10:00:12"], "sensor":["sensor_A","sensor_C"], "data":[30.0,300.0], "deviation":[1.29,1.0], "alert_level":["Moderate","Minor"], "signal":["OrangeLight","YellowLight"] }; Through the above steps, the corresponding early warning signals were successfully activated according to the results of the anomaly detection process, and timely response and processing were ensured. This not only improves the reliability and response speed of the system, but also provides a solid foundation for subsequent anomaly processing and improvement.

[0067] Step 7: Using the warning signal to update the safety status display panel; specifically including: Defining the layout and elements of the safety status display panel and determining the layout Panel_Layout of each warning signal on the panel can ensure that each warning signal can be displayed clearly and intuitively on the panel. This step provides the basis for subsequent mapping operations, enabling the system to accurately display the warning signal at the specified location.

[0068] Based on the selected warning signal set Signal_Set, each warning signal is mapped to the corresponding position of the safety status display panel. The formula is: Display_Pos=Map(Signal_Set,Panel_Layout), where Display_Pos represents the specific position of the warning signal on the panel, and Map is the mapping operation; through this formula, different warning signals can be placed in predefined positions to ensure the orderliness and consistency of information display.

[0069] According to the mapping results, the status of each warning signal is updated in real time on the security status display panel, including color, icon and text description, to ensure that monitoring personnel can obtain the latest abnormal information in a timely manner. This step improves the system's visualization effect and user interaction experience.

[0070] A snapshot of the updated safety status display panel is recorded and archived, and a log file is generated to track all update activities. This step ensures system traceability and data integrity.

[0071] Example 7 Assume that there is a monitoring area with three sensors A, B, and C, which are used to monitor temperature, humidity, and light intensity respectively. After the anomaly detection process, the following warning signals are obtained: Sensor A (temperature): orange warning light (Moderate); Sensor C (light intensity): yellow warning light (Minor); The layout of the safety status display panel (Panel_Layout) is as follows: Top left: sensor A (temperature); Top right: Sensor B (humidity); Bottom left: sensor C (light intensity); Lower right corner: spare area; Implementation steps: Define the layout and elements of the safety status display panel, and determine the layout of the safety status display panel for each warning signal on the panel Panel_Layout: Define the layout of the safety status display panel as follows: { "layout":[ {"position":"Upper left corner","sensor":"sensor_A"}, {"position":"Upper right corner","sensor":"sensor_B"}, {"position":"lower left corner","sensor":"sensor_C"}, {"position":"lower right corner","sensor":"spare area"} ] }; Based on the selected warning signal set Signal_Set, each warning signal is mapped to the corresponding position of the safety status display panel: Warning signal set (Signal_Set): { "time":["10:00:12","10:00:12"], "sensor":["sensor_A","sensor_C"], "data":[30.0,300.0], "deviation":[1.29,1.0], "alert_level":["Moderate","Minor"], "signal":["OrangeLight","YellowLight"] }; Use the formula Display_Pos=Map(Signal_Set,Panel_Layout) to perform the mapping operation: { "left_top":{"sensor":"sensor_A","signal":"OrangeLight"}, "right_top":{"sensor":"sensor_B","signal":"None"}, "left_bottom":{"sensor":"sensor_C","signal":"YellowLight"}, "right_bottom":{"sensor":"Backup area","signal":"None"} }; According to the mapping results, the status of each warning signal is updated in real time on the safety status display panel, including color, icon and text description: Update the status of the safety status display panel: { "left_top":{"sensor":"sensor_A","color":"orange","icon":"warning_triangle","description":"ModerateAnomalyDetected"}, "right_top":{"sensor":"sensor_B","color":"green","icon":"checkmark","description":"Normal"}, "left_bottom":{"sensor":"sensor_C","color":"yellow","icon":"exclamation_mark","description":"MinorAnomalyDetected"}, "right_bottom":{"sensor":"Backup area","color":"gray","icon":"none","description":"NoData"} }; Record a snapshot of the Security Status Dashboard after it is updated and archive it, and generate a log file to track all update activity: Record a snapshot of the Security Status Dashboard after it is updated: { "timestamp":"2025-03-03T10:00:12", "panel_snapshot":{ "left_top":{"sensor":"sensor_A","color":"orange","icon":"warning_triangle","description":"ModerateAnomalyDetected"}, "right_top":{"sensor":"sensor_B","color":"green","icon":"checkmark","description":"Normal"}, "left_bottom":{"sensor":"sensor_C","color":"yellow","icon":"exclamation_mark","description":"MinorAnomalyDetected"}, "right_bottom":{"sensor":"Backup area","color":"gray","icon":"none","description":"NoData"} } }; Generate a log file to track all update activity: { "log_entries":[ { "timestamp":"2025-03-03T10:00:12", "sensor":"sensor_A", "action":"update", "details":{"color":"orange","icon":"warning_triangle","description":"ModerateAnomalyDetected"} }, { "timestamp":"2025-03-03T10:00:12", "sensor":"sensor_C", "action":"update", "details":{"color":"yellow","icon":"exclamation_mark","description":"MinorAnomalyDetected"} } ] }.

