An emergency communication method and system for law enforcement recorders

Abstract

This invention discloses an emergency communication method and system for law enforcement recorders, relating to the field of communication networking technology. The method includes: when multiple law enforcement recorders enter each other's short-range communication range, they authenticate each other via a wireless ad hoc network protocol to form a dynamic collaborative communication group; a master sensing node continuously captures environmental video streams, obtains dynamic risk entropy values ​​and alarm data through image recognition, encapsulates them into emergency data packets, and sends them to a master relay node; the master relay node transmits the emergency data packets to the command center; when the performance status of the master sensing node or master relay node is detected to be deteriorating, a dynamic election process is triggered to re-elect the corresponding node. This invention dynamically elects the optimal master node through a lightweight distributed consensus algorithm based on real-time resource state vectors, achieving adaptive node switching and ensuring the continuous and stable operation of the emergency data acquisition and transmission link.

Description

Technical Field

[0001] This invention relates to the field of communication networking technology, and in particular to an emergency communication method and system for law enforcement recorders. Background Technology

[0002] With the increasing intelligence of mobile law enforcement equipment, its collaborative networking and emergency communication capabilities have become key development areas. Currently, law enforcement recorders generally support short-range wireless communication technologies (such as Wi-Fi Direct), enabling rapid networking between devices via ad-hoc networks and ensuring network security through digital certificate authentication mechanisms. Meanwhile, the application of distributed consensus algorithms and lightweight target detection models in edge computing devices is becoming increasingly mature, providing a technological foundation for multi-device collaborative perception and decision-making.

[0003] However, existing technologies mostly employ a network topology with fixed master nodes or manually designated relay nodes, lacking a dynamic adaptation mechanism to device resource status and communication link quality. Once the master node fails due to excessive computing load or signal quality deterioration, the entire collaborative group will be unable to continuously perform critical tasks, leading to interruptions in emergency data uploads or blind spots in risk perception, severely impacting real-time response capabilities and system reliability at law enforcement sites. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an emergency communication method for law enforcement recorders to solve the problem of overall functional failure of the collaborative group caused by the performance degradation of a single point in the fixed master node mode.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides an emergency communication method for law enforcement recorders, comprising: when multiple law enforcement recorders enter each other's short-range communication range, they authenticate each other through a wireless ad hoc network protocol to form a dynamic cooperative communication group; each law enforcement recorder in the dynamic cooperative communication group broadcasts its own real-time resource status vector to each other; by combining the real-time resource status vectors received by each law enforcement recorder, a master sensing node and a master relay node are dynamically elected by running a lightweight distributed consensus algorithm; the master sensing node continuously captures environmental video streams, obtains dynamic risk entropy values ​​and alarm data through image recognition, encapsulates them into emergency data packets, and sends them to the master relay node; the master relay node transmits the emergency data packets to the command center, and when the performance status of the master sensing node or the master relay node is detected to be deteriorated, a dynamic election process is triggered and a corresponding node is re-elected.

[0008] As a preferred embodiment of the emergency communication method of the law enforcement recorder described in this invention, the short-range communication range between them refers to the physical distance range that can maintain stable data exchange, determined by short-range wireless communication.

[0009] The formation of the dynamic collaborative communication group includes each law enforcement recorder authenticating itself by exchanging pre-set digital certificates, negotiating to generate a unique group identifier, and establishing a shared key for communication within the dynamic collaborative communication group.

[0010] As a preferred embodiment of the emergency communication method for law enforcement recorders described in this invention, the mutual broadcasting of their real-time resource status vectors refers to the set of current working status parameters sent by each law enforcement recorder to all other law enforcement recorders in the dynamic collaborative communication group when the resource status changes.

[0011] As a preferred embodiment of the emergency communication method of the law enforcement recorder described in this invention, the command center refers to a central dispatch entity used to receive, process and respond to emergency data packets from the law enforcement recorder.

[0012] The specific steps for dynamically electing the primary sensing node and the primary relay node are as follows:

[0013] A comprehensive performance score is calculated for each law enforcement recorder based on the real-time resource state vector;

[0014] Within the dynamic collaborative communication group, a voting proposal is initiated, and each law enforcement recorder votes to recommend candidates for the main sensing node and the main relay node based on a comprehensive performance score.

