Optical cable routing anti-intrusion method and system

Abstract

The invention discloses an optical cable routing anti-intrusion method and system, and the method comprises the following steps: laying a sensor network at key nodes along an optical cable, monitoring the data of a surrounding environment in real time, constructing an environment reference model, and integrating the model to a GIS platform, thereby achieving the risk assessment of different regions. Different monitoring frequencies and sensitivities are set for a high-risk area and a low-risk area according to an evaluation result, an anomaly detection module compares real-time signal data with an environment reference model, high-credibility abnormal signals are screened out through self-adaptive threshold adjustment, and when a physical barrier module detects abnormal external force, the physical barrier module sends out the abnormal external force. Triggering the area and the adjacent area to enter a high-frequency monitoring mode to form a dense monitoring area, positioning an intrusion position through a sensor network and a GIS platform, generating a path prediction model, and dynamically adjusting the frequency, sensitivity and alarm threshold value of the monitoring area based on data analysis and GIS geographic information. The adaptability and the resource configuration efficiency of the system in different environments are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of optical cable line security protection, and in particular to an optical cable routing anti-intrusion method and system. Background Art

[0002] With the widespread application of fiber-optic communications, optical cable lines have become an important infrastructure for information transmission. However, optical cable lines are usually deployed in the wild or remote areas. The environment in these areas is complex and easily interfered by human or natural factors. Traditional optical cable protection methods mainly rely on regular manual inspections and simple monitoring equipment. These methods have great limitations and cannot meet the real-time intrusion monitoring needs over long distances and in complex terrains. Especially when the optical cable route is damaged or invaded, there is a lack of effective response mechanisms and accurate positioning methods, making it difficult for maintenance personnel to grasp the intrusion location in time and deal with it, resulting in serious threats to the security and stability of the communication network.

[0003] Existing optical cable intrusion prevention technologies usually adopt centralized sensor deployment or local monitoring equipment, such as video surveillance and infrared sensors. However, these methods have many defects in practical applications. The monitoring coverage is limited. Traditional monitoring equipment is difficult to cover long-distance optical cable lines, especially in remote areas with complex terrain. Its monitoring range is severely limited. Centralized sensor deployment requires dense sensor nodes, which is often difficult to achieve in practical applications, especially in mountainous areas or wild environments. Insufficient coverage leads to an increase in monitoring blind spots, making it difficult to detect intrusions in a timely manner.

[0004] Existing protection measures usually rely on regular inspections and cannot achieve real-time monitoring and response. This approach not only increases maintenance costs, but also easily causes optical cable lines to be damaged during inspection intervals, and cannot effectively prevent intrusion incidents. When intrusion or sabotage occurs, traditional means lack the ability to respond quickly, resulting in an increased risk of communication interruption or optical cable damage, and cannot meet the needs of efficient intrusion prevention.

[0005] In actual applications, the environmental factors around optical cable lines are relatively complex. Common factors such as wind, rain, and animal activities will interfere with the signal acquisition of the sensor, resulting in a high false alarm rate. Existing detection technologies are difficult to effectively distinguish between environmental noise and real intrusion behavior, and cannot accurately identify and filter non-intrusion signals, resulting in the monitoring efficiency and accuracy of the protection system being affected, and it is difficult to make reliable judgments on abnormal signals.

[0006] The existing system also faces the problem of low positioning accuracy and unclear intrusion location, especially in long-distance optical cable lines. Many monitoring devices are unable to accurately locate the specific location of the intrusion behavior, and maintenance personnel are unable to determine the intrusion point in a timely and accurate manner, which delays the response and repair time, resulting in low emergency response efficiency and affecting the protection effect of the optical cable. Traditional optical cable intrusion prevention technology still mainly relies on static monitoring strategies, which are difficult to dynamically adjust according to actual conditions. Due to the lack of analysis and feedback on historical intrusion events, the existing system is difficult to gradually optimize the monitoring strategy and cannot adapt to complex environmental changes. The protection effect gradually weakens in long-term application.

[0007] Therefore, how to provide an optical cable routing anti-intrusion method and system is a problem that those skilled in the art need to solve urgently. Summary of the invention

[0008] One purpose of the present invention is to propose an optical cable routing anti-intrusion method and system. The present invention realizes real-time monitoring and intelligent protection of optical cable lines through distributed sensor networks, environmental benchmark models, GIS platform integration and dynamic risk assessment, makes full use of the layout of sensor networks, multi-layer processing of environmental data and the combination of GIS geographic information, describes in detail how to collect environmental data, intrusion detection and path prediction for key nodes along the optical cable, and dynamically adjusts monitoring parameters through historical data analysis to provide optimal resource configuration, and has the advantages of wide monitoring range, low false alarm rate, accurate positioning and fast response speed.

[0009] An optical cable routing anti-intrusion method and system according to an embodiment of the present invention comprises the following steps:

[0010] S1. Distributed sensor devices are deployed at key nodes of the optical cable line to collect initial environmental data around the optical cable, generate an environmental benchmark model including geographical conditions, climate characteristics and background noise, and integrate the benchmark model into the GIS management platform;

[0011] S2. Based on the environmental benchmark model, the optical cable routing is risk graded and divided into high-risk and low-risk areas. The corresponding monitoring frequency and sensitivity are set for different risk areas. The high-risk area is configured as an intensive monitoring mode, and the low-risk area is configured as a periodic monitoring mode to achieve resource allocation optimization;

[0012] S3. During real-time monitoring, the collected signal data is compared with the environmental benchmark model, abnormal signals are identified through the self-calibration anomaly detection algorithm, and the detection threshold is adaptively adjusted according to environmental noise, natural changes and daily vibration dynamic data to form screened high-confidence abnormal signal data;

[0013] S4. Physical barriers are deployed in key areas of the optical cable line. When the physical barriers detect abnormal external forces, the sensors in the area and its adjacent areas are triggered to enter a high-frequency monitoring state. Based on the screened high-reliability abnormal signal data, a dense monitoring area is formed locally to optimize the ability to capture intrusion features.

[0014] S5. After confirming the intrusion, the location of the intrusion event is accurately located through the GIS platform, and the geographic coordinates, event information and path guidance are transmitted to the monitoring center through the 4G / 5G network. At the same time, an early warning signal is sent to the mobile terminal of the maintenance personnel to support response and on-site processing;

[0015] S6. The time, location, characteristic data and processing results of each intrusion event are recorded in the historical database, and the monitoring frequency, sensitivity and alarm threshold of each monitoring area are dynamically adjusted based on historical data analysis and geographic information of the GIS platform to optimize the system's adaptability and response efficiency in different environments.