[0072] Through the above steps, the early warning signal was successfully used to update the security status display panel, and timely response and processing were ensured. This not only improved the system's visualization effect and user interaction experience, but also provided a solid foundation for subsequent data analysis and troubleshooting.

[0073] Step 8: Optimizing the collection strategy of the initial environment data set based on the feedback from the safety status display panel; specifically including: By analyzing the feedback information on the security status display panel, we can identify the areas and time periods that frequently trigger warning signals. This step helps determine the areas that need to be focused on in the system and provides a basis for the subsequent optimization of data collection strategies.

[0074] Based on the high-frequency warning areas High_Areas and time periods High_Times, the effectiveness of the current data collection strategy is evaluated, and the coverage of each sensor in these areas and time periods is calculated. The formula is: , where Coverage represents the coverage rate, Collected_D is the amount of data collected, and Required_D is the amount of data required; through this formula, the data coverage of each sensor in the high-frequency warning area and time period can be quantified, thereby evaluating the effectiveness of the current data collection strategy.

[0075] According to the coverage results, the deployment density and sampling frequency of sensors in high-frequency warning areas and time periods are adjusted to ensure more comprehensive and timely data collection. The sensor configuration parameters are updated, recorded as Config_Params, which can make data collection more comprehensive and timely, reduce blind spots and improve the response speed of the system. After updating the sensor configuration parameters (Config_Params), it can better adapt to actual needs and improve overall performance.

[0076] Apply the new sensor configuration parameters Config_Params to the data acquisition system and verify its effect. Confirm the optimization effect by comparing the quality of the new and old data sets and the frequency of warning signals. Record the optimization results and update the acquisition strategy document.

[0077] Example 8 Assume that there is a monitoring area with three sensors A, B, and C, which are used to monitor temperature, humidity, and light intensity respectively. After a period of operation, the areas and time periods where warning signals are frequently triggered on the safety status display panel are as follows: High_Areas: Area 1: upper left corner (sensor A); Area 2: lower left corner (sensor C); High_Times: Time period 1: 10:00:00-12:00:00; Time period 2: 16:00:00-18:00:00; Implementation steps: Analyze the feedback information on the safety status display panel and identify the areas High_Areas and time periods High_Times that frequently trigger warning signals: Analyze the feedback information on the safety status display panel and identify the following high-frequency warning areas and time periods: High_Areas=["Upper left corner","Lower left corner"]; High_Times=["10:00:00-12:00:00","16:00:00-18:00:00"]; Based on the high-frequency warning areas High_Areas and time periods High_Times, evaluate the effectiveness of the current data collection strategy and calculate the coverage of each sensor in these areas and time periods: Assume that in the high-frequency warning areas and time periods, the data collection of each sensor is as follows: Sensor A (Temperature): Amount of collected data (Collected_D_A): 500; Required data volume (Required_D_A): 1000; Sensor C (light intensity): Amount of collected data (Collected_D_C): 400; Required data volume (Required_D_C): 800; Using the formula Calculate coverage: Sensor A (Temperature): ; Sensor C (light intensity): ; According to the coverage results, adjust the deployment density and sampling frequency of sensors in high-frequency warning areas and time periods to ensure more comprehensive and timely data collection, and update the sensor configuration parameters, recorded as Config_Params: According to the coverage results (50%), adjust the deployment density and sampling frequency of sensors A and C: Sensor A (Temperature): New sampling frequency: once every 5 minutes (originally once every 10 minutes); New deployment density: one more sensor (originally one sensor); Sensor C (light intensity): New sampling frequency: once every 5 minutes (originally once every 10 minutes); New deployment density: one more sensor (originally one sensor); Updated sensor configuration parameters (Config_Params): Config_Params={ "sensor_A":{"sampling_rate":"5minutes","density":2}, "sensor_C":{"sampling_rate":"5minutes","density":2} }.