[0015] All votes are collected and tallied, and the law enforcement recorder that receives the most votes is determined as the primary sensing node; at the same time, the law enforcement recorder that scores the highest in the single item of signal quality of the communication link with the command center is determined as the primary relay node.

[0016] As a preferred embodiment of the emergency communication method of the law enforcement recorder described in this invention, in the process of the main sensing node continuously capturing environmental video stream, the image sensor image quality and computing load of the main sensing node are dynamically evaluated at a preset time interval, and the evaluation results are updated to the real-time resource status vector of the main sensing node itself.

[0017] As a preferred embodiment of the emergency communication method for the law enforcement recorder described in this invention, the specific steps for obtaining dynamic risk entropy values ​​and alarm data through image recognition and encapsulating them into an emergency data packet are as follows.

[0018] Decode and extract frames from continuously captured environmental video streams;

[0019] Each frame is analyzed using a deep learning-based object detection model to identify dangerous objects and abnormal behaviors of a pre-defined category.

[0020] Based on the evolution of hazardous objects and abnormal behaviors over time, calculate dynamic risk entropy values ​​and extract alarm data;

[0021] The dynamic risk entropy value, alarm data, and the timestamp and device identifier of the main sensing node are encapsulated together into an emergency data packet.

[0022] As a preferred embodiment of the emergency communication method of the law enforcement recorder described in this invention, when the main relay node transmits emergency data packets to the command center, it uses differential service code points to mark the emergency data packets.

[0023] As a preferred embodiment of the emergency communication method of the law enforcement recorder described in this invention, the performance degradation refers to the current load rate of the computing unit in the real-time resource status vector of the main sensing node continuously exceeding the first threshold for a predetermined time, or the signal quality index of the communication link with the command center in the real-time resource status vector of the main relay node continuously falling below the second threshold for a predetermined time.

[0024] As a preferred embodiment of the emergency communication method of the law enforcement recorder described in this invention, after triggering the dynamic election process and re-electing the corresponding node, the newly elected main sensing node or main relay node broadcasts a role switching completion notification to the dynamic collaborative communication group.

[0025] After receiving the notification that the role switch is complete, the original primary sensing node or the original primary relay node stops executing the corresponding original role tasks; the newly elected primary sensing node begins to take over the task of capturing and analyzing environmental video streams, or the newly elected primary relay node begins to take over the task of transmitting emergency data packets to the command center.

[0026] Secondly, this invention provides an emergency communication system for law enforcement recorders, comprising a network authentication module, a dynamic election module, a risk perception module, and a relay control module. The network authentication module is used to mutually authenticate each other via a wireless ad hoc network protocol when multiple law enforcement recorders enter each other's short-range communication range, forming a dynamic collaborative communication group. Each law enforcement recorder in the dynamic collaborative communication group broadcasts its own real-time resource status vector to each other. The dynamic election module is used to synthesize the real-time resource status vectors received by each law enforcement recorder and dynamically elect a master perception node and a master relay node by running a lightweight distributed consensus algorithm. The risk perception module is used for the master perception node to continuously capture environmental video streams, obtain dynamic risk entropy values ​​and alarm data through image recognition, encapsulate them into emergency data packets, and send them to the master relay node. The relay control module is used for the master relay node to transmit emergency data packets to the command center, and when the performance status of the master perception node or the master relay node is detected to be deteriorating, it triggers a dynamic election process and re-elects the corresponding node.

[0027] The beneficial effects of this invention are as follows: By dynamically electing the master sensing node and the master relay node through a lightweight distributed consensus algorithm, comprehensive performance evaluation and voting decisions are made based on multi-dimensional parameters such as battery power, computing load, memory utilization, and signal quality index in the real-time resource state vector. This ensures that the elected master node is always the best device in the current group, realizes adaptive switching of the master node, avoids task interruption caused by the performance degradation of a single device, and ensures the continuous and stable operation of emergency data collection and transmission links through periodic resource state updates and triggering re-election. This significantly improves the robustness and execution efficiency of multi-device collaboration. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. 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.

[0029] Figure 1 This is a flowchart illustrating emergency communication methods for law enforcement recorders.

[0030] Figure 2 This is a flowchart of the dynamic election process.

[0031] Figure 3 This is a schematic diagram of the risk perception module.

