A multi-node ad hoc network cooperative monitoring method for regional nuclear radiation

Abstract

The application provides a multi-node self-organizing network cooperative monitoring method for regional nuclear radiation, relates to the technical field of nuclear radiation monitoring, uses mobile monitoring modules with edge computing capability, and uses a plurality of the mobile monitoring modules as monitoring nodes to form a wireless self-organizing network in a target region; each monitoring node synchronously performs real-time measurement of nuclear radiation in the target region to obtain local measurement data. The application preinstalls a radiation diffusion prediction model in each mobile monitoring node and gives the edge computing capability; each node independently generates a prediction trend based on local and adjacent node data, and cooperatively identifies a key region in a distributed manner by exchanging data; even if individual nodes fail or part of the communication links are interrupted, the remaining nodes can still continue to perform the prediction, cooperation and adjustment tasks based on local information, the whole system shows high self-organization and robustness, and the fatal defect that a single point failure of a center node leads to paralysis of the whole network can be overcome.

Description

Technical Field

[0001] This invention relates to the field of nuclear radiation monitoring technology, and in particular to a multi-node self-organizing network collaborative monitoring method for regional nuclear radiation. Background Technology

[0002] Nuclear radiation monitoring is a core component of nuclear safety, emergency response, and environmental assessment. Regional nuclear radiation monitoring can provide real-time insights into the spatial distribution and temporal evolution of radiation fields over large geographical areas, which is crucial for monitoring the vicinity of nuclear facilities, assessing the consequences of accidents, and making emergency decisions. Traditional monitoring methods primarily rely on fixed monitoring stations and manual inspections, which suffer from inherent limitations such as inflexible deployment, numerous coverage blind spots, and slow response times, making it difficult to meet the modern demands for rapid, accurate, and adaptive tracking and monitoring of sudden and mobile radiation contamination. With the development of wireless sensor networks and the Internet of Things (IoT) technologies, regional monitoring using multi-node networking has become an important research direction.

[0003] Currently, common network monitoring systems aggregate collected radiation data from monitoring nodes via wireless networks to a central server, where the server performs unified data processing, analysis, and mapping. This approach involves highly centralized decision-making and control; all data processing and command generation depend on the central node. If the central node fails or the communication link is interrupted, the entire system will be paralyzed. Furthermore, nodes are typically statically deployed and cannot autonomously adjust their spatial layout and monitoring strategies based on real-time dynamic changes in the radiation field, resulting in extremely limited ability to track rapidly changing radiation clouds. Summary of the Invention

[0004] In view of this, the present invention proposes a multi-node self-organizing network collaborative monitoring method for regional nuclear radiation. By pre-setting a radiation diffusion prediction model in each mobile monitoring node and endowing it with edge computing capabilities, each node independently generates prediction trends based on local and neighboring node data, and jointly identifies key areas in a distributed collaborative manner by exchanging data. The architecture ensures that even if individual nodes fail or some communication links are interrupted, the remaining nodes can still continue to perform prediction, collaboration and adjustment tasks based on local information. The entire system exhibits high self-organization and robustness, and can overcome the fatal defect that a single point of failure of the central node will lead to the paralysis of the entire network.

[0005] The technical solution of this invention is implemented as follows: On one hand, this invention provides a multi-node ad hoc network collaborative monitoring method for regional nuclear radiation, characterized by using mobile monitoring modules with edge computing capabilities, and using multiple mobile monitoring modules as monitoring nodes to form a wireless ad hoc network within a target area. The method includes: Each monitoring node synchronously performs real-time nuclear radiation measurements in the target area to obtain its own local measurement data, which includes nuclear radiation data and environmental parameter data. The nuclear radiation data includes at least the gamma radiation dose rate. Each monitoring node runs a pre-set radiation diffusion prediction model based on local measurement data to generate local short-term radiation trend prediction data. The radiation diffusion prediction model takes local measurement data and data from adjacent monitoring nodes as input and outputs short-term radiation trend prediction data corresponding to the monitoring node. Each monitoring node exchanges short-term radiation trend prediction data through a wireless self-organizing network, and based on the exchanged data, identifies key areas that need to be monitored in a distributed and collaborative manner. Each monitoring node adjusts its own operating parameters based on the information of the key area to obtain the adjusted target distribution network. The adjustment of its own operating parameters includes at least some monitoring nodes moving to the key area to enhance the monitoring coverage of the key area. Data is collected from all monitoring nodes in the target distribution network, and spatiotemporal fusion processing is performed to generate a dynamic distribution map of the regional radiation field.

[0006] Based on the above technical solutions, preferably, each monitoring node synchronously performs real-time nuclear radiation measurements in the target area to obtain its own local measurement data, specifically including: Each monitoring node receives a unified time signal and calibrates its local clock based on the time signal to achieve time synchronization of all monitoring nodes in the network. During the synchronized time period, the monitoring node synchronously collects nuclear radiation data and environmental parameter data, wherein the nuclear radiation data includes at least the gamma radiation dose rate. The monitoring node performs local preprocessing on the collected raw data. The preprocessing includes at least data validity verification, background noise reduction, and unification of physical units. The pre-processed raw data is appended with corresponding timestamps and monitoring node location information, and then encapsulated into structured local measurement data.