[0078] Apply the new sensor configuration parameters Config_Params to the data acquisition system and verify its effect. Confirm the optimization effect by comparing the quality and warning signal frequency of the new and old data sets, record the optimization results and update the acquisition strategy document: After applying the new sensor configuration parameters (Config_Params), re-collect data and verify the effect: New dataset quality: Data integrity has been significantly improved, with coverage increasing from 50% to over 90%.

[0079] Warning signal frequency: In high-frequency warning areas and time periods, the frequency of warning signals decreased by about 30%, indicating that abnormal events were monitored and handled more effectively.

[0080] Record the optimization results and update the collection strategy document: { "optimization_results":[ { "timestamp":"2025-03-03T10:00:00", "sensor":"sensor_A", "old_coverage":50, "new_coverage":92, "old_sampling_rate":"10minutes", "new_sampling_rate":"5minutes", "old_density":1, "new_density":2, "old_alert_frequency":10, "new_alert_frequency":7 }, { "timestamp":"2025-03-03T10:00:00", "sensor":"sensor_C", "old_coverage":50, "new_coverage":91, "old_sampling_rate":"10minutes", "new_sampling_rate":"5minutes", "old_density":1, "new_density":2, "old_alert_frequency":8, "new_alert_frequency":5 } ] }.

[0081] Through the above steps, we successfully optimized the collection strategy of the initial environmental data set based on the feedback from the safety status display panel. This not only improved the completeness and timeliness of data collection, but also reduced the frequency of early warning signals and improved the overall performance of the system.

[0082] On the other hand, the present invention proposes a multi-sensor fusion security monitoring system, such as Figure 2 As shown, including: The data acquisition module is used to obtain the initial environmental data set of the sensor; specifically: define the boundary range of the monitoring area and set the data collection points; deploy sensors based on the data collection points and start preliminary data collection; normalize the preliminary data to make the data format of each sensor consistent, the formula is D_norm=(D-Min_D) / (Max_D-Min_D), where D_norm represents the normalized data, D is the original data, Min_D and Max_D are the minimum and maximum values ​​of the data respectively; integrate the normalized data into a unified data structure to form an initial environmental data set.

[0083] The data synchronization module is used to start the data synchronization process based on the initial environmental data set; specifically: analyze the timestamp information in the initial environmental data set to determine the time base of each sensor data; calculate the time difference of the arrival of each sensor data based on the time base, the formula is Delta_T=T_sensor-T_ref, where Delta_T represents the time difference, T_sensor is the timestamp of the sensor data, and T_ref is the reference time base; adjust the data sending frequency of each sensor according to the time difference to align the time of all sensor data; integrate the time-aligned data streams into a unified time axis to form a synchronized data stream.

[0084] The rate adjustment and time alignment module is used to adjust the data flow rate of each sensor through the data synchronization process to achieve time alignment; specifically: measure the initial rate of each sensor data flow and record it as a reference rate value, denoted as Base_Rate; calculate the difference between the actual rate of each sensor and the reference rate based on the reference rate value, the formula is Diff_Rate=Rate_sensor-Base_Rate, where Diff_Rate represents the rate difference and Rate_sensor is the actual rate of the sensor; adjust the data sending interval of each sensor according to the rate difference so that the data rates of all sensors tend to be consistent, the adjustment formula is Interval_T=Ref_T / (1+Diff_Rate), where Interval_T represents the new data sending interval and Ref_T is the reference time interval; apply the adjusted data sending interval to each sensor, and verify whether the data flow achieves time alignment to ensure that all sensor data are synchronized on a unified time axis.