[0032] Figure 4 This is a schematic diagram of the relay control module. Detailed Implementation

[0033] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0034] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0035] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0036] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides an emergency communication method for a law enforcement recorder, comprising the following steps:

[0037] S1: When multiple law enforcement recorders enter each other's short-range communication range, they authenticate each other through a wireless ad hoc network protocol to form a dynamic cooperative communication group; each law enforcement recorder in the dynamic cooperative communication group broadcasts its own real-time resource status vector to each other.

[0038] It should be noted that the real-time resource status vector broadcast by each other refers to the set of its current working status parameters sent by each law enforcement recorder to all other law enforcement recorders in the dynamic collaborative communication group when the resource status changes.

[0039] S1.1: The short-range communication range between each other refers to the physical distance range that can maintain stable data exchange, as determined by short-range wireless communication.

[0040] Specifically, the body camera continuously scans for short-range wireless communication signals in the surrounding area via Wi-Fi Direct. The body camera measures the received signal strength (RSS) readings between itself and other body cameras. When the RSS reading consistently exceeds a stable communication threshold, the body camera determines that it has entered the short-range communication range of the other device. The stable communication threshold is the minimum signal strength limit determined based on the receiver sensitivity requirements of the wireless communication protocol and channel noise characteristics, used to determine whether a reliable data connection can be maintained between the devices. For example, the stable communication threshold might be set to -70 dBm for a duration of 5 seconds.

[0041] S1.2: Forming a dynamic collaborative communication group involves each law enforcement recorder authenticating itself by exchanging pre-set digital certificates, negotiating to generate a unique group identifier, and establishing a shared key for communication within the dynamic collaborative communication group.

[0042] Specifically, once a law enforcement recorder enters the short-range communication range, it initiates a wireless ad-hoc network protocol, using the IEEE 802.11s protocol to establish an ad-hoc network between devices. The recorder initializes its network interface, configures ad-hoc mode, and continuously monitors the wireless channel and responds to network requests.

[0043] The body cameras connect via an established ad-hoc network and exchange pre-installed X.509 digital certificates. Each body camera stores a digital certificate issued by a trusted certificate authority. Upon receiving a digital certificate, the body camera uses public key infrastructure to verify its validity, including checking the certificate signature, validity period, and revocation status.

[0044] After successful authentication, the body cameras negotiate and generate a unique group identifier. Each body camera shares its own device identifier (such as MAC address), and uses the SHA-256 hash function to hash all device identifiers, taking the first 64 bits of the hash value as the unique group identifier;

[0045] If authentication fails—that is, if any certificate to be verified fails any of the three checks: certificate signature, validity period, and revocation status—the body camera immediately terminates the current network formation process. The body camera that triggered the authentication failure will record a failure event log (including the failed certificate identifier and the reason for the failure) and send an authentication failure alarm message to the command center via the wireless communication interface. Simultaneously, the body camera will not join the dynamic collaborative communication group and will continue to scan for signals from other body cameras to attempt new network authentication. The command center is the central dispatch entity used to receive, process, and respond to emergency data packets from body cameras; the command center is connected to the body cameras via a wireless communication network.

[0046] It should be noted that the first 64 bits of the hash value are used as the unique group identifier because the 64-bit length ensures uniqueness while taking into account computational efficiency, storage space and network transmission overhead, which is a common balance choice in engineering practice.

[0047] Furthermore, the body cameras use the elliptic curve Diffie-Hellman key exchange protocol to generate symmetric keys. Each body camera generates a temporary key pair, exchanges public keys, and derives a shared key for group communication by calculating a shared secret. The shared key is used to encrypt all communication data within the dynamic collaborative communication group.

[0048] After the shared key is established, the body camera confirms the connection by broadcasting an encrypted message, including a unique group identifier and shared key information. Once all body cameras have confirmed the connection, the dynamic collaborative communication group is officially established.

[0049] Specifically, after the dynamic collaborative communication group is formed, each law enforcement recorder begins broadcasting its own real-time resource status vector to each other. The real-time resource status vector includes operating status parameters such as battery level, computing unit load rate, memory utilization, and signal quality index of the communication link with the command center. A change in resource status refers to a change in the current value of any operating status parameter of the law enforcement recorder. For example, when the battery level drops by 5% or the computing unit load rate changes by more than 10%, the law enforcement recorder sends its real-time resource status vector to all other law enforcement recorders in the dynamic collaborative communication group.

