Perimeter intrusion detection system based on optical fiber sensing

Abstract

The invention discloses a perimeter intrusion detection system based on optical fiber sensing, and relates to the technical field of safety monitoring, and the system comprises the following components: an optical fiber sensing subsystem, a multi-mode sensor array, a data fusion processing center and an alarm and response module. By adopting the data fusion processing center, the system has strong data processing and intelligent decision-making capabilities, can perform real-time preprocessing, feature extraction, time synchronization and space calibration on original data from the optical fiber sensing subsystem and the multi-mode sensor array, ensures the consistency and accuracy of the data, and meanwhile, ensures the real-time performance of the optical fiber sensing subsystem and the multi-mode sensor array. By applying an advanced data fusion algorithm and integrating feature information of a plurality of sensors, intelligent fusion decision is carried out, rapid identification and accurate classification of intrusion behaviors are realized, and the efficient data processing mechanism not only improves the response speed of the system, but also greatly reduces the occurrence of misinformation and missing report, and improves the safety of the system. And powerful technical support is provided for perimeter safety protection work.

Description

Technical Field

[0001] The present invention relates to the technical field of security monitoring, and in particular to a perimeter intrusion detection system based on optical fiber sensing. Background Art

[0002] With the continuous development of modern society, perimeter security has become an important topic of concern to all walks of life. Traditional perimeter intrusion detection systems mostly use single sensor technology. Although these systems can realize the function of intrusion detection to a certain extent, their accuracy and reliability are often seriously affected in complex environments and severe weather conditions. At the same time, single sensor technology has blind spots and it is difficult to fully cover the perimeter area, which brings many challenges to security prevention work.

[0003] The data fusion processing center of the traditional perimeter intrusion detection system has many shortcomings. The traditional method is difficult to integrate the feature information from multiple sensors and make intelligent fusion decisions. It is impossible to perform pattern recognition and classification on the fused data and distinguish different types of intrusion behaviors. Therefore, it is particularly important to develop a perimeter intrusion detection system based on fiber optic sensing.

[0004] To sum up, the application of traditional perimeter intrusion detection systems and their data fusion processing centers in complex environments has many limitations. In order to improve the accuracy and reliability of perimeter security prevention, there is an urgent need for a new and efficient perimeter intrusion detection system. The perimeter intrusion detection system based on fiber optic sensing proposed in the present invention realizes comprehensive and real-time monitoring of the perimeter area by combining fiber optic sensing technology, multimodal sensor array and data fusion algorithm, effectively improves the system's recognition ability and response speed to complex intrusion behaviors, and provides strong technical support for perimeter security prevention work. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and provide a perimeter intrusion detection system based on fiber optic sensing. The system realizes comprehensive and real-time monitoring of the perimeter area by combining fiber optic sensing technology, multimodal sensor array and data fusion algorithm, effectively improves the system's recognition ability and response speed to complex intrusion behaviors, and provides strong technical support for perimeter security prevention work.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a perimeter intrusion detection system based on optical fiber sensing, the system comprising an optical fiber sensing subsystem, a multimodal sensor array, a data fusion processing center and an alarm and response module;

[0007] The optical fiber sensing subsystem: adopts distributed optical fiber sensing technology, and monitors the physical quantities of vibration and stress changes caused by intrusion in real time through optical cables laid on perimeter fences or underground. The high sensitivity of the optical fiber sensor can capture tiny disturbance signals and provide basic monitoring data for the system;

[0008] The multimodal sensor array: infrared sensors, radar sensors and sound sensors are deployed at key locations of the perimeter to form a multimodal sensor array. The infrared sensor captures the body temperature information of the intruder. The radar sensor measures the speed, distance and direction of the intruder by transmitting and receiving electromagnetic waves. The sound sensor records the sound characteristics of the intrusion scene.