[0007] Based on the above technical solutions, preferably, each monitoring node runs a pre-set radiation diffusion prediction model based on local measurement data to generate local short-term radiation trend prediction data, specifically including: The monitoring node acquires local measurement data and obtains its corresponding local measurement data from neighboring monitoring nodes through a wireless ad hoc network. The monitoring node takes local measurement data, data from adjacent monitoring nodes, and locally collected environmental parameter data as a set of input vectors and inputs them into the preset radiation diffusion prediction model. The radiation diffusion prediction model performs calculations at the monitoring nodes and outputs the predicted radiation intensity value after a predetermined time in the future, as well as the spatial gradient trend of the predicted radiation intensity value at the monitoring node location. The monitoring nodes encapsulate the predicted radiation intensity values ​​and spatial gradient change trends to generate structured short-term radiation trend prediction data.

[0008] Based on the above technical solutions, preferably, each monitoring node exchanges short-term radiation trend prediction data through a wireless ad hoc network, and identifies key areas requiring focused monitoring in a distributed and collaborative manner based on the exchanged data, specifically including: Each monitoring node broadcasts its own generated short-term radiation trend prediction data through a wireless ad hoc network and receives short-term radiation trend prediction data from at least one other monitoring node. Each monitoring node constructs a comprehensive trend distribution map covering the prediction area of ​​all nodes within its corresponding communication range, based on the spatial gradient change trend in its own and received short-term radiation trend prediction data. Based on the comprehensive trend distribution map, each monitoring node evaluates and votes on areas with significant gradient changes and predicted radiation intensity exceeding the threshold, according to preset judgment rules, as candidate areas. Each monitoring node exchanges voting information and uses a distributed consensus algorithm to confirm and merge candidate key areas, outputting key areas that need to be monitored in a network-wide consensus. The key area is a set of coordinates.

[0009] Based on the above technical solutions, preferably, each monitoring node adjusts its own operating parameters according to the information of the key area to obtain an adjusted target distribution network, specifically including: Each monitoring node acquires the coordinates of the key area and simultaneously acquires the real-time status information of other monitoring nodes within the ad hoc network. The status information includes at least the node location and node type. Each monitoring node generates a parameter adjustment instruction for itself based on a preset collaborative scheduling strategy, its own status, and the acquired global network information. The parameter adjustment instruction includes at least the target movement vector and the updated sampling frequency. The monitoring node adjusts according to the parameter adjustment command. After the adjustment is completed, the adjusted target distribution network is obtained, and its latest status information is updated and broadcast.

[0010] Based on the above technical solutions, a preferred approach involves collecting data from all monitoring nodes in the target distribution network, performing spatiotemporal fusion processing, and generating a dynamic distribution map of the regional radiation field. Specifically, this includes: Collect nuclear radiation data from all monitoring nodes after adjusting their operating parameters. The nuclear radiation data also includes location information and timestamps. All nuclear radiation data are subjected to spatiotemporal alignment preprocessing, which includes unifying the data to the same reference time window based on timestamps and mapping the data to a unified spatial coordinate system based on location information. Based on the spatiotemporally aligned preprocessed nuclear radiation data, a spatial interpolation algorithm is used to generate a spatial radiation intensity distribution surface covering the entire target area. By performing a time-series correlation analysis between the distribution surface generated in the current time window and the historical distribution surfaces of multiple consecutive time windows, a dynamic distribution map describing the changes in radiation field intensity and diffusion direction is generated.

[0011] Based on the above technical solution, preferably, before each monitoring node synchronously performs real-time nuclear radiation measurement in the target area and obtains its own local measurement data, the following further step is taken: Multiple monitoring nodes are deployed in the target area; Each monitoring node identifies other monitoring nodes within its communication range by listening to and responding to broadcast beacons, and negotiates to establish an initial wireless multi-hop communication link, forming a wireless ad hoc network. In the formed wireless ad hoc network, each monitoring node exchanges initial location information; The monitoring nodes enter a ready state and wait for a unified synchronization measurement command, which enables each monitoring node to synchronously perform real-time nuclear radiation measurements in the target area.

[0012] On the other hand, the present invention provides a multi-node self-organizing network collaborative monitoring system for regional nuclear radiation, used to implement the above-mentioned multi-node self-organizing network collaborative monitoring method for regional nuclear radiation. The system includes multiple monitoring nodes and a data processing center, wherein... The monitoring nodes are mobile and include at least: Nuclear radiation sensor, used for nuclear radiation measurement; Edge computing units are pre-configured with radiation diffusion prediction models; and Wireless communication module, used for network communication between nodes; The multiple monitoring nodes form a wireless ad hoc network within the target area through their respective wireless communication modules; The edge computing unit is configured to perform the following operations: Control the nuclear radiation sensor to perform synchronous measurements and generate local measurement data; Run the radiation diffusion prediction model to generate local short-term radiation trend prediction data based on local measurement data and data obtained from neighboring nodes; The radiation short-term trend prediction data is exchanged through a wireless ad hoc network, and key areas are identified in a distributed and collaborative manner. Based on the identified key areas, generate and execute instructions for adjusting the working parameters of the monitoring nodes themselves. The data processing center is configured to perform the following operations: It aggregates monitoring data collected by all monitoring nodes in the wireless ad hoc network; The aggregated data undergoes spatiotemporal fusion processing to generate a dynamic distribution map of the regional radiation field.