[0050] S2: By integrating the real-time resource status vectors received by each law enforcement recorder, a lightweight distributed consensus algorithm is used to dynamically elect the main sensing node and the main relay node.

[0051] S2.1: Calculate the comprehensive performance score for each law enforcement recorder based on the real-time resource state vector.

[0052] Specifically, the battery level, compute load, memory utilization, and signal quality index parameters are extracted from the real-time resource state vector. Battery level, compute load, and memory utilization are expressed as percentage values, and the signal quality index is a value normalized to 0-100, where 100 represents optimal quality. The overall performance score is calculated using a weighted summation formula.

[0053] S2.2: Initiate a voting proposal within the dynamic collaborative communication group, and each law enforcement recorder votes to recommend candidates for the main sensing node and the main relay node based on comprehensive performance scores.

[0054] Specifically, after completing the comprehensive performance score calculation, each law enforcement recorder within the dynamic collaborative communication group initiates a voting proposal using a lightweight distributed consensus algorithm. The voting proposal is triggered by any law enforcement recorder within the dynamic collaborative communication group broadcasting a voting request message, which includes a voting identifier and a unique group identifier.

[0055] It should be noted that lightweight distributed consensus algorithms specifically refer to low-overhead consensus mechanisms designed for resource-constrained law enforcement recorder hardware environments. Lightweight distributed consensus algorithms simplify the classic distributed election concept, reducing computational and energy consumption by decreasing communication rounds, compressing message load, and limiting the size of participating nodes. Specifically, implementations can adopt existing lightweight protocol variants such as the Gossip protocol or the Bully algorithm, initiating proposals by broadcasting voting request messages (including voting identifiers and unique group identifiers), and relying on local interactions between nodes rather than global synchronization to achieve consensus, thus meeting the need for fast and reliable election of master nodes in dynamic networking environments.

[0056] After receiving the voting request message, each law enforcement recorder executes steps A1 and A2 for voting recommendation:

[0057] Step A1: Recommend candidate main sensing node based on comprehensive performance score: Each law enforcement recorder selects the law enforcement recorder with the highest comprehensive performance score as the candidate main sensing node;

[0058] Step A2: Recommend candidate for primary relay node based on signal quality index: Each law enforcement recorder selects the law enforcement recorder with the highest signal quality index as the candidate for primary relay node.

[0059] Each law enforcement recorder encapsulates the voting recommendation results into a voting response message. The voting response message contains the device identifiers of the recommended primary sensing node candidate and the recommended primary relay node candidate, and is then encrypted using the shared key for communication within the dynamic collaborative communication group before being broadcast to all other law enforcement recorders within the dynamic collaborative communication group.

[0060] S2.3: Collect and count all votes, and determine the law enforcement recorder that receives the most votes as the primary sensing node; at the same time, determine the law enforcement recorder that scores the highest in the single item of signal quality of the communication link with the command center backend as the primary relay node.

[0061] It should be noted that the signal quality of the command center's back-end communication link specifically refers to the state quantification index of the wireless communication link between the main relay node and the command center's back-end server. The state quantification index is an independent data item (i.e., a "single item") of the real-time resource state vector;

[0062] Specifically, the primary relay node periodically measures the physical layer parameters (such as Received Signal Strength Indicator (RSSI) or Signal-to-Noise Ratio (SNR)) of the communication link with the command center. These physical layer parameters are converted into a normalized, dimensionless value, typically ranging from 0 to 100, using a preset mapping relationship (such as a piecewise linear function or lookup table). This value represents the best quality, while 0 represents a connection interruption. The converted value is then used as a "signal quality index" to update the real-time resource status vector. This index is specifically used to assess the health of the communication link between the primary relay node and the command center, and serves as the core basis for dynamically electing the primary relay node.

[0063] Furthermore, each law enforcement recorder within the dynamic collaborative communication group collects all voting response messages, decrypts them, and extracts the device identifiers of the recommended primary sensing node candidate and the recommended primary relay node candidate. The number of votes recommended for the primary sensing node candidate is counted, and the law enforcement recorder device identifier with the most votes is determined as the primary sensing node. Simultaneously, the signal quality index in the real-time resource state vectors of each law enforcement recorder within the dynamic collaborative communication group is compared, and the law enforcement recorder device identifier with the highest signal quality index value is determined as the primary relay node.