[0009] The data fusion processing center: receives raw data from the optical fiber sensing subsystem and the multimodal sensor array in real time, pre-processes the received data, extracts key features from the pre-processed data, and these features can characterize different types of intrusion behaviors, performs time synchronization and spatial calibration on data from different sensors, performs correlation analysis across sensor data, and identifies the inherent connections and correlations between different sensor data;

[0010] Once an intrusion is identified, the data fusion processing center will immediately trigger the alarm mechanism, send alarm information to relevant personnel, and link other security equipment to respond quickly and effectively according to the system configuration. The alarm information and processing results will be fed back to the system administrator or user. According to the actual operation situation and user feedback, the data fusion algorithm and system parameters will be continuously optimized to improve the overall performance of the system.

[0011] The alarm and response module: When the data fusion processing center confirms that an intrusion has occurred, the alarm mechanism is immediately triggered to notify security personnel through sound and light alarms, text message notifications, and remote monitoring screen push. At the same time, the system can link other security equipment to achieve rapid response and effective interception.

[0012] Furthermore, the fiber optic sensing subsystem perimeter intrusion detection system defines the monitoring range, accuracy and environmental adaptability requirements, adopts fiber optic cable layout and system architecture, lays and installs fiber optic cables, configures and calibrates fiber optic sensors, and realizes real-time data collection and pre-processing of vibration and stress changes. After the data is transmitted to the processing center for storage, it is combined with multi-modal sensor data for comprehensive analysis to identify intrusion behavior and trigger the alarm mechanism and security equipment to respond in conjunction.

[0013] Furthermore, the multimodal sensor array demand analysis needs to identify the key intrusion paths and risk points of the perimeter, and then determine the optimal deployment locations for infrared, radar and sound sensors. The monitoring needs and environmental conditions must be considered during selection, and tools and materials must be prepared and environmental conditions must be checked before installation.

[0014] Furthermore, the multimodal sensor array is then installed and the sensors are adjusted to the optimal state, the equipment is connected to the data processing center, communication testing and individual debugging are performed, the multimodal sensor array is integrated into the perimeter system, linkage testing and performance optimization are performed, and finally the sensor array is regularly maintained to monitor the operating status and data.

[0015] Furthermore, the data fusion algorithm in the data fusion processing center is:

[0016] Among them: F represents the fusion result, which is used for intelligent fusion decision-making, S i represents the characteristic information of the i-th sensor, W i represents the corresponding weight, α is the adjustment coefficient of this part, C j represents different environmental noise and false alarm signal characteristics, m is the total number of such characteristics, and the function of filtering environmental noise and false alarm signals is achieved by taking the average value and combining it with the coefficient β to offset it. k represents the characteristic value in the kth intrusion behavior model or rule base, is the average value of these characteristic values. This part measures the degree of deviation from the standard model by calculating the variance. γ is its adjustment coefficient. According to the preset intrusion behavior model and rule base, the fused data is pattern recognized and classified to distinguish different types of intrusion behaviors. Where P represents the result of pattern recognition, which is used to distinguish different types of intrusion behaviors; ω is the adjustment coefficient, which is used to control the overall impact; q is the number of different types in the preset intrusion behavior model and rule base; and D l represents the weight coefficient of the lth intrusion behavior type, This part is an exponential function used to measure the fusion data feature vector F f The characteristic vector of the lth intrusion behavior type represents R l The similarity between f is the fused data feature vector, R l is the characteristic vector representation of the lth intrusion behavior type in the model and rule base, |F f -R l | represents the distance between the fused data features and the features of the lth intrusion behavior type, and σ is a scale parameter used to control the influence of the distance on the exponential function and adjust the sensitivity of pattern recognition.

[0017] Furthermore, when making data fusion decisions, the data fusion processing center adjusts the coefficients α, β and γ and optimizes them through machine learning algorithms to improve the accuracy and reliability of the fusion results. The weight W iDynamic allocation is performed based on the sensor's performance and role in the system to give full play to the advantages of each sensor.

[0018] Furthermore, the data fusion processing center performs pattern recognition and classification on the fused data according to the preset intrusion behavior model and rule base. When distinguishing different types of intrusion behaviors, the adjustment coefficient ω is dynamically adjusted according to the security requirements of different scenarios to adapt to different sensitivity requirements. The weight coefficient D l The setting is based on historical intrusion data and expert experience to accurately reflect the importance of different intrusion behavior types, and the scale parameter σ is optimized through experiments and simulations to ensure that the response of the exponential function to distance is consistent with the actual situation.