[0013] On the other hand, the present invention provides an electronic device including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the above-described method.

[0014] On the other hand, the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0015] The multi-node self-organizing network collaborative monitoring method for regional nuclear radiation of the present invention has the following advantages over the prior art: 1. By pre-setting a radiation diffusion prediction model in each mobile monitoring node and giving it edge computing capabilities, each node independently generates prediction trends based on local and neighboring node data, and jointly identifies key areas in a distributed collaborative manner by exchanging data. The architecture ensures that even if individual nodes fail or some communication links are interrupted, the remaining nodes can still continue to perform prediction, collaboration and adjustment tasks based on local information. The whole system exhibits high self-organization and robustness, and can overcome the fatal defect that a single point failure of the central node will lead to the paralysis of the entire network. 2. By enabling monitoring nodes to autonomously make decisions and adjust their own operating parameters based on the coordinates of key areas identified collaboratively, at least some nodes are driven to actively move towards the predicted high-concentration paths or key areas. This transforms the monitoring network from a static and passive data acquisition network into a dynamic and proactive situational awareness and tracking network. The network topology and sensor resources can be adaptively reconstructed according to the real-time evolution of the radiation field, enhancing the coverage density and sampling frequency of high-risk areas. This enables precise tracking of moving and spreading radiation clouds, significantly improving the timeliness, flexibility, and overall efficiency of monitoring. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram illustrating the steps of the multi-node self-organizing network collaborative monitoring method for regional nuclear radiation according to the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0019] like Figure 1 As shown, the core of the multi-node self-organizing network collaborative monitoring method for regional nuclear radiation of the present invention lies in utilizing mobile monitoring modules with edge computing capabilities, hereinafter collectively referred to as monitoring nodes. Multiple such monitoring nodes autonomously form a wireless self-organizing network within the target area, and realize a complete monitoring process from prediction and decision-making to active tracking based on distributed intelligent collaboration. Initially, system initialization and deployment are required to lay the foundation for subsequent collaborative monitoring.

[0020] During initialization and deployment, multiple monitoring nodes need to be deployed in the target area.

[0021] In some embodiments, the monitoring node integrates a nuclear radiation sensor, a microprocessor unit, a wireless communication module, and a mobile platform. Specifically, the nuclear radiation sensor may be a gamma dose rate meter, the microprocessor unit serves as an edge computing unit, and the mobile platform can be a wheeled or tracked ground mobile platform, or it may be a flying device such as a drone. The deployment method can be flexibly selected according to the scenario. For example, in emergency response, it can be quickly deployed via drone airdrop or vehicle-borne deployment; in regular monitoring areas, it can be pre-positioned manually or on fixed bases. The target area refers to the specific geographical area that needs to be monitored, such as the area surrounding a nuclear facility, a suspected contaminated area, or a border crossing.

[0022] After the initial deployment of monitoring nodes, each node identifies other monitoring nodes within its communication range by listening to and responding to broadcast beacons, and negotiates to establish an initial multi-hop wireless communication link, forming a wireless ad hoc network. Specifically, each monitoring node periodically sends a beacon frame containing its own ID. When a node receives a beacon from an unknown neighbor, the two parties exchange link quality information and negotiate to establish a reliable communication path according to a preset routing protocol. This process occurs in parallel among all nodes, ultimately forming one or more interconnected subnets, ensuring that there are at least one relayable path between any two nodes in the network. For example, node A may establish an indirect connection with node C through node B as a relay. This self-organizing characteristic allows the network to expand rapidly and adaptively without relying on pre-set infrastructure.

[0023] After the wireless ad hoc network is formed, each monitoring node exchanges initial location information. Location information can be obtained through the node's built-in Global Navigation Satellite System (GNSS) module, or estimated in indoor / underground environments without GNSS signals using indoor positioning technologies such as ultra-wideband (UWB) or Bluetooth beacons, or through network topology-based positioning algorithms. Each node encapsulates its coordinates into data packets and broadcasts them through the established ad hoc network links. Thus, each node not only knows its own location but also gradually constructs an initial network topology map containing the location information of most nodes in the network. This location information serves as the basic spatial reference for subsequent spatial prediction, collaborative decision-making, and mobility scheduling.

[0024] After establishing the spatial reference, the monitoring nodes enter a ready state and await a unified synchronization measurement command. This command enables all monitoring nodes to synchronously conduct real-time nuclear radiation measurements within the target area. The synchronization measurement command can be issued by a designated node in the network or injected by an external command center via a long-wave link or satellite communication. The command may include an absolute timestamp as the start time of the network-wide synchronization measurement. Upon receiving the command, each node calibrates its local clock using a high-precision clock synchronization protocol to ensure a high degree of uniformity in the network's time reference. Once the time specified in the command is reached, all nodes simultaneously activate their nuclear radiation sensors and begin synchronous data acquisition. This synchronization mechanism is crucial for subsequent spatiotemporal correlation analysis of the data, eliminating data fusion errors caused by asynchronous sampling times and providing temporal consistency for generating accurate dynamic distribution maps.

[0025] After initialization is completed and the system is ready, the core collaborative monitoring process of the present invention is started, and each monitoring node synchronously performs real-time nuclear radiation measurement in the target area to obtain its own local measurement data.