[0016] Optionally, the S1 specifically includes:

[0017] S11. Distributed optical fiber sensors are deployed along the optical cable line. The sensors are deployed at a preset dynamic spacing. The spacing is adaptively adjusted according to the complexity of the terrain along the line, the frequency of environmental changes and the risk level, so as to form a sensor network covering the optical cable line, which is used to collect data on the vibration frequency, temperature, humidity and background noise intensity of the environment around the optical cable;

[0018] S12. Perform multi-layer preprocessing on the collected environmental data, clean the data to remove outliers, eliminate short-term fluctuation interference through dynamic smoothing algorithm, and generate a standardized data set D 标准 , and is associated with the sensor location;

[0019] S13. Based on the standardized dataset D 标准 , combined with the geographical characteristics and historical climate data along the optical cable, to generate the environmental benchmark model M 环境 The model includes geographical condition benchmark, climate characteristic benchmark and noise benchmark. The geographical condition benchmark is based on terrain undulation and human activity distribution information, the climate characteristic benchmark is based on historical temperature, humidity and seasonal variation, and the noise benchmark determines the environmental noise characteristics through hierarchical spectrum analysis.

[0020] S14, the environmental benchmark model M 环境 It is integrated into the GIS management platform and stored in the monitoring system database. The benchmark model is updated at set time intervals through an adaptive update algorithm to adapt to dynamic changes in the environment and continuously optimize the accuracy of signal comparison and anomaly detection.

[0021] Optionally, the S2 specifically includes:

[0022] S21, based on the generated environmental benchmark model M 环境 , using the data of geographical condition benchmark, climate characteristic benchmark and noise benchmark, combined with historical intrusion event data, the risk assessment factor R is calculated and generated to divide the risk area along the optical cable:

[0023]

[0024] Where n is the number of monitoring points along the line, W G , W C and W N is the weight coefficient of geographical conditions, climate characteristics and noise benchmark, g i 、c i and n i represent the geographical condition value, climate characteristic value and noise reference value of the ith monitoring point respectively, and f G (g i ), f C (c i ) and f N (n i ) is the normalized function of each benchmark parameter, H(E) is the entropy value of the frequency of historical intrusion events, P(I) is the predicted intrusion probability function, α, β and γ are adjustment coefficients;

[0025] S22. According to the risk assessment factor R, the monitoring area along the optical cable is divided into high-risk area, second-high-risk area and low-risk area, with the high-risk area as the priority monitoring area, the second-high-risk area as the key monitoring area, and the low-risk area as the periodic monitoring area, and the corresponding monitoring levels are set;

[0026] S23. Set up a high-frequency intensive monitoring mode in high-risk areas, and set the monitoring frequency to f 高 , the sensitivity is set to S 高 , the frequency and sensitivity are adaptively adjusted through real-time data feedback, and the monitoring intensity is automatically adjusted in response to fluctuations in the risk assessment factor R;

[0027] S24: Set up adaptive monitoring mode in the second highest risk area, and set the monitoring frequency to f 次高 , the sensitivity is set to S 次高 , monitoring frequency and sensitivity are based on the environmental benchmark model M 环境 Dynamic adjustment of historical climate change data and current environmental fluctuations;

[0028] S25. Set a low-frequency periodic monitoring mode in low-risk areas, and set the monitoring frequency to f 低 , the sensitivity is set to S 低, low-risk areas maintain low-frequency monitoring mode and sensitivity within the set period to achieve basic monitoring coverage and resource optimization;

[0029] S26. Establish a parameter database for monitoring configurations in different risk areas and compare the monitoring configuration parameters with the environmental benchmark model M 环境 Based on the real-time data and the changes in the risk assessment factor R, the monitoring parameters are updated regularly and adaptively adjusted;

[0030] S27. Set time intervals and environmental change trigger mechanisms so that the adaptive monitoring configuration database can periodically recalculate the risk assessment factor R, and update the monitoring frequency and sensitivity of high-risk, second-highest-risk, and low-risk areas based on dynamic changes in the environmental benchmark model, so that the monitoring system matches environmental changes and achieves effectiveness in resource allocation and anomaly detection.

[0031] Optionally, the S3 specifically includes:

[0032] S31. During real-time monitoring, the collected signal data is compared with the standardized data set D 标准 and environmental benchmark model M 环境 Comparison is performed to detect abnormal fluctuations in the signal and to identify the characteristics of the abnormal signal based on the geographic condition benchmark, climate characteristic benchmark and noise benchmark;

[0033] S32, using the self-calibration anomaly detection algorithm to compare and analyze the results, dynamically adjust the detection threshold according to the benchmark parameters in the environmental benchmark model, and generate an adaptive abnormal signal threshold T 自适应 :

[0034] T 自适应 =δ·(W G ·G 基准 +W C ·C 基准 +W N ·N 基准 )+κ·H(E)e -λt ;

[0035] Among them, W G , W C and W N are the weighting factors of geographical conditions, climate characteristics and noise benchmark, G 基准 , C 基准 and N 基准 denote geographic, climate and noise baseline parameters respectively, δ is the adjustment coefficient, H(E) denotes the entropy value of historical invasion events, κ is the adjustment factor, and λ is the time decay coefficient;

[0036] S33, comparing the signal data with the generated adaptive threshold T 自适应Compare and filter out highly reliable abnormal signal data D 异常 ,This dataset reflects possible intrusion events after threshold adjustment;

[0037] S34, the screened high-confidence abnormal signal data D 异常 The data is stored in the monitoring system database and associated with historical anomaly data for long-term analysis and optimization of subsequent anomaly detection models, thereby optimizing the system's adaptability to complex environments and the accuracy of identifying intrusion behaviors.

[0038] Optionally, the S4 specifically includes:

[0039] S41. Deploy physical barrier devices in key areas of optical cables. When the physical barrier detects abnormal external force, the collected signal data is compared with the screened high-reliability abnormal signal data D 异常 Perform a comparison to confirm whether the physical interference meets the characteristics of the intrusion signal;

[0040] S42, after confirming that the abnormal signal meets the intrusion characteristics, the distributed optical fiber sensors in the adjacent areas are triggered to enter the high-frequency monitoring state, and the monitoring frequency is set to f 高 , and dynamically adjust based on the risk assessment factor R:

[0041]

[0042] Among them, f 0 is the basic monitoring frequency, α is the adjustment coefficient, R is the risk assessment factor, g i ,c i ,n i Respectively represent the geographical, climate and noise characteristic values ​​of the ith monitoring point;

[0043] S43, using the high-risk area parameters in the adaptive monitoring configuration database, compare the signal data of the densely monitored area with the adaptive abnormal signal threshold T 自适应 Compare and select intrusion feature data with higher credibility D 高可信 :

[0044]

[0045] Among them, γ is the adjustment factor, H(E) is the entropy value of historical intrusion events, P(I) is the predicted intrusion probability, μ is the adjustment coefficient, R 临界 is the critical risk assessment factor;

[0046] S44, the selected high-credibility intrusion feature data D 高可信It is transmitted to the monitoring system database and combined with historical intrusion data to generate real-time warning information, which is sent to the remote monitoring center and maintenance personnel to form a local intensive monitoring area to optimize the capture capability of intrusion characteristics and support rapid response.