[0019] Furthermore, when identifying intrusion behavior, the data fusion processing center combines real-time environmental data to calibrate and correct sensor data to improve the stability and accuracy of the system in different environments.

[0020] Furthermore, the data fusion processing center of the alarm and response module comprehensively analyzes the fiber optic sensing subsystem and multimodal sensor data. After confirming an intrusion incident, it immediately triggers the alarm mechanism, activates the sound and light alarm, and notifies the security personnel. At the same time, it pushes the on-site image, and the system links the security equipment cameras and lighting. The security personnel respond quickly and go to deal with the situation. After the incident is handled, the relevant information is recorded, the system monitoring status is restored, and the damaged equipment is checked and repaired.

[0021] Compared with the existing technology, a perimeter intrusion detection system based on optical fiber sensing has the following beneficial effects:

[0022] 1. This system adopts a data fusion processing center, which has powerful data processing and intelligent decision-making capabilities. It can perform real-time preprocessing, feature extraction, time synchronization and spatial calibration on the raw data from the fiber optic sensing subsystem and the multimodal sensor array to ensure the consistency and accuracy of the data. At the same time, by using advanced data fusion algorithms, integrating the feature information of multiple sensors, and making intelligent fusion decisions, it realizes the rapid identification and accurate classification of intrusion behaviors. This efficient data processing mechanism not only improves the response speed of the system, but also greatly reduces the occurrence of false alarms and missed alarms, providing strong technical support for perimeter security prevention work.

[0023] 2. The present invention realizes comprehensive and real-time monitoring of the perimeter area by combining fiber optic sensing technology and multimodal sensor array. The high sensitivity of the fiber optic sensor can capture tiny disturbance signals, while the multimodal sensor array provides rich intruder feature information. This multi-dimensional monitoring method significantly improves the system's ability to identify complex intrusion behaviors, allowing the system to maintain a high degree of accuracy and reliability in complex environments and severe weather conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 An operation flow chart of a perimeter intrusion detection system based on fiber optic sensing. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] Embodiment 1

[0028] Clarify the monitoring range, accuracy requirements, and key indicators of environmental adaptability of the perimeter intrusion detection system.

[0029] According to the actual situation of the perimeter fence or underground pipeline, design the laying path and density of the optical cable to ensure that it can fully cover the monitoring area and capture the vibration and stress changes caused by the intrusion behavior. Determine the interface and communication protocol between the optical fiber sensing subsystem and the multimodal sensor array, data fusion processing center, and alarm and response module. Select optical cables suitable for distributed optical fiber sensing technology to ensure that they have good transmission performance and mechanical strength. According to the designed layout, lay the optical cable on the perimeter fence or in the underground pipeline. Pay attention to keeping the optical cable flat and fixed to avoid additional vibration or stress during the monitoring process. Connect both ends of the optical cable to the input / output port of the optical fiber sensor to ensure a firm connection and stable signal transmission.

[0030] According to the system requirements, configure the parameters of the fiber optic sensor to optimize its response to vibration and stress changes. After the laying is completed and the system is connected correctly, perform a calibration test on the fiber optic sensor. By simulating intrusion behavior, observe and record the response of the sensor to ensure that it can accurately capture small disturbance signals. Start the fiber optic sensing subsystem and start real-time data collection of physical quantities of vibration and stress changes caused by intrusion behavior. Perform denoising, filtering, and amplification preprocessing operations on the collected raw data to improve the signal-to-noise ratio and availability of the data. At the same time, perform time synchronization and format standardization on the data to facilitate subsequent data fusion and analysis.