[0085] The data framework construction module is used to construct a unified data framework using the time-aligned data; specifically: define a unified time base and data format standard to ensure that all sensor data follow the same rules; based on the time base and data format standard, convert the time-aligned data into a unified format, recorded as Unified_D; sort and integrate the data in the unified format, and arrange the data of each sensor in chronological order. The formula is Sorted_D=Sort(Unified_D,Time_Key), where Sorted_D represents the sorted data set and Time_Key is the timestamp key value; create a global data structure and insert the sorted data set into the structure to form a complete unified data framework.

[0086] The anomaly detection module is used to execute the anomaly detection process within the unified data framework; specifically: set the threshold range for anomaly detection, determine the upper and lower limits based on historical data and system requirements, and record them as Upper_Limit and Lower_Limit; calculate the data deviation at each time point based on the time series data in the unified data framework, and the formula is Deviation=(Unified_D-Avg_D) / Std_D, where Deviation represents the deviation, Unified_D is the data point in the unified data framework, Avg_D is the average value, and Std_D is the standard deviation; compare the deviation with the threshold range, mark the data points that exceed the threshold as potential anomaly points, and record them as Anomaly_Points; verify and classify the potential anomaly points, determine whether they are real abnormal events based on the continuity rule, and update the anomaly record table.

[0087] The warning signal activation module is used to activate the corresponding warning signal according to the result of the anomaly detection process; specifically: define different types of anomaly levels and their corresponding warning signals, set the threshold range based on the result of the anomaly detection process, recorded as Level_Thresholds; determine the anomaly level of each anomaly point based on the deviation and type of the potential anomaly point Anomaly_Points, the formula is Alert_Level=Match(Deviation,Level_Thresholds), where Alert_Level represents the anomaly level and Match is a matching operation; according to the result of the anomaly level, select the corresponding warning signal mode, map the anomaly level to a predefined warning signal set, recorded as Signal_Set; activate the selected warning signal and send it to the monitoring center or related equipment to ensure timely response and processing, and update the warning record table to track the processing progress.

[0088] The status display panel update module is used to update the safety status display panel using the warning signal; specifically: define the layout and elements of the safety status display panel, determine the display position and format of each warning signal on the panel, recorded as Panel_Layout; based on the selected warning signal Signal_Set, map each warning signal to the corresponding position of the safety status display panel, the formula is Display_Pos=Map(Signal_Set,Panel_Layout), where Display_Pos represents the specific position of the warning signal on the panel, and Map is the mapping operation; according to the mapping result, update the status of each warning signal in real time on the safety status display panel, including color, icon and text description, to ensure that the information is accurate; record the snapshot of the updated safety status display panel and archive it for subsequent query and analysis, and generate a log file to track all update activities.

[0089] The strategy optimization module is used to optimize the collection strategy of the initial environmental data set according to the feedback from the safety status display panel; specifically, the feedback information on the safety status display panel is analyzed to identify the areas and time periods that frequently trigger warning signals, which are recorded as High_Areas and High_Times; based on the high-frequency warning areas High_Areas and time periods High_Times, the effectiveness of the current data collection strategy is evaluated, and the coverage of each sensor in these areas and time periods is calculated. The formula is , where Coverage represents coverage, Collected_D is the amount of data collected, and Required_D is the amount of data required; according to the coverage result, adjust the deployment density and sampling frequency of sensors in high-frequency warning areas and time periods to ensure more comprehensive and timely data collection, and update the sensor configuration parameters, recorded as Config_Params; apply the new sensor configuration parameters Config_Params to the data collection system and verify its effect. By comparing the quality of the new and old data sets and the frequency of warning signals, confirm the optimization effect, record the optimization results and update the collection strategy document.