[0026] During the acquisition of local measurement data, each monitoring node receives a unified time synchronization signal and calibrates its local clock based on this signal, achieving time synchronization across all monitoring nodes in the network. The time synchronization signal can utilize precise clock signals provided by global navigation satellite systems such as GPS and BeiDou. In environments where satellite signals cannot be directly received, time can be transmitted and distributed within the ad hoc network via a few synchronized gateway nodes using a precise time synchronization protocol. For example, a node deployed in an open area, upon receiving a GPS second pulse signal, not only calibrates its own clock but also acts as a time source, transmitting its high-precision time information to nodes deeper in the network that cannot directly receive satellite signals through a multi-hop approach. This hybrid time synchronization mechanism ensures that regardless of a node's location, its internal clock remains highly consistent with a unified time reference, providing a common time scale for synchronized actions across the entire network.

[0027] Based on time synchronization, within a unified planned time period, all monitoring nodes synchronously activate their sensors to collect nuclear radiation data and related environmental parameter data. The nuclear radiation data includes at least the gamma radiation dose rate. The core of nuclear radiation data acquisition is the gamma radiation dose rate, which is the most direct physical quantity characterizing the level of environmental radiation. It can be measured by scintillator detectors or semiconductor detectors built into the monitoring nodes. Simultaneously, to support subsequent radiation diffusion prediction models, the monitoring nodes also need to synchronously collect key environmental parameters, including atmospheric temperature, humidity, air pressure, wind speed, and wind direction. These environmental parameters directly affect the transport and diffusion of radioactive materials in the atmosphere. For example, nodes simultaneously sample at the exact second, obtaining the dose rate and wind direction at that moment. This synchronous acquisition mechanism ensures the comparability of data from different spatial locations in the temporal dimension.

[0028] Before transmission, the collected raw data needs to be preprocessed locally by the edge computing unit of the monitoring node to improve data quality and usability. This preprocessing includes at least data validity verification, background noise reduction, and physical unit standardization. Data validity verification, based on the sensor range and physical possibilities, removes obviously erroneous outliers. Background noise reduction, based on detector characteristics and laboratory calibration results, subtracts the instrument's background count from the raw reading to obtain the net radiation signal. Physical unit standardization converts the raw AD count or voltage signal into standard physical units using a preset calibration coefficient. For example, the microprocessor built into the monitoring node runs a lightweight filtering algorithm to perform median filtering on the dose rate of several consecutive sampling points to suppress transient spike interference. The filtered value is then multiplied by a calibration factor to obtain the standard value. This preprocessing process is performed at the source of data generation, effectively reducing the network transmission burden and providing clean input data for subsequent model calculations.

[0029] Finally, the preprocessed raw data is appended with corresponding timestamps and monitoring node location information, and encapsulated into structured local measurement data. The timestamps are taken from a synchronized high-precision local clock, and the location information comes from the real-time output of the GNSS module or network positioning algorithm.

[0030] After obtaining structured local measurement data, each monitoring node needs to run a pre-set radiation diffusion prediction model based on the local measurement data to generate local short-term radiation trend prediction data. The radiation diffusion prediction model takes local measurement data, environmental parameters and data from adjacent monitoring nodes as inputs and outputs short-term radiation trend prediction data corresponding to the monitoring node.

[0031] In some specific embodiments, the monitoring node first needs to acquire local measurement data and then obtain its corresponding local measurement data from neighboring monitoring nodes via a wireless ad hoc network. In addition to using its own generated local measurement data packets, the node must also obtain its corresponding local measurement data from one or more neighboring monitoring nodes within its communication range through the established wireless ad hoc network. This process can be achieved through periodic data broadcasting or an on-demand request-response mechanism. For example, node A listens for broadcast data packets from neighboring nodes B and C. These data packets contain the dose rate, location, and environmental parameters measured by B and C within the same time window. Node A correlates and caches this external data with its own data to form a local dataset, providing spatial context information for prediction. This data exchange is the foundation of distributed collaborative sensing, enabling a single node to transcend the physical limitations of its own sensors and perceive the preliminary outline of the radiation field over a larger area.

[0032] The monitoring node integrates and formats local data, received neighbor data, and locally collected environmental parameter data to construct a set of regular input vectors for calculation by a pre-defined radiation diffusion prediction model. Constructing the input vector requires normalizing heterogeneous data and organizing it into the format specified by the model. For example, for a prediction model based on a lightweight neural network, its input vector can be sequentially concatenated with the following features: the node's own gamma dose rate, temperature, humidity, wind speed, and wind direction; the gamma dose rates from the three nearest neighbor nodes and their azimuth and distance relative to the current node; and temporal features. These raw data are standardized to eliminate the influence of dimensions and improve model convergence and prediction stability. By constructing this comprehensive input vector that integrates local spatiotemporal and environmental information, the model can perform inference in a richer feature space.