[0047] Optionally, the S5 specifically includes:

[0048] S51, after confirming the high-confidence intrusion feature data D 高可信 Finally, the distributed sensor network is used to accurately locate the geographical location of the intrusion signal, and the location coordinates (x, y, z) of the intrusion point are calculated and determined based on the GIS platform and the risk assessment factor R:

[0049]

[0050] Where n is the number of reference monitoring points used for positioning, R i is the risk assessment factor of the ith monitoring point, is the spatial coordinate of the i-th monitoring point (x i ,y i ,z i );

[0051] S52, compare the located intrusion location data (x, y, z) with the environmental benchmark model M 环境 Compare and combine the geographical condition benchmark and climate characteristic benchmark to calculate the path risk coefficient R 路径 :

[0052]

[0053] Where m is the number of monitoring areas involved in the path evaluation, W G and W C are weight factors for geographical conditions and climate characteristics, G 基准,j and C 基准,j denote the geographic and climatic benchmark parameters of the jth region, respectively;

[0054] S53, based on the path risk coefficient R 路径 , generate an intrusion path prediction model, dynamically identify the optimal response path from the intrusion point to the key node, mark the areas with high path risk and generate path guidance, and provide real-time path guidance for the emergency response team;

[0055] S54. The path prediction model and intrusion location data are stored in the monitoring system database through the 4G / 5G network, the warning information and path guidance are transmitted to the maintenance personnel's mobile terminal through the GIS platform, and the relevant data are stored in the monitoring system database to support path optimization, intrusion feature analysis and dynamic adjustment of protection strategies.

[0056] Optionally, the S6 specifically includes:

[0057] S61, each detected intrusion event information, including the intrusion time t 入侵 , location coordinates (x, y, z), risk assessment factor R and high-confidence intrusion feature data D 高可信 , recorded in the monitoring system database, and associated with the geographic information of the GIS platform to form an event record set E 记录 For the management and analysis of historical data;

[0058] S62, event record set E 记录 Perform cluster analysis and calculate the feature similarity S between events 事件 , generate an intrusion event clustering model:

[0059]

[0060] Where n is the number of historical events, R i and D 高可信,i are the risk assessment factor and high-confidence intrusion feature data of the i-th historical event respectively;

[0061] S63. Based on the intrusion event clustering model, the frequency distribution of each cluster is calculated, and the historical event frequency is weighted using the time decay factor λ to obtain the weighted intrusion frequency F 加权 :

[0062]

[0063] Where m is the number of events in the clustering model, f j is the frequency of occurrence of the jth event, t 当前 is the current time, t j is the occurrence time of the jth event, λ is the time decay factor;

[0064] S64, according to the weighted intrusion frequency F 加权 Based on the results of event similarity analysis and combined with the geographic information provided by the GIS platform, the monitoring frequency, sensitivity and alarm threshold of each monitoring area are dynamically adjusted, and the updated monitoring parameters are stored in the monitoring system database to optimize the system's adaptability and response efficiency in different environments and achieve optimization of resource allocation.

[0065] Optional modules include:

[0066] Distributed sensor module: Deploy sensor networks along optical cable lines to collect environmental data and build environmental benchmark models to adapt to different geographical and climatic conditions;

[0067] Environmental benchmark model module: pre-processes the collected environmental data, generates standardized data sets, builds benchmark models of geography, climate, and noise, supports subsequent risk assessment and anomaly detection, and integrates into the GIS platform;

[0068] Risk assessment module: Based on the environmental baseline model and historical intrusion data, the risk assessment factor is calculated, the monitoring area is divided into high-risk, second-highest-risk and low-risk areas, and the monitoring frequency and sensitivity are set respectively to optimize resource allocation;

[0069] Anomaly detection module: compares signal data with the environmental benchmark model and selects highly reliable abnormal signals through dynamic threshold adjustment;

[0070] Physical barrier trigger module: A physical barrier is set up in the key area of ​​the optical cable. When abnormal external force is detected, high-frequency monitoring of the area and adjacent areas is triggered to form a dense monitoring area;

[0071] Intrusion positioning module: uses distributed sensors and high-reliability abnormal signal data, combined with the GIS platform for positioning, calculates the location coordinates of the intrusion event, and transmits the location information to the monitoring system database;

[0072] Path prediction and response module: Generates an intrusion path prediction model based on path risk assessment, dynamically determines the optimal response path from the intrusion point to the key node, marks high-risk paths, and provides real-time path guidance and response suggestions for the emergency response team;

[0073] Historical data recording and adaptive adjustment module: records the data of each intrusion event, and dynamically adjusts it through cluster analysis and weighted frequency combined with GIS geographic information to optimize the monitoring frequency, sensitivity and alarm threshold of the monitoring area, and optimize the system's adaptive ability and resource allocation efficiency.

[0074] The beneficial effects of the present invention are:

[0075] The system of the present invention deploys a distributed fiber optic sensor network along the optical cable, and combined with an innovative environmental benchmark model, can effectively adapt to different geographical conditions and climate changes, and comprehensively collect and monitor environmental data such as vibration, temperature, humidity and background noise in real time. Based on an adaptive risk assessment model, the system can dynamically classify the areas along the optical cable line, and divide the monitoring areas into high-risk, second-highest-risk and low-risk areas, thereby achieving optimal allocation of resources and multi-level monitoring strategies. This effectively solves the problems of incomplete monitoring coverage and low inspection efficiency in the prior art, and achieves a wider monitoring scope and more reasonable monitoring resource allocation.

[0076] The system's anomaly detection module adopts a self-calibration anomaly detection algorithm and an adaptive threshold adjustment mechanism, which can effectively distinguish intrusion signals from environmental noise in complex environments, significantly reduce the false alarm rate, and ensure the monitoring efficiency and accuracy of the system. Through the physical barrier trigger mechanism, once abnormal external force is detected, it can trigger high-frequency monitoring of the area and adjacent areas to form a local intensive monitoring area, thereby quickly capturing intrusion characteristics and improving the system's response speed. Combined with the innovative path prediction module, the system can accurately locate the location of the intrusion incident and provide the emergency response team with the optimal path guidance based on the path risk assessment model, effectively shortening the response time and further improving the efficiency of intrusion handling.