[0031] The pre-processed data is transmitted to the data fusion processing center by wired or wireless means to ensure the real-time and reliability of data transmission and avoid data loss or delay. A database or data warehouse is established in the data fusion processing center to store the collected fiber optic sensor data and subsequent processing results to ensure data security and traceability. The data fusion processing center receives and processes the data transmitted by the fiber optic sensor subsystem in real time, combines the data of the multi-modal sensor array for comprehensive analysis, processes and analyzes the comprehensive data through preset algorithms or models, identifies potential intrusion behaviors, and immediately triggers the alarm mechanism and notifies relevant personnel once an intrusion incident is confirmed. At the same time, other security equipment is linked to respond quickly and effectively according to the system configuration.

[0032] Analyze the key intrusion paths and potential risk points of the perimeter, determine the best deployment locations for infrared sensors, radar sensors, and sound sensors, ensure that they can fully cover and complement each other's monitoring blind spots, select appropriate infrared sensors, radar sensors, and sound sensors according to monitoring requirements and environmental conditions, ensure that the selected equipment has sufficient sensitivity, stability, and anti-interference capabilities, prepare necessary installation tools and materials, check the environmental conditions of the installation location, ensure that they meet the equipment requirements, install infrared sensors at the selected locations, adjust their angles and heights to best capture the intruder's body temperature information, install radar sensors, ensure that they can clearly transmit and receive electromagnetic waves, measure the intruder's speed, distance, and direction, deploy sound sensors, and select locations that can clearly record the sound characteristics of the intrusion scene to avoid background noise interference;

[0033] Use cables to connect each sensor to the data fusion processing center or the corresponding data acquisition equipment, conduct communication tests between devices to ensure the accuracy and real-time performance of data transmission, debug each sensor individually to ensure its normal operation and accurate reflection of monitoring data, integrate the multimodal sensor array into the perimeter intrusion detection system, conduct linkage tests with other parts of the fiber optic sensing subsystem, simulate intrusion scenarios, test the comprehensive monitoring effect of the multimodal sensor array, adjust sensor parameters and system settings according to the test results, optimize monitoring performance, regularly inspect and maintain the multimodal sensor array to ensure its normal operation and the accuracy of monitoring data, monitor the operating status and monitoring data of the system, and promptly detect and handle abnormal situations.

[0034] Receive raw data from the fiber optic sensing subsystem and multimodal sensor array in real time, pre-process the received data, including but not limited to denoising, filtering, and data format unification to improve data quality and reduce the impact of environmental noise on subsequent analysis, extract key features from the pre-processed data that can characterize different types of intrusion behaviors, perform time synchronization and spatial calibration on data from different sensors to ensure data consistency, perform correlation analysis across sensor data, and identify the inherent connections and correlations between different sensor data;

[0035] Use data fusion algorithms to integrate feature information from multiple sensors and make intelligent fusion decisions. Among them: F represents the fusion result, which is used for intelligent fusion decision-making, S i represents the characteristic information of the i-th sensor, W i represents the corresponding weight, α is the adjustment coefficient of this part, C j represents different environmental noise and false alarm signal characteristics, m is the total number of such characteristics, and the function of filtering environmental noise and false alarm signals is achieved by taking the average value and combining it with the coefficient β to offset it. k represents the characteristic value in the kth intrusion behavior model or rule base, is the average value of these characteristic values. This part measures the degree of deviation from the standard model by calculating the variance. γ is its adjustment coefficient. According to the preset intrusion behavior model and rule base, the fused data is pattern recognized and classified to distinguish different types of intrusion behaviors. Where P represents the result of pattern recognition, which is used to distinguish different types of intrusion behaviors; ω is the adjustment coefficient, which is used to control the overall impact; q is the number of different types in the preset intrusion behavior model and rule base; and D l represents the weight coefficient of the lth intrusion behavior type, This part is an exponential function used to measure the fusion data feature vector F f The characteristic vector of the lth intrusion behavior type represents R l The similarity between f is the fused data feature vector, R l is the characteristic vector representation of the lth intrusion behavior type in the model and rule base, |F f -R l | represents the distance between the fused data features and the features of the lth intrusion behavior type. σ is a scale parameter used to control the influence of distance on the exponential function and adjust the sensitivity of pattern recognition. At the same time, the algorithm must also have the ability to effectively filter environmental noise and false alarm signals to improve the accuracy and reliability of the system.