[0033] The radiation diffusion prediction model performs calculations at the monitoring node, outputting the predicted radiation intensity value after a predetermined time interval, as well as the spatial gradient trend of the predicted radiation intensity value at the monitoring node location. The radiation diffusion prediction model can be a compressed machine learning model, such as a gradient boosting tree or a small convolutional neural network, or a parameterized simplified physical diffusion model, such as a fast solver for a Gaussian plume model at the edge. Since this is not related to the core innovation of this invention, it will not be described in detail here. The model's output includes two parts: first, the predicted radiation intensity value after a predetermined time interval, for example, an estimate of the gamma dose rate at the node location 5 or 10 minutes later; second, the spatial gradient trend of the predicted radiation intensity value at the current node location, which can be represented by a vector indicating the rate of change of the dose rate in the east-west and north-south directions. The vector and direction intuitively indicate the direction of the fastest increase in radiation concentration, i.e., the possible approach path of the radiation cloud. For example, if the model output predicts that the dose rate will rise to 150 nSv / h in 5 minutes, and the gradient vector points northeast, this strongly suggests that the radiation threat is approaching from the northeast.

[0034] The monitoring nodes encapsulate the predicted radiation intensity values ​​and spatial gradient change trends to generate structured short-term radiation trend prediction data.

[0035] After obtaining their respective short-term radiation trend prediction data, the present invention can enter the distributed group decision-making stage, that is, each monitoring node exchanges these prediction data through a wireless ad hoc network, and based on the prediction information of all nodes, identifies the key areas of the entire network that need to be monitored in a fully distributed and collaborative manner.

[0036] In some embodiments, each monitoring node first broadcasts its own generated short-term radiation trend prediction data via a wireless ad hoc network and receives short-term radiation trend prediction data from at least one other monitoring node. Simultaneously, each node continuously listens to the channel, receiving similar prediction data packets from at least one, typically multiple, other neighboring nodes within its communication range. For example, node A broadcasts its prediction packet, and simultaneously receives prediction packets from nodes B and C. This combination of active broadcasting and passive reception ensures that prediction information propagates rapidly within the local network, allowing each node to acquire a dataset reflecting multi-view prediction results for the local area. This mechanism inherently provides redundancy; even if a single data packet is lost, information can still be obtained through other paths or from other nodes.

[0037] Each monitoring node constructs a comprehensive trend distribution map covering the prediction areas of all nodes within its corresponding communication range, based on the spatial gradient change trends in its own data and the received short-term radiation trend prediction data. This comprehensive trend distribution map is essentially a digital representation of a two-dimensional scalar field and a two-dimensional vector field, and nodes can use algorithms such as inverse distance weighted interpolation. For a geographic coordinate point P(x,y) to be interpolated, its comprehensive gradient intensity value G... c (P) can be calculated using the following formula:

[0038]

[0039]

[0040] Where n is the total number of valid prediction data packets obtained by the node. It is the spatial gradient vector in the i-th data packet. Let P be the magnitude of the gradient vector, representing the drasticness of the gradient change. i Let d(P,P) be the location coordinates of the node that emits the i-th prediction data. i Let P be the distance from point P to node P. i The Euclidean distance, k is the attenuation coefficient used to control the degree of influence of distance on weight, w i The weights are calculated based on distance. The closer the node is, the greater the weight of the information it provides. Through this formula, the node can integrate discrete gradient observations into a continuous spatial distribution perception, thereby providing a quantitative basis for subsequent local judgments.

[0041] Based on the comprehensive trend distribution map, each monitoring node evaluates and votes for areas with significant gradient changes and predicted radiation intensity exceeding a threshold, according to preset judgment rules, as candidate areas. In a specific embodiment, the judgment rules are based on quantifiable technical indicators, transforming complex situational judgments into explicit mathematical conditions. Each node applies judgment logic to each analysis unit of the trend map:

[0042] If G is satisfied: c (P)>G th And D p (P)>D th Then, position P is marked as a potential key point, where G th D is the preset threshold for the significance of gradient change. p (P) represents the predicted radiation intensity value at this location obtained through interpolation, and D... th This is the preset threshold for predicted radiation intensity.

[0043] The node will set the geometric center coordinates of all marked connected regions {C1,C2,...,C...} mThe set of candidate regions supported by this node's vote is output.

[0044] Finally, in order to form a unified action goal across the entire network, each monitoring node needs to exchange voting information and use a distributed consensus algorithm to confirm and merge candidate key areas, and output the key areas that need to be monitored in a key manner that are agreed upon by the entire network. The key area is a set of coordinates.

[0045] Specifically, each node collects votes from the entire network within a certain period. For all collected candidate coordinates, a density-based clustering algorithm is used for aggregation. For example, the DBSCAN algorithm groups points that meet specific conditions into the same cluster. For each point in a cluster, it must contain at least MinPts other points within its neighborhood of radius ε. For point C... i Its neighborhood N ε (C i ) is defined as:

[0046] N ε (C i )={C j ∈AllCandidates|distance(C i C j )≤ε} If |N ε (C i If |≥MinPts, then C i It was regarded as the core point and began to form a cluster.

[0047] Each cluster represents a potential threat area that is independently discovered and collectively monitored by multiple nodes. The center coordinates of all points within each cluster are calculated and used as the coordinates K of the consensus critical region. l :

[0048]

[0049] Through multiple rounds of communication or verification, the network can reach a consensus on the final set of critical areas {K1, K2, ...} without a central coordinator. This coordinate set serves as the global action guide for all subsequent resource scheduling and movement adjustments by all nodes, ensuring the coordination and efficiency of group behavior.