[0077] Through innovative historical data clustering analysis and time-attenuated weighted frequency models, the system dynamically adjusts the monitoring frequency and sensitivity of each monitoring area based on the feature similarity and time weighting of intrusion events to achieve adaptive intelligent adjustment. Through in-depth use of historical data, the system can automatically optimize monitoring strategies according to environmental and event changes, so that in complex and changing environments, the protection system has higher adaptability and stability. This data-driven dynamic adjustment mechanism not only reduces maintenance costs, but also improves the long-term protection efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0079] Figure 1 A flowchart of an optical cable routing anti-intrusion method and system proposed by the present invention;

[0080] Figure 2 A flow chart of data collection and benchmark model construction for the distributed sensor module and the environmental benchmark model module in the present invention;

[0081] Figure 3 It is a dynamic adjustment flow chart of the historical data recording and adaptive adjustment module in the present invention. DETAILED DESCRIPTION

[0082] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0083] refer to Figure 1-3 , a method and system for preventing intrusion in optical cable routing, comprising the following steps:

[0084] S1. Distributed sensor devices are deployed at key nodes of the optical cable line to collect initial environmental data around the optical cable, generate an environmental benchmark model including geographical conditions, climate characteristics and background noise, and integrate the benchmark model into the GIS management platform;

[0085] S2. Based on the environmental benchmark model, the optical cable routing is risk graded and divided into high-risk and low-risk areas. The corresponding monitoring frequency and sensitivity are set for different risk areas. The high-risk area is configured as an intensive monitoring mode, and the low-risk area is configured as a periodic monitoring mode to achieve resource allocation optimization;

[0086] S3. During real-time monitoring, the collected signal data is compared with the environmental benchmark model, abnormal signals are identified through the self-calibration anomaly detection algorithm, and the detection threshold is adaptively adjusted according to environmental noise, natural changes and daily vibration dynamic data to form screened high-confidence abnormal signal data;

[0087] S4. Physical barriers are deployed in key areas of the optical cable line. When the physical barriers detect abnormal external forces, the sensors in the area and its adjacent areas are triggered to enter a high-frequency monitoring state. Based on the screened high-reliability abnormal signal data, a dense monitoring area is formed locally to optimize the ability to capture intrusion features.

[0088] S5. After confirming the intrusion, the location of the intrusion event is accurately located through the GIS platform, and the geographic coordinates, event information and path guidance are transmitted to the monitoring center through the 4G / 5G network. At the same time, an early warning signal is sent to the mobile terminal of the maintenance personnel to support response and on-site processing;

[0089] S6. The time, location, characteristic data and processing results of each intrusion event are recorded in the historical database, and the monitoring frequency, sensitivity and alarm threshold of each monitoring area are dynamically adjusted based on historical data analysis and geographic information of the GIS platform to optimize the system's adaptability and response efficiency in different environments.

[0090] In this implementation, S1 specifically includes:

[0091] S11. Distributed optical fiber sensors are deployed along the optical cable line. The sensors are deployed at a preset dynamic spacing. The spacing is adaptively adjusted according to the complexity of the terrain along the line, the frequency of environmental changes and the risk level, so as to form a sensor network covering the optical cable line, which is used to collect data on the vibration frequency, temperature, humidity and background noise intensity of the environment around the optical cable;

[0092] S12. Perform multi-layer preprocessing on the collected environmental data, clean the data to remove outliers, eliminate short-term fluctuation interference through dynamic smoothing algorithm, and generate a standardized data set D 标准 , and is associated with the sensor location;

[0093] S13. Based on the standardized dataset D 标准 , combined with the geographical characteristics and historical climate data along the optical cable, to generate the environmental benchmark model M 环境 The model includes geographical condition benchmark, climate characteristic benchmark and noise benchmark. The geographical condition benchmark is based on terrain undulation and human activity distribution information, the climate characteristic benchmark is based on historical temperature, humidity and seasonal variation, and the noise benchmark determines the environmental noise characteristics through hierarchical spectrum analysis.

[0094] S14, the environmental benchmark model M 环境 It is integrated into the GIS management platform and stored in the monitoring system database. The benchmark model is updated at set time intervals through an adaptive update algorithm to adapt to dynamic changes in the environment and continuously optimize the accuracy of signal comparison and anomaly detection.

[0095] In this implementation, S2 specifically includes:

[0096] S21, based on the generated environmental benchmark model M 环境 , using the data of geographical condition benchmark, climate characteristic benchmark and noise benchmark, combined with historical intrusion event data, the risk assessment factor R is calculated and generated to divide the risk area along the optical cable:

[0097]

[0098] Where n is the number of monitoring points along the line, W G , W C and W N is the weight coefficient of geographical conditions, climate characteristics and noise benchmark, g i 、c i and n i represent the geographical condition value, climate characteristic value and noise reference value of the ith monitoring point respectively, and f G (g i ), f C (c i ) and f N (n i ) is the normalized function of each benchmark parameter, H(E) is the entropy value of the frequency of historical intrusion events, P(I) is the predicted intrusion probability function, α, β and γ are adjustment coefficients;

[0099] S22. According to the risk assessment factor R, the monitoring area along the optical cable is divided into high-risk area, second-high-risk area and low-risk area, with the high-risk area as the priority monitoring area, the second-high-risk area as the key monitoring area, and the low-risk area as the periodic monitoring area, and the corresponding monitoring levels are set;

[0100] S23. Set up a high-frequency intensive monitoring mode in high-risk areas, and set the monitoring frequency to f 高 , the sensitivity is set to S 高 , the frequency and sensitivity are adaptively adjusted through real-time data feedback, and the monitoring intensity is automatically adjusted in response to fluctuations in the risk assessment factor R;

[0101] S24: Set up adaptive monitoring mode in the second highest risk area, and set the monitoring frequency to f 次高 , the sensitivity is set to S 次高 , monitoring frequency and sensitivity are based on the environmental benchmark model M 环境 Dynamic adjustment of historical climate change data and current environmental fluctuations;

[0102] S25. Set a low-frequency periodic monitoring mode in low-risk areas, and set the monitoring frequency to f 低 , the sensitivity is set to S 低 , low-risk areas maintain low-frequency monitoring mode and sensitivity within the set period to achieve basic monitoring coverage and resource optimization;

[0103] S26. Establish a parameter database for monitoring configurations in different risk areas and compare the monitoring configuration parameters with the environmental benchmark model M 环境 Based on the real-time data and the changes in the risk assessment factor R, the monitoring parameters are updated regularly and adaptively adjusted;

[0104] S27. Set time intervals and environmental change trigger mechanisms so that the adaptive monitoring configuration database can periodically recalculate the risk assessment factor R, and update the monitoring frequency and sensitivity of high-risk, second-highest-risk, and low-risk areas based on dynamic changes in the environmental benchmark model, so that the monitoring system matches environmental changes and achieves effectiveness in resource allocation and anomaly detection.