[0036] Once an intrusion is identified, the data fusion processing center will immediately trigger the alarm mechanism, send alarm information to relevant personnel, and according to the system configuration, link other security equipment for rapid response and effective interception. The alarm information and processing results will be fed back to the system administrator or user for subsequent analysis and evaluation. According to the actual operation situation and user feedback, the data fusion algorithm and system parameters will be continuously optimized to improve the overall performance of the system.

[0037] After the data fusion processing center conducts a comprehensive analysis of the data from the fiber optic sensing subsystem and the multi-modal sensor array, it confirms the occurrence of an intrusion event, verifies the authenticity and urgency of the intrusion event, and eliminates the interference of false alarms and environmental noise. Once the intrusion event is confirmed, the alarm mechanism is immediately triggered, the sound and light alarm equipment is activated, and obvious sound and light signals are emitted near the monitoring center or the perimeter to attract the attention of security personnel. Through the SMS notification system, the intrusion event alarm information is sent to the designated security personnel’s mobile phone number, including the event type, location, and time key information. The push function of the remote monitoring system is used to push the real-time picture or video clips of the intrusion scene to the security personnel so that they can quickly understand the situation on the scene;

[0038] The system automatically or according to preset rules links with other security equipment. If necessary, it can control the access control system to block the intrusion path, or activate the automatic sprinkler device and the physical barrier of the power grid to intercept.

[0039] After receiving the alarm notification, the security personnel will immediately formulate a response strategy based on the information provided and the remote monitoring screen, quickly go to the intrusion site to deal with it, record the intrusion incident, including the time, location, handling process, and result information of the incident, review and analyze the system, evaluate the efficiency and effectiveness of the alarm and response mechanism, and put forward improvement suggestions. If needed, report the intrusion incident and the handling situation to the relevant departments or superior units. After the intrusion incident is handled, restore the system to normal monitoring status, and check and repair security equipment or facilities that may be damaged during the incident handling process.

[0040] Embodiment 2

[0041] A layer of optical cable is laid at the top and bottom of the campus wall to ensure that the vibration and stress changes caused by intrusion can be captured in all directions. A sensor node is set at a certain distance along the wall to ensure full coverage and sensitivity of the signal. High-sensitivity optical fiber sensors are selected to ensure that tiny disturbance signals can be captured. Each sensor node is connected to the optical cable through a dedicated interface and is equipped with a signal amplifier and pre-processing circuit to improve signal stability and transmission quality.

[0042] Infrared sensors are deployed at key entrances and weak points of possible intrusions in the campus wall to capture the body temperature information of intruders. Infrared sensors use non-contact measurement to ensure that they can work effectively at night or in severe weather conditions. Radar sensors are installed on the outside of the wall to measure the speed, distance and direction of intruders by transmitting and receiving electromagnetic waves. The high-precision measurement capability of the radar sensor provides the system with the detailed movement trajectory of the intruder. Sound sensors are set near the campus wall to record the sound characteristics of the intrusion scene. The sound sensor combined with infrared and radar sensors can provide a more comprehensive description of the intrusion behavior.

[0043] The data fusion processing center receives the raw data from the fiber optic sensing subsystem and the multimodal sensor array in real time, pre-processes the received data, including denoising, filtering and data format unification to improve data quality, and uses data fusion algorithms to integrate feature information from multiple sensors to make intelligent fusion decisions. Among them: F represents the fusion result, which is used for intelligent fusion decision-making, S i represents the characteristic information of the i-th sensor, W i The corresponding weights are set according to the importance and reliability of different sensors in the campus wall security system. α is the adjustment coefficient of this part. C j represents different environmental noise and false alarm signal characteristics, m is the total number of such characteristics, and the function of filtering environmental noise and false alarm signals is achieved by taking the average value and combining it with the coefficient β to offset it. k It represents the characteristic value in the kth intrusion behavior model or rule base, E is the average value of these characteristic values, which measures the degree of deviation from the standard model by calculating the variance, and γ is its adjustment coefficient. According to the preset intrusion behavior model and rule base, the fused data is pattern recognized and classified to distinguish different types of intrusion behaviors. Among them, P represents the result of pattern recognition, which is used to distinguish different types of intrusion behaviors, such as climbing the wall, destroying the wall, and trying to break into the specific intrusion behavior type. ω is the adjustment coefficient, which is used to control the overall impact degree. q is the number of different types in the preset intrusion behavior model and rule library. D l represents the weight coefficient of the lth intrusion behavior type, This part is an exponential function used to measure the fusion data feature vector F f The characteristic vector of the lth intrusion behavior type represents R l The similarity between f is the fused data feature vector, R l is the characteristic vector representation of the lth intrusion behavior type in the model and rule base,