[0050] After determining the coordinate set of key areas through distributed consensus, each monitoring node autonomously adjusts its operating parameters based on the information of that key area. The core of this process is to drive at least some mobile nodes to converge on the key area, thereby dynamically enhancing the network's monitoring coverage of key areas. This step represents a fundamental shift in the monitoring strategy from static preset to dynamic response, enabling the network to proactively track and contain radiation anomalies.

[0051] In some specific embodiments, each monitoring node acquires the coordinates of a key area and simultaneously obtains the real-time status information of other monitoring nodes within the ad hoc network. This status information includes at least the node's location and type. The coordinates of the key area can be treated as a set of coordinates, which can be broadcast and distributed within the network in a list format. Meanwhile, to make reasonable collaborative scheduling decisions, the monitoring nodes also need to acquire the status information of other nodes within the ad hoc network in real time. The status information includes at least: the node's real-time location, which is continuously updated via GNSS or positioning algorithms; and the node type, such as a high-speed UAV node, a high-precision fixed node, or a high-endurance vehicle-mounted node, etc. Different types determine their mobility, sensor accuracy, and energy characteristics. This information can be obtained through periodic heartbeat data packets or a dedicated neighbor state exchange protocol. For example, while receiving the coordinates of the key area, node A also learns from neighbor broadcasts that node B is 500 meters north of it and is patrolling; node C is 300 meters east of it and is stationary. Node A integrates this global information with its own status to form a localized network resource and situational view, serving as the basis for its decision-making.

[0052] Then, each monitoring node generates parameter adjustment instructions based on a preset collaborative scheduling strategy, its own status, and the acquired global network information. These instructions include at least a target movement vector and an updated sampling frequency. The collaborative scheduling strategy is a set of decision-making logic rules embedded in each node. Its core objective is to optimize the network's monitoring effect on key areas while considering energy balance and avoiding node action conflicts. The strategy's decision output includes at least two core instructions: one is the target movement vector, which indicates the direction and speed of movement. For movable nodes, this vector points to the nearest or most critical area requiring reinforcement, and its magnitude may be dynamically calculated based on the threat level and the node's own capabilities. For immovable nodes or nodes that do not need to move temporarily, this vector is zero. The other is the updated sampling frequency, which adjusts the sensor data acquisition cycle based on the node's relative position and importance to the critical area. For example, a node moving towards a critical area might increase its sampling frequency from 1 time / second to 5 times / second to provide more dense situational data; while a node far from a critical area might decrease its frequency to save energy.

[0053] Finally, after an adjustment action begins or is completed, the monitoring node needs to update its own status information in a timely manner and broadcast its latest status through the wireless ad hoc network. The updated status information includes at least the new location and the currently effective operating parameters. When a node broadcasts its new status, its neighboring nodes can perceive the changes in network layout and resource allocation in real time, which may trigger a new round of decision adjustments. For example, after node A begins moving towards a critical area, it immediately broadcasts: "Node A, new location (X', Y'), type vehicle-mounted, current sampling frequency 5Hz, target K1." Upon receiving this information, node B can, in the next round of decision-making, choose to move to another critical area K2 or adjust its speed because it perceives that node A has already moved towards K1, thereby achieving continuous optimization and conflict avoidance for the entire network.

[0054] After the monitoring nodes adjust their working parameters according to the decision and launch a new round of focused monitoring, the data collected by all monitoring nodes in the adjusted network are processed in a spatiotemporal fusion manner to generate a dynamic distribution map of the regional radiation field.

[0055] In one specific embodiment, monitoring data from all monitoring nodes after adjusting their operating parameters is first collected. This monitoring data includes location information and a timestamp. At the predetermined fusion period, the data processing center collects or receives the latest monitoring data uploaded by each monitoring node in the ad hoc network. The latest monitoring data is collected by each node at its new location and sampling frequency after adjusting its position and sampling frequency, thus better reflecting the true state and detailed characteristics of the current radiation field. Each data packet strictly adheres to a standard format, containing precise spatial coordinates and the acquisition time.

[0056] Next, all monitoring data undergoes spatiotemporal alignment preprocessing. This preprocessing includes unifying the data to the same reference time window based on timestamps and mapping the data to a unified spatial coordinate system based on location information. Specifically, time unification involves aligning all data to the same reference time window based on the timestamp of each data point. For data points that are not strictly simultaneous, algorithms such as linear interpolation can be used to convert them to estimated values ​​at the reference time. The spatial coordinate system involves transforming the location information of all nodes into a unified geographic coordinate system or a Cartesian coordinate system.

[0057] Then, based on the spatiotemporally aligned preprocessed monitoring data, a spatial interpolation algorithm is used to generate a spatial radiation intensity distribution surface covering the entire target area. The essence of the interpolation algorithm is to infer the values ​​at unknown locations based on the values ​​of known discrete points. In this scheme, commonly used algorithms include inverse distance weighted interpolation or Kriging interpolation. The data processing center divides the target area into regular grids. For each grid point, the estimated radiation intensity value is calculated based on the dose rate values ​​and distances of the surrounding known data points. For example, Kriging interpolation not only provides the best estimate for each grid point but also provides the variance of the estimate, characterizing the uncertainty of the interpolation result. The estimated values ​​of all grid points together constitute a digitized surface, which visually displays the continuous spatial distribution of radiation intensity at the reference time—that is, a static radiation hotspot map.