[0105] In this implementation, S3 specifically includes:

[0106] S31. During real-time monitoring, the collected signal data is compared with the standardized data set D 标准 and environmental benchmark model M 环境 Comparison is performed to detect abnormal fluctuations in the signal and to identify the characteristics of the abnormal signal based on the geographic condition benchmark, climate characteristic benchmark and noise benchmark;

[0107] S32, using the self-calibration anomaly detection algorithm to compare and analyze the results, dynamically adjust the detection threshold according to the benchmark parameters in the environmental benchmark model, and generate an adaptive abnormal signal threshold T 自适应 :

[0108] T 自适应 =δ·(W G ·G 基准 +WC ·C 基准 +W N ·N 基准 )+κ·H(E)·e -λt ;

[0109] Among them, W G , W C and W N are the weighting factors of geographical conditions, climate characteristics and noise benchmark, G 基准 , C 基准 and N 基准 denote geographic, climate and noise baseline parameters respectively, δ is the adjustment coefficient, H(E) denotes the entropy value of historical invasion events, κ is the adjustment factor, and λ is the time decay coefficient;

[0110] S33, comparing the signal data with the generated adaptive threshold T 自适应 Compare and filter out highly reliable abnormal signal data D 异常 ,This dataset reflects possible intrusion events after threshold adjustment;

[0111] S34, the screened high-confidence abnormal signal data D 异常 The data is stored in the monitoring system database and associated with historical anomaly data for long-term analysis and optimization of subsequent anomaly detection models, thereby optimizing the system's adaptability to complex environments and the accuracy of identifying intrusion behaviors.

[0112] In this implementation, S4 specifically includes:

[0113] S41. Deploy physical barrier devices in key areas of optical cables. When the physical barrier detects abnormal external force, the collected signal data is compared with the screened high-reliability abnormal signal data D 异常 Perform a comparison to confirm whether the physical interference meets the characteristics of the intrusion signal;

[0114] S42, after confirming that the abnormal signal meets the intrusion characteristics, the distributed optical fiber sensors in the adjacent areas are triggered to enter the high-frequency monitoring state, and the monitoring frequency is set to f 高 , and dynamically adjust based on the risk assessment factor R:

[0115]

[0116] Among them, f 0 is the basic monitoring frequency, α is the adjustment coefficient, R is the risk assessment factor, g i ,c i ,n i Respectively represent the geographical, climate and noise characteristic values ​​of the ith monitoring point;

[0117] S43, using the high-risk area parameters in the adaptive monitoring configuration database, compare the signal data of the densely monitored area with the adaptive abnormal signal threshold T 自适应 Compare and select intrusion feature data with higher credibility D 高可信 :

[0118]

[0119] Among them, γ is the adjustment factor, H(E) is the entropy value of historical intrusion events, P(I) is the predicted intrusion probability, μ is the adjustment coefficient, R 临界 is the critical risk assessment factor;

[0120] S44, the selected high-credibility intrusion feature data D 高可信 It is transmitted to the monitoring system database and combined with historical intrusion data to generate real-time warning information, which is sent to the remote monitoring center and maintenance personnel to form a local intensive monitoring area to optimize the capture capability of intrusion characteristics and support rapid response.

[0121] In this implementation manner, S5 specifically includes:

[0122] S51, after confirming the high-confidence intrusion feature data D 高可信 Finally, the distributed sensor network is used to accurately locate the geographical location of the intrusion signal, and the location coordinates (x, y, z) of the intrusion point are calculated and determined based on the GIS platform and the risk assessment factor R:

[0123]

[0124] Where n is the number of reference monitoring points used for positioning, R i is the risk assessment factor of the ith monitoring point, is the spatial coordinate of the i-th monitoring point (x i ,y i ,z i 0;

[0125] S52, compare the located intrusion location data (x, y, z) with the environmental benchmark model M 环境 Compare and combine the geographical condition benchmark and climate characteristic benchmark to calculate the path risk coefficient R 路径 :

[0126]

[0127] Where m is the number of monitoring areas involved in the path evaluation, W G and W C are weight factors for geographical conditions and climate characteristics, G 基准,j and C 基准,jdenote the geographic and climatic benchmark parameters of the jth region, respectively;

[0128] S53, based on the path risk coefficient R 路径 , generate an intrusion path prediction model, dynamically identify the optimal response path from the intrusion point to the key node, mark the areas with high path risk and generate path guidance, and provide real-time path guidance for the emergency response team;

[0129] S54. The path prediction model and intrusion location data are stored in the monitoring system database through the 4G / 5G network, the warning information and path guidance are transmitted to the maintenance personnel's mobile terminal through the GIS platform, and the relevant data are stored in the monitoring system database to support path optimization, intrusion feature analysis and dynamic adjustment of protection strategies.

[0130] In this implementation manner, S6 specifically includes:

[0131] S61, each detected intrusion event information, including the intrusion time t 入侵 , location coordinates (x, y, z), risk assessment factor R and high-confidence intrusion feature data D 高可信 , recorded in the monitoring system database, and associated with the geographic information of the GIS platform to form an event record set E 记录 For the management and analysis of historical data;

[0132] S62, event record set E 记录 Perform cluster analysis and calculate the feature similarity S between events 事件 , generate an intrusion event clustering model:

[0133]

[0134] Where n is the number of historical events, R i and D 高可信,i are the risk assessment factor and high-confidence intrusion feature data of the i-th historical event respectively;

[0135] S63. Based on the intrusion event clustering model, the frequency distribution of each cluster is calculated, and the historical event frequency is weighted using the time decay factor λ to obtain the weighted intrusion frequency F 加权 :

[0136]

[0137] Where m is the number of events in the clustering model, f j is the frequency of occurrence of the jth event, t 当前 is the current time, t j is the occurrence time of the jth event, λ is the time decay factor;

[0138] S64, according to the weighted intrusion frequency F 加权 Based on the results of event similarity analysis and combined with the geographic information provided by the GIS platform, the monitoring frequency, sensitivity and alarm threshold of each monitoring area are dynamically adjusted, and the updated monitoring parameters are stored in the monitoring system database to optimize the system's adaptability and response efficiency in different environments and achieve optimization of resource allocation.