[0090] In addition, when the above modules are executed, they are also used to implement other steps of the above multi-sensor fusion security monitoring method, which will not be described one by one here.

[0091] In summary, the present invention realizes time alignment of data by starting the data synchronization process and dynamically adjusting the data flow rate of each sensor, effectively solving the real-time challenge caused by the delay difference of sensor data; this method not only improves the real-time performance of the system and the consistency of data processing, but also enhances the consistency and reliability of data by building a unified data framework and executing anomaly detection process, and reduces data loss and inconsistency.

[0092] In addition, using a unified data framework for anomaly detection can more accurately identify potential risks and activate corresponding warning signals in a timely manner, thereby optimizing the system's warning response capabilities and significantly improving the overall performance and reliability of the multi-sensor fusion security monitoring system.

[0093] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A multi-sensor fusion security monitoring method, characterized in that: The following steps are involved: Acquire an initial environmental data set of the sensor, and start a data synchronization process based on the initial environmental data set; Adjusting the data flow rate of each sensor through the data synchronization process to achieve time alignment, and constructing a unified data framework using the time-aligned data; Executing an anomaly detection process within the unified data framework, and activating a corresponding warning signal based on the result of the anomaly detection process; The safety status display panel is updated using the early warning signal, and the acquisition strategy of the initial environmental data set is optimized according to the feedback of the safety status display panel.

2. A multi-sensor fusion security monitoring method according to claim 1, characterized in that: The step of obtaining an initial environment data set of a sensor includes: Define the boundaries of the monitoring area and set data collection points, deploy sensors based on the data collection points and start preliminary data collection; Normalize the preliminary data to make the data format of each sensor consistent. The formula is D_norm=(D-Min_D) / (Max_D-Min_D), where D_norm represents the normalized data, D is the original data, Min_D and Max_D are the minimum and maximum values ​​of the data respectively; Integrate the normalized data into a unified data structure to form the initial environment dataset.

3. A multi-sensor fusion security monitoring method according to claim 2, characterized in that: The starting of the data synchronization process based on the initial environment data set includes: Analyze the timestamp information in the initial environmental data set to determine the time base of each sensor data; Calculate the time difference of each sensor data arrival based on the time reference, the formula is Delta_T=T_sensor-T_ref, where Delta_T represents the time difference, T_sensor is the timestamp of the sensor data, and T_ref is the reference time reference; The data transmission frequency of each sensor is adjusted according to the time difference to align the time of all sensor data, and the time-aligned data streams are integrated into a unified time axis to form a synchronous data stream.

4. A multi-sensor fusion security monitoring method according to claim 3, characterized in that: The step of adjusting the data flow rate of each sensor through the data synchronization process to achieve time alignment includes: Measure the initial rate of each sensor data stream and record it as the base rate value Base_Rate; Calculate the difference between the actual rate of each sensor and the base rate based on the base rate value. The formula is Diff_Rate=Rate_sensor-Base_Rate, where Diff_Rate represents the rate difference and Rate_sensor is the actual rate of the sensor. According to the rate difference, the data transmission interval of each sensor is adjusted to make the data rate of all sensors consistent. The adjustment formula is Interval_T=Ref_T / (1+Diff_Rate), where Interval_T represents the new data transmission interval and Ref_T is the reference time interval. Apply the adjusted data sending interval to each sensor and verify whether the data stream is time-aligned to ensure that all sensor data are synchronized on the same timeline.

5. A multi-sensor fusion security monitoring method according to claim 4, characterized in that: The step of constructing a unified data framework using the time-aligned data includes: Define unified time base and data format standards to ensure that all sensor data follow the same rules; Based on the time base and data format standards, the time-aligned data is converted into unified format data Unified_D; Sort and integrate the data in a unified format, and arrange the data of each sensor in chronological order. The formula is Sorted_D=Sort(Unified_D,Time_Key), where Sorted_D represents the sorted data set and Time_Key is the timestamp key value; Create a global data structure and insert the sorted data set into the structure to form a complete unified data framework.