[0058] Finally, the distribution surface generated in the current time window is correlated with historical distribution surfaces from previous consecutive time windows to generate a dynamic distribution map describing changes in radiation field intensity and diffusion direction. This analysis can calculate and visualize a series of key dynamic features by comparing surfaces at different times, such as: which areas show significant increases or decreases in radiation intensity, displayed visually through color animation or contour line movement; estimating the overall movement direction and average migration speed of the radiation cloud by tracking the centroid movement trajectory of high-concentration areas or analyzing the temporal changes in concentration gradients; and dynamically displaying the expansion or contraction trend of contaminated areas exceeding a certain threshold over time. The final dynamic distribution map is presented in the form of video animation, sequence diagrams with timelines, or thematic maps overlaid with movement vectors, providing emergency command with visualized intelligence on the development trend of the situation.

[0059] The present invention provides a multi-node self-organizing network collaborative monitoring system for regional nuclear radiation, which is used to implement the above-mentioned multi-node self-organizing network collaborative monitoring method for regional nuclear radiation. The system includes multiple monitoring nodes and a data processing center.

[0060] The monitoring nodes are mobile and include at least a nuclear radiation sensor, an edge computing unit, and a wireless communication module. The nuclear radiation sensor is used for nuclear radiation measurement; the edge computing unit has a pre-built radiation spread prediction model; and the wireless communication module is used for network communication between nodes.

[0061] The multiple monitoring nodes form a wireless ad hoc network within the target area through their respective wireless communication modules. The edge computing unit is configured to perform the following operations: control the nuclear radiation sensor to perform synchronous measurements and generate local measurement data; run the radiation diffusion prediction model to generate local short-term radiation trend prediction data based on the local measurement data and data obtained from neighboring nodes; exchange short-term radiation trend prediction data through the wireless ad hoc network and identify key areas in a distributed and collaborative manner; and generate and execute adjustment instructions for the monitoring node's own operating parameters based on the identified key areas.

[0062] The data processing center is configured to perform the following operations: aggregate monitoring data collected by all monitoring nodes in the wireless ad hoc network; perform spatiotemporal fusion processing on the aggregated data to generate a dynamic distribution map of the regional radiation field.

[0063] The electronic device of the present invention includes a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the above-described method.

[0064] The present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0065] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0066] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0067] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the shown or discussed mutual couplings or direct couplings or communication connections may be through some service interfaces; indirect couplings or communication connections between apparatuses or units may be electrical or other forms.

[0068] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0069] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0070] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0071] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-node self-organizing network collaborative monitoring method for regional nuclear radiation, characterized in that, Using a mobile monitoring module with edge computing capabilities, and employing multiple such mobile monitoring modules as monitoring nodes, a wireless ad hoc network is formed within a target area. The method includes: Each monitoring node synchronously performs real-time nuclear radiation measurements in the target area to obtain its own local measurement data, which includes nuclear radiation data and environmental parameter data. The nuclear radiation data includes at least the gamma radiation dose rate. Each monitoring node runs a pre-set radiation diffusion prediction model based on local measurement data to generate local short-term radiation trend prediction data. The radiation diffusion prediction model takes local measurement data and data from adjacent monitoring nodes as input and outputs short-term radiation trend prediction data corresponding to the monitoring node. Each monitoring node exchanges short-term radiation trend prediction data through a wireless self-organizing network, and based on the exchanged data, identifies key areas that need to be monitored in a distributed and collaborative manner. Each monitoring node adjusts its own operating parameters based on the information of the key area to obtain the adjusted target distribution network. The adjustment of its own operating parameters includes at least some monitoring nodes moving to the key area to enhance the monitoring coverage of the key area. Data is collected from all monitoring nodes in the target distribution network, and spatiotemporal fusion processing is performed to generate a dynamic distribution map of the regional radiation field.

2. The multi-node self-organizing network collaborative monitoring method for regional nuclear radiation as described in claim 1, characterized in that, Each monitoring node synchronously performs real-time nuclear radiation measurements in the target area, obtaining its own local measurement data, specifically including: Each monitoring node receives a unified time signal and calibrates its local clock based on the time signal to achieve time synchronization of all monitoring nodes in the network. During the synchronized time period, the monitoring nodes simultaneously collect nuclear radiation data and environmental parameter data; The monitoring node performs local preprocessing on the collected raw data. The preprocessing includes at least data validity verification, background noise reduction, and unification of physical units. The pre-processed raw data is appended with corresponding timestamps and monitoring node location information, and then encapsulated into structured local measurement data.

3. The multi-node self-organizing network collaborative monitoring method for regional nuclear radiation as described in claim 1, characterized in that, Each monitoring node, based on local measurement data, runs a pre-set radiation diffusion prediction model to generate corresponding local short-term radiation trend prediction data, specifically including: The monitoring node acquires local measurement data and obtains its corresponding local measurement data from neighboring monitoring nodes through a wireless ad hoc network. The monitoring node takes local measurement data, data from adjacent monitoring nodes, and locally collected environmental parameter data as a set of input vectors and inputs them into the preset radiation diffusion prediction model. The radiation diffusion prediction model performs calculations at the monitoring nodes and outputs the predicted radiation intensity value after a predetermined time in the future, as well as the spatial gradient trend of the predicted radiation intensity value at the monitoring node location. The monitoring nodes encapsulate the predicted radiation intensity values ​​and spatial gradient change trends to generate structured short-term radiation trend prediction data.