[0044] |F f -Rl | represents the distance between the fused data features and the features of the lth intrusion behavior type. σ is a scale parameter used to control the influence of distance on the exponential function and adjust the sensitivity of pattern recognition. The system has the ability to effectively filter environmental noise and false alarm signals to improve accuracy and reliability.

[0045] When the data fusion processing center confirms that an intrusion has occurred, the alarm mechanism is immediately triggered. The alarm methods include sound and light alarms, SMS notifications, and remote monitoring screen push to ensure that security personnel can respond in a timely manner. The system can be linked to other security equipment. After receiving the alarm, the security personnel will quickly go to the scene to deal with it and record relevant information. After the incident is handled, the relevant information will be recorded and the system monitoring status will be restored. Damaged equipment will be regularly inspected and repaired to ensure the long-term stable operation of the system.

[0046] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A perimeter intrusion detection system based on optical fiber sensing, characterized in that: The system includes a fiber optic sensing subsystem, a multimodal sensor array, a data fusion processing center, and an alarm and response module; The optical fiber sensing subsystem: adopts distributed optical fiber sensing technology, and monitors the physical quantities of vibration and stress changes caused by intrusion in real time through optical cables laid on perimeter fences or underground. The high sensitivity of the optical fiber sensor can capture tiny disturbance signals and provide basic monitoring data for the system; The multimodal sensor array: infrared sensors, radar sensors and sound sensors are deployed at key locations of the perimeter to form a multimodal sensor array. The infrared sensor captures the body temperature information of the intruder. The radar sensor measures the speed, distance and direction of the intruder by transmitting and receiving electromagnetic waves. The sound sensor records the sound characteristics of the intrusion scene. The data fusion processing center: receives raw data from the optical fiber sensing subsystem and the multimodal sensor array in real time, pre-processes the received data, extracts key features from the pre-processed data, and these features can characterize different types of intrusion behaviors, performs time synchronization and spatial calibration on data from different sensors, performs correlation analysis across sensor data, and identifies the inherent connections and correlations between different sensor data; Once an intrusion is identified, the data fusion processing center will immediately trigger the alarm mechanism, send alarm information to relevant personnel, and link other security equipment to respond quickly and effectively according to the system configuration. The alarm information and processing results will be fed back to the system administrator or user. According to the actual operation situation and user feedback, the data fusion algorithm and system parameters will be continuously optimized to improve the overall performance of the system. The alarm and response module: When the data fusion processing center confirms that an intrusion has occurred, the alarm mechanism is immediately triggered to notify security personnel through sound and light alarms, text message notifications, and remote monitoring screen push. At the same time, the system can link other security equipment to achieve rapid response and effective interception.

2. The perimeter intrusion detection system based on optical fiber sensing according to claim 1, characterized in that: The fiber optic sensing subsystem perimeter intrusion detection system defines the monitoring range, accuracy and environmental adaptability requirements, adopts fiber optic cable layout and system architecture, lays and installs fiber optic cables, configures and calibrates fiber optic sensors, and realizes real-time data collection and preprocessing of vibration and stress changes. After transmission to the processing center for storage, it combines multi-modal sensor data for comprehensive analysis to identify intrusion behavior and trigger alarm mechanisms and security equipment to respond in conjunction.