6. A multi-sensor fusion security monitoring method according to claim 5, characterized in that: The anomaly detection process is performed within the unified data framework, including: Set the threshold range for anomaly detection and determine the upper limit Upper_Limit and lower limit Lower_Limit based on historical data and system requirements; Based on the time series data in the unified data framework, the data deviation at each time point is calculated using the formula Deviation = (Unified_D-Avg_D) / Std_D, where Deviation represents the deviation, Unified_D is the data point in the unified data framework, Avg_D is the average value, and Std_D is the standard deviation; Compare the deviation with the threshold range and mark the data points exceeding the threshold as potential anomaly points Anomaly_Points; Verify and classify potential anomalies, determine whether they are real abnormal events based on continuity rules, and update the anomaly record table.

7. A multi-sensor fusion security monitoring method according to claim 6, characterized in that: The activating a corresponding warning signal according to the result of the abnormality detection process includes: Define different types of abnormal levels and their corresponding warning signals, and set the abnormal level threshold range Level_Thresholds based on the results of the abnormal detection process; Based on the deviation and type of potential anomaly points Anomaly_Points, determine the anomaly level of each anomaly point. The formula is: Alert_Level=Match(Deviation,Level_Thresholds), where Alert_Level indicates the abnormal level and Match is the matching operation; According to the result of the abnormal level, the corresponding warning signal mode is selected, and the abnormal level is mapped to the predefined warning signal set Signal_Set; Activate selected warning signals and send them to the monitoring center to ensure timely response and processing, and update the warning record table to track the processing progress.

8. A multi-sensor fusion security monitoring method according to claim 7, characterized in that: The method of updating the safety status display panel using the early warning signal comprises: Define the layout and elements of the safety status display panel, and determine the layout Panel_Layout of the safety status display panel for each warning signal on the panel; Based on the selected warning signal set Signal_Set, each warning signal is mapped to the corresponding position of the safety status display panel. The formula is: Display_Pos=Map(Signal_Set,Panel_Layout), where Display_Pos indicates the specific position of the warning signal on the panel, and Map is the mapping operation; According to the mapping results, the status of each warning signal is updated in real time on the safety status display panel, including color, icon and text description; Records and archives a snapshot of the updated security status dashboard and generates a log file to track all update activity.

9. A multi-sensor fusion security monitoring method according to claim 8, characterized in that: Optimizing the collection strategy of the initial environment data set based on the feedback of the safety status display panel includes: Analyze the feedback information on the safety status display panel and identify the areas High_Areas and time periods High_Times that frequently trigger warning signals; Based on the high-frequency warning areas High_Areas and time periods High_Times, the effectiveness of the current data collection strategy is evaluated, and the coverage of each sensor in these areas and time periods is calculated. The formula is: , where Coverage represents coverage, Collected_D is the amount of data collected, and Required_D is the amount of data required; According to the coverage results, adjust the deployment density and sampling frequency of sensors in high-frequency warning areas and time periods to ensure more comprehensive and timely data collection, and update sensor configuration parameters, recorded as Config_Params; Apply the new sensor configuration parameters Config_Params to the data acquisition system and verify its effect. Confirm the optimization effect by comparing the quality of the new and old data sets and the frequency of warning signals. Record the optimization results and update the acquisition strategy document.

10. A multi-sensor fusion security monitoring system for implementing the method according to any one of claims 1 to 9, characterized in that: include: A data acquisition module, used to obtain the initial environmental data set of the sensor; A data synchronization module, used to start a data synchronization process based on the initial environment data set; A rate adjustment and time alignment module, used to adjust the data flow rate of each sensor through the data synchronization process to achieve time alignment; A data framework construction module, used to construct a unified data framework using the time-aligned data; An anomaly detection module, used to perform an anomaly detection process within the unified data framework; An early warning signal activation module is used to activate the corresponding early warning signal according to the result of the abnormality detection process; A status display panel update module, used to update the safety status display panel using the warning signal; A strategy optimization module is used to optimize the collection strategy of the initial environment data set according to the feedback of the security status display panel.

Images