[0139] In this embodiment, the following modules are included:

[0140] Distributed sensor module: Deploy sensor networks along optical cable lines to collect environmental data and build environmental benchmark models to adapt to different geographical and climatic conditions;

[0141] Environmental benchmark model module: pre-processes the collected environmental data, generates standardized data sets, builds benchmark models of geography, climate, and noise, supports subsequent risk assessment and anomaly detection, and integrates into the GIS platform;

[0142] Risk assessment module: Based on the environmental baseline model and historical intrusion data, the risk assessment factor is calculated, the monitoring area is divided into high-risk, second-highest-risk and low-risk areas, and the monitoring frequency and sensitivity are set respectively to optimize resource allocation;

[0143] Anomaly detection module: compares signal data with the environmental benchmark model and selects highly reliable abnormal signals through dynamic threshold adjustment;

[0144] Physical barrier trigger module: A physical barrier is set up in the key area of ​​the optical cable. When abnormal external force is detected, high-frequency monitoring of the area and adjacent areas is triggered to form a dense monitoring area;

[0145] Intrusion positioning module: uses distributed sensors and high-reliability abnormal signal data, combined with the GIS platform for positioning, calculates the location coordinates of the intrusion event, and transmits the location information to the monitoring system database;

[0146] Path prediction and response module: Generates an intrusion path prediction model based on path risk assessment, dynamically determines the optimal response path from the intrusion point to the key node, marks high-risk paths, and provides real-time path guidance and response suggestions for the emergency response team;

[0147] Historical data recording and adaptive adjustment module: records the data of each intrusion event, and dynamically adjusts it through cluster analysis and weighted frequency combined with GIS geographic information to optimize the monitoring frequency, sensitivity and alarm threshold of the monitoring area, and optimize the system's adaptive ability and resource allocation efficiency.

[0148] Embodiment 1:

[0149] In order to verify the feasibility of the present invention in implementation, the present invention is applied to the actual application of a communication line in a mountainous area. A section of an optical cable line of about 60 kilometers needs to be deployed with an anti-intrusion monitoring system to ensure the security and stability of the communication network. The line passes through multiple complex terrain areas, including valleys, rivers, forests, etc., and some areas are difficult to inspect and maintain due to remoteness or inconvenient transportation. The area is characterized by frequent wild animal activities and occasional risks of human construction. These factors increase the possibility of intrusion and damage to the optical cable. In the past inspection process, the line has suffered physical damage many times, which not only caused communication interruption, but also caused high maintenance costs. The communication company decided to introduce the optical cable routing anti-intrusion method and system of the present invention to improve the protection effect through an intelligent monitoring system.

[0150] A distributed sensor network is deployed at key nodes along the optical cable, especially in areas vulnerable to damage such as poles, valleys, and tunnels. Sensors are deployed to collect environmental data such as vibration, temperature, humidity, and background noise. Different sensors, such as infrared, microwave, and vibration sensors, are used to monitor environmental changes in real time. The sensor spacing is dynamically adjusted according to the complexity of the terrain, the frequency of environmental changes, and the risk level. For example, the density of sensor deployment in the pole area is high, with a spacing of about 30 meters, while the spacing in the low-risk area is about 100 meters. These data are transmitted back to the monitoring center in real time through the 4G / 5G network and integrated with the GIS management platform within the system to form an environmental benchmark model.

[0151] After a week of basic data collection, the system generated an environmental baseline model and conducted regional risk classification. According to the analysis, the 15th to 25th kilometers of the line are high-risk areas, mainly because of the complex terrain and proximity to residential areas, which may pose a risk of illegal construction. The system sets this area as a high-risk intensive monitoring area with a monitoring frequency of once every 10 seconds and high sensitivity to ensure that any abnormal fluctuations can be captured in a timely manner. Other areas are divided into second-highest risk and low risk, and the monitoring frequency is set to once per minute.

[0152] During actual operation, the system can not only automatically screen out abnormal signals with high credibility, but also effectively distinguish between environmental noise and intrusion signals. At 4 a.m. one day, the system detected abnormal signal fluctuations at the 19th kilometer position, which lasted for 5 seconds. The signal strength exceeded the set threshold. Through adaptive threshold adjustment and data comparison, the system preliminarily determined that the signal was an abnormal external force. After the physical barrier module was activated, the sensors from the 18th to the 20th kilometer were set to high-frequency monitoring mode, and the collection frequency was adjusted to once every 5 seconds, forming a local intensive monitoring area. The system found that the signal frequency and intensity changes in the area continued to increase, and finally determined it to be an intrusion.

[0153] After confirming the intrusion, the system accurately located the geographic coordinates of the intrusion point at 19.3 kilometers based on the geographic information of the GIS platform, and sent the positioning information, warning information and path guidance to the monitoring center and the maintenance personnel's mobile terminals through the 4G network. The maintenance personnel arrived at the scene within 30 minutes along the predetermined route based on the path guidance generated by the system. After arriving at the scene, the maintenance personnel found that someone was conducting illegal excavation. The system's warning information enabled them to stop the destructive behavior in time and avoid damage to the optical cable.

[0154] The system also recorded the time, location, signal strength and other information of the intrusion incident and stored it in the historical database for subsequent data analysis. In the following months, the system continuously updated and optimized monitoring parameters through cluster analysis of similar incidents and combined with geographic information from the GIS platform, increasing the monitoring frequency in high-risk areas and maintaining basic monitoring coverage in low-risk areas. According to data statistics from the first quarter, the false alarm rate of the optical cable line dropped from 5% of the traditional system to 1.5%, the response time was reduced from an average of 50 minutes to less than 30 minutes, and the maintenance cost was reduced by more than 20%.

[0155] Table 1 Optical cable routing anti-intrusion system data sheet

[0156]

[0157]

[0158] From the above table data, we can see that in the high-risk area of ​​15 to 25 kilometers, the system sets the monitoring frequency to once every 10 seconds to deal with the potential intrusion risk in this area. For the densely monitored area of ​​18 to 20 kilometers, the system automatically switches to a high-frequency monitoring mode of once every 5 seconds after detecting abnormal external forces, forming local intensive monitoring to ensure accurate capture of abnormal behavior. When an abnormal signal lasting for 5 seconds is detected at the 19th kilometer, the system can control the offset of the intrusion position within 10 meters and complete the emergency response within 30 minutes. Quarterly data statistics show that through the analysis of historical data and dynamic adjustment of parameters, the average response time of the system is maintained within 30 minutes, the false alarm rate is significantly reduced to 1.5%, and the maintenance cost is also reduced by more than 20%.

[0159] In summary, the optical cable routing anti-intrusion system of the present invention has demonstrated excellent performance in the protection application of optical cable lines in mountainous areas. Through distributed sensing, dynamic risk classification and adaptive monitoring strategies, it has not only successfully solved the monitoring blind spots and false alarm problems caused by complex terrain and environmental interference, but also achieved fast and accurate intrusion response, greatly improving the safety and maintenance efficiency of optical cable lines.