[0064] A better approach is to use a lightweight distributed consensus algorithm for voting and decision-making based on real-time resource state vectors. Compared with conventional networking methods that rely on fixed master nodes or manual designation, this improves the adaptability and operational reliability of the device group, effectively avoids overall functional failure caused by the performance degradation of designated devices, and ensures that the optimal device always plays a key role in emergency situations, thereby enhancing the efficiency and stability of multi-device collaboration.

[0065] S3: The main sensing node continuously captures environmental video streams, obtains dynamic risk entropy values ​​and alarm data through image recognition, encapsulates them into emergency data packets, and sends them to the main relay node.

[0066] S3.1: During the continuous capture of environmental video stream by the main sensing node, the image sensor image quality and computational load of the main sensing node are dynamically evaluated at preset time intervals, and the evaluation results are updated to the real-time resource state vector of the main sensing node itself.

[0067] Specifically, the master sensing node continuously captures environmental video streams through its built-in image sensor. The master sensing node performs evaluation operations at fixed time intervals (determined by the device's preset configuration). Image sensor image quality evaluation is achieved by calculating the sharpness index of video frames, which is calculated using the Brenner gradient function to measure the rate of change of grayscale values ​​of pixels in the frame image. Computational load evaluation is achieved by reading the current CPU utilization percentage of the device's operating system. The master sensing node updates the obtained sharpness index value and CPU utilization percentage to the corresponding image sensor image quality field and computational load field in the real-time resource status vector.

[0068] It should be noted that the device's preset configuration refers to a set of parameters pre-stored in the law enforcement recorder's non-volatile memory, used to define the time intervals for various operations. Specifically, the fixed time interval values ​​are written into the device firmware by the device manufacturer during the production process or by the administrator before deployment using a dedicated configuration tool.

[0069] S3.2: Decode and extract frames from the continuously captured environmental video stream; use a deep learning-based object detection model to analyze each frame and identify dangerous objects and abnormal behaviors of preset categories.

[0070] Specifically, the master sensing node performs H.264 decoding on the continuously captured environmental video stream, and then extracts video frames at a fixed frame rate. The YOLOv4 deep learning-based object detection model is used to analyze each frame.

[0071] It should be noted that the pre-training process of the YOLOv4 object detection model uses a training set containing pre-defined category labeled data such as weapons, vehicles, crowds, running actions, and falling actions. The backpropagation algorithm is used to optimize the neural network weights of the YOLOv4 object detection model, the cross-entropy loss function is used to calculate the error between the predicted bounding box and the ground truth label, and the gradient descent method is used to iteratively update the object detection model parameters until the loss converges.

[0072] The YOLOv4 object detection model, after pre-training, can identify dangerous objects and abnormal behaviors of preset categories. The model outputs the category label and confidence score of the dangerous objects and abnormal behaviors detected in each frame, retaining only detection results with confidence scores exceeding a recognition threshold.

[0073] The identification threshold refers to the preset threshold value used by the YOLOv4 object detection model to filter low-confidence detection results during the inference process. The preset threshold value is determined by the object detection model during the training phase based on the performance of the validation set and is fixed in the object detection model configuration file. Only detection results with confidence scores higher than the preset threshold value are output to ensure the reliability of identification.

[0074] Example: For instance, if the recognition threshold is set to 0.7, when the target detection model identifies a "weapon" target in an image frame, but the confidence score of this target is only 0.65 (below 0.7), the main perception node will discard this detection result and will not include the detection result in the calculation of dynamic risk entropy or alarm data.

[0075] S3.3: Calculate the dynamic risk entropy value and extract alarm data based on the evolution of dangerous objects and abnormal behaviors over time.

[0076] Specifically, the main sensing node analyzes the evolution of dangerous objects and abnormal behaviors over time based on the detection results of the target detection model in continuous video frames.

[0077] A fixed time window is set in the main sensing node to statistically analyze the occurrence of various dangerous objects and abnormal behaviors within the fixed time window. Subsequently, the main sensing node calculates the frequency probability of each preset category occurring within the fixed time window. ,in It represents any one of the preset categories; the length of the fixed time window is uniformly set by the device's preset configuration, including the length of the analysis period used to statistically analyze the occurrence of dangerous objects and abnormal behaviors.