3. The perimeter intrusion detection system based on optical fiber sensing according to claim 1, characterized in that: The multimodal sensor array demand analysis needs to identify the key intrusion paths and risk points of the perimeter, and then determine the optimal deployment locations for infrared, radar, and sound sensors. The monitoring needs and environmental conditions must be considered during selection, and tools and materials must be prepared and environmental conditions checked before installation.

4. The perimeter intrusion detection system based on optical fiber sensing according to claim 3, characterized in that: The multimodal sensor array is then installed and the sensors are adjusted to the optimal state, the equipment is connected to the data processing center, communication testing and individual debugging are performed, the multimodal sensor array is integrated into the perimeter system, linkage testing and performance optimization are performed, and finally the sensor array is regularly maintained to monitor the operating status and data.

5. The perimeter intrusion detection system based on optical fiber sensing according to claim 1, characterized in that: The data fusion algorithm in the data fusion processing center is: Among them: F represents the fusion result, which is used for intelligent fusion decision-making, S i represents the characteristic information of the i-th sensor, W i represents the corresponding weight, α is the adjustment coefficient of this part, C j represents different environmental noise and false alarm signal characteristics, m is the total number of such characteristics, and the function of filtering environmental noise and false alarm signals is achieved by taking the average value and combining it with the coefficient β to offset it. k represents the characteristic value in the kth intrusion behavior model or rule base, is the average value of these characteristic values. This part measures the degree of deviation from the standard model by calculating the variance. γ is its adjustment coefficient. According to the preset intrusion behavior model and rule base, the fused data is pattern recognized and classified to distinguish different types of intrusion behaviors. Where P represents the result of pattern recognition, which is used to distinguish different types of intrusion behaviors; ω is the adjustment coefficient, which is used to control the overall impact; q is the number of different types in the preset intrusion behavior model and rule base; and D l represents the weight coefficient of the lth intrusion behavior type, This part is an exponential function used to measure the fusion data feature vector F f The characteristic vector of the lth intrusion behavior type represents R l The similarity between f is the fused data feature vector, R l is the characteristic vector representation of the lth intrusion behavior type in the model and rule base, |F f -R l | represents the distance between the fused data features and the features of the lth intrusion behavior type, and σ is a scale parameter used to control the influence of the distance on the exponential function and adjust the sensitivity of pattern recognition.

6. The perimeter intrusion detection system based on optical fiber sensing according to claim 5, characterized in that: When making data fusion decisions, the data fusion processing center adjusts the coefficients α, β and γ and optimizes them through machine learning algorithms to improve the accuracy and reliability of the fusion results. i Dynamic allocation is performed based on the sensor's performance and role in the system to give full play to the advantages of each sensor.

7. The perimeter intrusion detection system based on optical fiber sensing according to claim 1, characterized in that: The data fusion processing center performs pattern recognition and classification on the fused data according to the preset intrusion behavior model and rule base. When distinguishing different types of intrusion behaviors, the adjustment coefficient ω is dynamically adjusted according to the security requirements of different scenarios to adapt to different sensitivity requirements. The weight coefficient D l The scale parameter σ is set based on historical intrusion data and expert experience to accurately reflect the importance of different intrusion behavior types, and is optimized through experiments and simulations.

8. The perimeter intrusion detection system based on optical fiber sensing according to claim 1, characterized in that: When identifying intrusion behavior, the data fusion processing center combines the real-time environmental data to calibrate and correct the sensor data.

9. The perimeter intrusion detection system based on optical fiber sensing according to claim 1, characterized in that: The data fusion processing center of the alarm and response module comprehensively analyzes the fiber optic sensing subsystem and multimodal sensor data. After confirming an intrusion event, it immediately triggers the alarm mechanism, activates the sound and light alarm, and notifies the security personnel. At the same time, it pushes the on-site image, and the system links the security equipment cameras and lighting. The security personnel respond quickly and go to deal with the situation. After the incident is handled, the relevant information is recorded, the system monitoring status is restored, and the damaged equipment is checked and repaired.

Images