4. The multi-node self-organizing network collaborative monitoring method for regional nuclear radiation as described in claim 1, characterized in that, Each monitoring node exchanges short-term radiation trend prediction data through a wireless ad hoc network, and based on the exchanged data, identifies key areas requiring focused monitoring in a distributed and collaborative manner, specifically including: Each monitoring node broadcasts its own generated short-term radiation trend prediction data through a wireless ad hoc network and receives short-term radiation trend prediction data from at least one other monitoring node. Each monitoring node constructs a comprehensive trend distribution map covering the prediction area of ​​all nodes within its corresponding communication range, based on the spatial gradient change trend in its own and received short-term radiation trend prediction data. Based on the comprehensive trend distribution map, each monitoring node evaluates and votes on areas with significant gradient changes and predicted radiation intensity exceeding the threshold, according to preset judgment rules, as candidate areas. Each monitoring node exchanges voting information and uses a distributed consensus algorithm to confirm and merge candidate key areas, outputting key areas that need to be monitored in a network-wide consensus. The key area is a set of coordinates.

5. The multi-node self-organizing network collaborative monitoring method for regional nuclear radiation as described in claim 1, characterized in that, Each monitoring node adjusts its own operating parameters based on the information from the key area to obtain the adjusted target distribution network, specifically including: Each monitoring node acquires the coordinates of the key area and simultaneously acquires the real-time status information of other monitoring nodes within the ad hoc network. The status information includes at least the node location and node type. Each monitoring node generates a parameter adjustment instruction for itself based on a preset collaborative scheduling strategy, its own status, and the acquired global network information. The parameter adjustment instruction includes at least the target movement vector and the updated sampling frequency. The monitoring node adjusts according to the parameter adjustment command. After the adjustment is completed, the adjusted target distribution network is obtained, and its latest status information is updated and broadcast.

6. The multi-node self-organizing network collaborative monitoring method for regional nuclear radiation as described in claim 1, characterized in that, The process of collecting data from all monitoring nodes in the target distribution network, performing spatiotemporal fusion processing, and generating a dynamic distribution map of the regional radiation field specifically includes: Collect nuclear radiation data from all monitoring nodes after adjusting their operating parameters. The nuclear radiation data also includes location information and timestamps. All nuclear radiation data are subjected to spatiotemporal alignment preprocessing, which includes unifying the data to the same reference time window based on timestamps and mapping the data to a unified spatial coordinate system based on location information. Based on the spatiotemporally aligned preprocessed nuclear radiation data, a spatial interpolation algorithm is used to generate a spatial radiation intensity distribution surface covering the entire target area. By performing a time-series correlation analysis between the distribution surface generated in the current time window and the historical distribution surfaces of multiple consecutive time windows, a dynamic distribution map describing the changes in radiation field intensity and diffusion direction is generated.

7. The multi-node self-organizing network collaborative monitoring method for regional nuclear radiation as described in claim 1, characterized in that, Before each monitoring node synchronously performs real-time nuclear radiation measurements in the target area and obtains its own local measurement data, the process also includes: Multiple monitoring nodes are deployed in the target area; Each monitoring node identifies other monitoring nodes within its communication range by listening to and responding to broadcast beacons, and negotiates to establish an initial wireless multi-hop communication link, forming a wireless ad hoc network. In the formed wireless ad hoc network, each monitoring node exchanges initial location information; The monitoring nodes enter a ready state and wait for a unified synchronization measurement command, which enables each monitoring node to synchronously perform real-time nuclear radiation measurements in the target area.

8. A multi-node self-organizing network collaborative monitoring system for regional nuclear radiation, characterized in that, For implementing the multi-node self-organizing network collaborative monitoring method for regional nuclear radiation as described in any one of claims 1 to 7, the system comprises multiple monitoring nodes and a data processing center, wherein... The monitoring nodes are mobile and include at least: Nuclear radiation sensor, used for nuclear radiation measurement; Edge computing units are pre-configured with radiation diffusion prediction models; and Wireless communication module, used for network communication between nodes; The multiple monitoring nodes form a wireless ad hoc network within the target area through their respective wireless communication modules; The edge computing unit is configured to perform the following operations: Control the nuclear radiation sensor to perform synchronous measurements and generate local measurement data; Run the radiation diffusion prediction model to generate local short-term radiation trend prediction data based on local measurement data and data obtained from neighboring nodes; The radiation short-term trend prediction data is exchanged through a wireless ad hoc network, and key areas are identified in a distributed and collaborative manner. Based on the identified key areas, generate and execute instructions for adjusting the working parameters of the monitoring nodes themselves. The data processing center is configured to perform the following operations: It aggregates monitoring data collected by all monitoring nodes in the wireless ad hoc network; The aggregated data undergoes spatiotemporal fusion processing to generate a dynamic distribution map of the regional radiation field.

9. An electronic device, characterized in that, The device includes a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1-7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method described in any one of claims 1-7.

Images