[0078] based on The main sensing node uses the information entropy formula to calculate the dynamic risk entropy value. Dynamic risk entropy value The magnitude of the value reflects the degree of uncertainty of the dangerous situation within the time window;

[0079] The formula for information entropy is:

[0080] ;

[0081] In the formula, Represents any one of the preset categories. This represents the frequency probability of a predefined category of hazardous objects or abnormal behaviors occurring within a time window. With the natural logarithm e as the base, express The contribution value to the overall dynamic risk entropy value. This represents the sum of all possible negative contributions, converted to a positive value.

[0082] While calculating the dynamic risk entropy value, the main sensing node extracts alarm data. The alarm data includes at least the location, type, and trajectory of the hazardous target. The location of the hazardous target is represented by the center coordinates of the bounding box output by the YOLOv4 target detection model; the type of the hazardous target is represented by the category label output by the target detection model; and the trajectory of the hazardous target is represented by the changing path of the center coordinates of the bounding box of the same target in consecutive video frames.

[0083] S3.4: Encapsulate the dynamic risk entropy value, alarm data, and the timestamp and device identifier of the main sensing node into an emergency data packet.

[0084] Specifically, the emergency data packet uses JSON format to organize data fields, including the risk_entropy field, the alert_data field (containing subfields: position, type, trajectory), the timestamp field, and the device_id field. After encapsulation, the primary sensing node encrypts the emergency data packet using the shared key for communication within the dynamic collaborative communication group and sends it to the primary relay node.

[0085] The preferred approach is to use the information entropy formula to quantify the evolution of dangerous targets over time. Compared with existing alarm mechanisms based on simple threshold triggering or single qualitative judgment, this approach can effectively quantify risk uncertainty and provide multi-dimensional comprehensive assessment, significantly reducing false alarms and missed alarms. This allows the command center to obtain more accurate and forward-looking early warning information, facilitating timely and precise responses, thereby significantly improving law enforcement safety and the level of intelligence in on-site handling.

[0086] S4: The primary relay node transmits emergency data packets to the command center. When the performance status of the primary sensing node or the primary relay node is detected to be deteriorating, a dynamic election process is triggered and the corresponding node is re-elected.

[0087] S4.1: When the primary relay node transmits an emergency data packet to the command center, it uses differential service code points to mark the emergency data packet.

[0088] Specifically, after receiving the encrypted emergency data packet from the primary sensing node, the primary relay node decrypts and verifies the emergency data packet to ensure data integrity. Upon successful verification, the primary relay node repackages the emergency data packet into a data format conforming to the IP network transmission protocol. In the Type of Service field of the IP packet header, based on the differential service architecture defined by the Internet Engineering Task Force, a differential service code point value corresponding to the Accelerated Forwarding PHB (Per-Hop Behavior) is set to mark the emergency data packet as high-priority emergency data requiring low latency and low packet loss protection. The primary relay node then transmits the marked emergency data packet to the command center server via the established wireless communication network connection.

[0089] It should be noted that the command center refers to the central dispatch entity used to receive, process and respond to emergency data packets from law enforcement recorders; the command center server IP address is pre-configured in the main relay node equipment configuration.

[0090] S4.2: Continuously monitor performance status indicators and determine performance degradation and trigger conditions.

[0091] It should be noted that performance degradation refers to the current load rate of the computing unit in the real-time resource status vector of the main sensing node continuously exceeding the first threshold for a predetermined time, or the signal quality index of the communication link with the command center in the real-time resource status vector of the main relay node continuously falling below the second threshold for a predetermined time.

[0092] Specifically, all law enforcement recorders within the dynamic collaborative communication group continuously monitor their own real-time resource status vectors. The main sensing node monitors the current load rate parameter of the computing unit in the real-time resource status vector. The load rate parameter is obtained by reading the current CPU utilization percentage of the device's operating system. The main relay node monitors the signal quality index parameter of the communication link with the command center's backend in the real-time resource status vector. The signal quality index parameter is obtained by measuring the received signal strength indication value of the wireless communication interface and converting it into a normalized value of 0-100. The monitoring operation is executed periodically at fixed time intervals.