[0160] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An optical cable routing anti-intrusion method, characterized in that: The steps include: S1. Distributed sensor devices are deployed at key nodes of the optical cable line to collect initial environmental data around the optical cable, generate an environmental benchmark model including geographical conditions, climate characteristics and background noise, and integrate the benchmark model into the GIS management platform; S2. Based on the environmental benchmark model, the optical cable routing is risk graded and divided into high-risk and low-risk areas. The corresponding monitoring frequency and sensitivity are set for different risk areas. The high-risk area is configured as an intensive monitoring mode, and the low-risk area is configured as a periodic monitoring mode to achieve resource allocation optimization; S3. During real-time monitoring, the collected signal data is compared with the environmental benchmark model, abnormal signals are identified through the self-calibration anomaly detection algorithm, and the detection threshold is adaptively adjusted according to environmental noise, natural changes and daily vibration dynamic data to form screened high-confidence abnormal signal data; S4. Physical barriers are deployed in key areas of the optical cable line. When the physical barriers detect abnormal external forces, the sensors in the area and its adjacent areas are triggered to enter a high-frequency monitoring state. Based on the screened high-reliability abnormal signal data, a dense monitoring area is formed locally to optimize the ability to capture intrusion features. S5. After confirming the intrusion, the location of the intrusion event is accurately located through the GIS platform, and the geographic coordinates, event information and path guidance are transmitted to the monitoring center through the 4G / 5G network. At the same time, an early warning signal is sent to the mobile terminal of the maintenance personnel to support response and on-site processing; S6. The time, location, characteristic data and processing results of each intrusion event are recorded in the historical database, and the monitoring frequency, sensitivity and alarm threshold of each monitoring area are dynamically adjusted based on historical data analysis and geographic information of the GIS platform to optimize the system's adaptability and response efficiency in different environments.

2. The optical cable routing anti-intrusion method according to claim 1, characterized in that: The S1 specifically includes: S11. Distributed optical fiber sensors are deployed along the optical cable line. The sensors are deployed at a preset dynamic spacing. The spacing is adaptively adjusted according to the complexity of the terrain along the line, the frequency of environmental changes and the risk level, so as to form a sensor network covering the optical cable line, which is used to collect data on the vibration frequency, temperature, humidity and background noise intensity of the environment around the optical cable; S12. Perform multi-layer preprocessing on the collected environmental data, clean the data to remove outliers, eliminate short-term fluctuation interference through dynamic smoothing algorithm, and generate a standardized data set D 标准 , and is associated with the sensor location; S13. Based on the standardized dataset D 标准 , combined with the geographical characteristics and historical climate data along the optical cable, to generate the environmental benchmark model M 环境 The model includes geographical condition benchmark, climate characteristic benchmark and noise benchmark. The geographical condition benchmark is based on terrain undulation and human activity distribution information, the climate characteristic benchmark is based on historical temperature, humidity and seasonal variation, and the noise benchmark determines the environmental noise characteristics through hierarchical spectrum analysis. S14, the environmental benchmark model M 环境 It is integrated into the GIS management platform and stored in the monitoring system database. The benchmark model is updated at set time intervals through an adaptive update algorithm to adapt to dynamic changes in the environment and continuously optimize the accuracy of signal comparison and anomaly detection.

3. The optical cable routing anti-intrusion method according to claim 1, characterized in that: The S2 specifically includes: S21, based on the generated environmental benchmark model M 环境 , using the data of geographical condition benchmark, climate characteristic benchmark and noise benchmark, combined with historical intrusion event data, the risk assessment factor R is calculated and generated to divide the risk area along the optical cable: Where n is the number of monitoring points along the line, W G , W C and W N is the weight coefficient of geographical conditions, climate characteristics and noise benchmark, g i 、c i and n i represent the geographical condition value, climate characteristic value and noise reference value of the ith monitoring point respectively, and f G (g i ), f C (c i ) and f N (n i ) is the normalized function of each benchmark parameter, H(E) is the entropy value of the frequency of historical intrusion events, P(I) is the predicted intrusion probability function, and α, β and γ are adjustment coefficients; S22. According to the risk assessment factor R, the monitoring area along the optical cable is divided into high-risk area, second-high-risk area and low-risk area, with the high-risk area as the priority monitoring area, the second-high-risk area as the key monitoring area, and the low-risk area as the periodic monitoring area, and the corresponding monitoring levels are set; S23. Set up a high-frequency intensive monitoring mode in high-risk areas, and set the monitoring frequency to f 高 , the sensitivity is set to S 高 , the frequency and sensitivity are adaptively adjusted through real-time data feedback, and the monitoring intensity is automatically adjusted in response to fluctuations in the risk assessment factor R; S24: Set up adaptive monitoring mode in the second highest risk area, and set the monitoring frequency to f 次高 , the sensitivity is set to S 次高 , monitoring frequency and sensitivity are based on the environmental benchmark model M 环境 Dynamic adjustment of historical climate change data and current environmental fluctuations; S25. Set a low-frequency periodic monitoring mode in low-risk areas, and set the monitoring frequency to f 低 , the sensitivity is set to S 低 , low-risk areas maintain low-frequency monitoring mode and sensitivity within the set period to achieve basic monitoring coverage and resource optimization; S26. Establish a parameter database for monitoring configurations in different risk areas and compare the monitoring configuration parameters with the environmental benchmark model M 环境 Based on the real-time data and the changes in the risk assessment factor R, the monitoring parameters are updated regularly and adaptively adjusted; S27. Set time intervals and environmental change trigger mechanisms so that the adaptive monitoring configuration database can periodically recalculate the risk assessment factor R, and update the monitoring frequency and sensitivity of high-risk, second-highest-risk, and low-risk areas based on dynamic changes in the environmental benchmark model, so that the monitoring system matches environmental changes and achieves effectiveness in resource allocation and anomaly detection.

4. The optical cable routing anti-intrusion method according to claim 1, characterized in that: The S3 specifically includes: S31. During real-time monitoring, the collected signal data is compared with the standardized data set D 标准 and environmental benchmark model M 环境 Comparison is performed to detect abnormal fluctuations in the signal and to identify the characteristics of the abnormal signal based on the geographic condition benchmark, climate characteristic benchmark and noise benchmark; S32, using the self-calibration anomaly detection algorithm to compare and analyze the results, dynamically adjust the detection threshold according to the benchmark parameters in the environmental benchmark model, and generate an adaptive abnormal signal threshold T 自适应 : T 自适应 =δ·(W G ·G 基准 +W C ·C 基准 +W N ·N 基准 )+κ·H(E)·e -λt ; Among them, W G , W C and W N are the weighting factors of geographical conditions, climate characteristics and noise benchmark, G 基准 , C 基准 and N 基准 denote geographic, climate and noise baseline parameters respectively, δ is the adjustment coefficient, H(E) denotes the entropy value of historical invasion events, κ is the adjustment factor, and λ is the time decay coefficient; S33, comparing the signal data with the generated adaptive threshold T 自适应 Compare and filter out highly reliable abnormal signal data D 异常 ,This dataset reflects possible intrusion events after threshold adjustment; S34, the screened high-confidence abnormal signal data D 异常 The data is stored in the monitoring system database and associated with historical anomaly data for long-term analysis and optimization of subsequent anomaly detection models, thereby optimizing the system's adaptability to complex environments and the accuracy of identifying intrusion behaviors.

5. The optical cable routing anti-intrusion method according to claim 1, characterized in that: The S4 specifically includes: S41. Deploy physical barrier devices in key areas of optical cables. When the physical barrier detects abnormal external force, the collected signal data is compared with the screened high-reliability abnormal signal data D 异常 Perform a comparison to confirm whether the physical interference meets the characteristics of the intrusion signal; S42, after confirming that the abnormal signal meets the intrusion characteristics, the distributed optical fiber sensors in the adjacent areas are triggered to enter the high-frequency monitoring state, and the monitoring frequency is set to f 高 , and dynamically adjust based on the risk assessment factor R: Among them, f0 is the basic monitoring frequency, α is the adjustment coefficient, R is the risk assessment factor, g i ,c i ,n i represent the geographical, climatic and noise characteristic values ​​of the ith monitoring point respectively; S43, using the high-risk area parameters in the adaptive monitoring configuration database, compare the signal data of the densely monitored area with the adaptive abnormal signal threshold T 自适应 Compare and select intrusion feature data with higher credibility D 高可信 : Among them, γ is the adjustment factor, H(E) is the entropy value of historical intrusion events, P(I) is the predicted intrusion probability, μ is the adjustment coefficient, R 临界 is the critical risk assessment factor; S44, the selected high-credibility intrusion feature data D 高可信 It is transmitted to the monitoring system database and combined with historical intrusion data to generate real-time warning information, which is sent to the remote monitoring center and maintenance personnel to form a local intensive monitoring area to optimize the capture capability of intrusion characteristics and support rapid response.

6. The optical cable routing anti-intrusion method according to claim 1, characterized in that: The S5 specifically includes: S51, after confirming the high-confidence intrusion feature data D 高可信 Finally, the distributed sensor network is used to accurately locate the geographical location of the intrusion signal, and the location coordinates (x, y, z) of the intrusion point are calculated and determined based on the GIS platform and the risk assessment factor R: Where n is the number of reference monitoring points used for positioning, R i is the risk assessment factor of the ith monitoring point, is the spatial coordinate of the i-th monitoring point (x i ,y i ,z i ); S52, the located intrusion location data ( x,y,z ) and the environmental benchmark model M 环境 Compare and combine the geographical condition benchmark and climate characteristic benchmark to calculate the path risk coefficient R 路径 : Where m is the number of monitoring areas involved in the path evaluation, W G and W C are weight factors for geographical conditions and climate characteristics, G 基准,j and C 基准,j denote the geographic and climatic benchmark parameters of the jth region, respectively; S53, based on the path risk coefficient R 路径 , generate an intrusion path prediction model, dynamically identify the optimal response path from the intrusion point to the key node, mark the areas with high path risk and generate path guidance, and provide real-time path guidance for the emergency response team; S54. The path prediction model and intrusion location data are stored in the monitoring system database through the 4G / 5G network, the warning information and path guidance are transmitted to the maintenance personnel's mobile terminal through the GIS platform, and the relevant data are stored in the monitoring system database to support path optimization, intrusion feature analysis and dynamic adjustment of protection strategies.

7. The optical cable routing anti-intrusion method according to claim 1, characterized in that: The S6 specifically includes: S61, each detected intrusion event information, including the intrusion time t 入侵 , location coordinates (x, y, z), risk assessment factor R and high-confidence intrusion feature data D 高可信 , recorded in the monitoring system database, and associated with the geographic information of the GIS platform to form an event record set E 记录 For the management and analysis of historical data; S62, event record set E 记录 Perform cluster analysis and calculate the feature similarity S between events 事件 , generate an intrusion event clustering model: Where n is the number of historical events, R i and D 高可信,i are the risk assessment factor and high-confidence intrusion feature data of the i-th historical event respectively; S63. Based on the intrusion event clustering model, the frequency distribution of each cluster is calculated, and the historical event frequency is weighted using the time decay factor λ to obtain the weighted intrusion frequency F 加权 : Where m is the number of events in the clustering model, f j is the frequency of occurrence of the jth event, t 当前 is the current time, t j is the occurrence time of the jth event, λ is the time decay factor; S64, according to the weighted intrusion frequency F 加权 Based on the results of event similarity analysis and combined with the geographic information provided by the GIS platform, the monitoring frequency, sensitivity and alarm threshold of each monitoring area are dynamically adjusted, and the updated monitoring parameters are stored in the monitoring system database to optimize the system's adaptability and response efficiency in different environments and achieve optimization of resource allocation.

8. An optical cable routing anti-intrusion system, characterized in that: Includes the following modules: Distributed sensor module: Deploy sensor networks along optical cable lines to collect environmental data and build environmental benchmark models to adapt to different geographical and climatic conditions; Environmental benchmark model module: pre-processes the collected environmental data, generates standardized data sets, builds benchmark models of geography, climate, and noise, supports subsequent risk assessment and anomaly detection, and integrates into the GIS platform; Risk assessment module: Based on the environmental baseline model and historical intrusion data, the risk assessment factor is calculated, the monitoring area is divided into high-risk, second-highest-risk and low-risk areas, and the monitoring frequency and sensitivity are set respectively to optimize resource allocation; Anomaly detection module: compares signal data with the environmental benchmark model and selects highly reliable anomaly signals through dynamic threshold adjustment; Physical barrier trigger module: A physical barrier is set up in the key area of ​​the optical cable. When abnormal external force is detected, high-frequency monitoring of the area and adjacent areas is triggered to form a dense monitoring area; Intrusion positioning module: uses distributed sensors and high-reliability abnormal signal data, combined with the GIS platform for positioning, calculates the location coordinates of the intrusion event, and transmits the location information to the monitoring system database; Path prediction and response module: Generates an intrusion path prediction model based on path risk assessment, dynamically determines the optimal response path from the intrusion point to the key node, marks high-risk paths, and provides real-time path guidance and response suggestions for the emergency response team; Historical data recording and adaptive adjustment module: records the data of each intrusion event, and dynamically adjusts it through cluster analysis and weighted frequency combined with GIS geographic information to optimize the monitoring frequency, sensitivity and alarm threshold of the monitoring area, and optimize the system's adaptive ability and resource allocation efficiency.

Images