[0093] When the current load rate of the computing unit of the primary sensing node continuously exceeds the first threshold for a predetermined duration, the performance of the primary sensing node is deemed to have deteriorated. When the signal quality index of the communication link between the primary relay node and the command center in the real-time resource status vector continuously falls below the second threshold for a predetermined duration, the performance of the primary relay node is deemed to have deteriorated. It should be noted that the specific values ​​of the first threshold, the second threshold, and the predetermined time are determined by the device's preset configuration, and the setting standards comprehensively consider the hardware performance limits of the law enforcement recorder and the minimum communication quality requirements required to maintain reliable communication with the command center.

[0094] When any law enforcement recorder detects a deterioration in its own performance or learns of a deterioration in the performance of other law enforcement recorders through real-time resource status vector broadcasting, it immediately broadcasts a request message to the dynamic collaborative communication group to trigger a dynamic election process.

[0095] Furthermore, after receiving the request message to trigger the dynamic election process, the dynamic collaborative communication group re-elects a master sensing node or master relay node according to the lightweight distributed consensus algorithm in step S2. The newly elected master sensing node or master relay node generates a role switch completion notification message;

[0096] The role switchover completion notification message contains the device identifier and role type of the newly elected node. It is encrypted using the shared key for communication within the dynamic collaborative communication group and broadcast to the group. Upon receiving and decrypting the role switchover completion notification message, the original primary sensing node or primary relay node ceases executing its original environmental video stream capture and analysis task or emergency data packet transmission task. The newly elected primary sensing node immediately begins taking over the environmental video stream capture and analysis task, or the newly elected primary relay node immediately begins taking over the transmission of emergency data packets to the command center.

[0097] This embodiment also provides an emergency communication system for law enforcement recorders, including: a network authentication module, a dynamic election module, a risk perception module, and a relay control module; the network authentication module is used to authenticate each other via a wireless ad hoc network protocol when multiple law enforcement recorders enter each other's short-range communication range, forming a dynamic collaborative communication group; each law enforcement recorder in the dynamic collaborative communication group broadcasts its own real-time resource status vector to each other; the dynamic election module is used to integrate the real-time resource status vectors received by each law enforcement recorder and dynamically elect a master perception node and a master relay node by running a lightweight distributed consensus algorithm; the risk perception module is used for the master perception node to continuously capture environmental video streams, obtain dynamic risk entropy values ​​and alarm data through image recognition, encapsulate them into emergency data packets, and send them to the master relay node; the relay control module is used for the master relay node to transmit emergency data packets to the command center, and when the performance status of the master perception node or the master relay node is detected to be deteriorating, a dynamic election process is triggered and a corresponding node is re-elected.

[0098] In summary, this invention dynamically elects the master sensing node and master relay node through a lightweight distributed consensus algorithm. Based on multi-dimensional parameters such as battery power, computing load, memory utilization, and signal quality index in the real-time resource state vector, it conducts comprehensive performance evaluation and voting decisions to ensure that the elected master node is always the best device in the current group. This achieves adaptive switching of the master node and avoids task interruption caused by the performance degradation of a single device. At the same time, through periodic resource state updates and triggering re-election, it ensures the continuous and stable operation of emergency data collection and transmission links, significantly improving the robustness and execution efficiency of multi-device collaboration.

[0099] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for emergency state communication of a law enforcement recording device, the method comprising: Comprising, When multiple law enforcement recorders enter each other's close communication range, mutual authentication is performed through a wireless ad hoc network protocol to form a dynamic cooperative communication group; Each law enforcement recorder in the dynamic cooperative communication group broadcasts its real-time resource state vector to each other; The main sensing node and the main relay node are dynamically elected by integrating the real-time resource state vectors received by each law enforcement recorder and running a lightweight distributed consensus algorithm, and the specific steps are as follows, Based on the real-time resource state vector, the comprehensive performance score of each law enforcement recorder is calculated; A voting proposal is initiated within the dynamic cooperative communication group, and each law enforcement recorder votes for the main sensing node candidate and the main relay node candidate based on the comprehensive performance score; All votes are collected and counted, and the law enforcement recorder that receives the most votes is determined as the main sensing node; at the same time, the law enforcement recorder with the highest score in the signal quality of the communication link with the background of the command center is determined as the main relay node; the command center refers to a central dispatching entity for receiving, processing and responding to emergency data packets from law enforcement recorders; The main sensing node continuously captures environmental video streams, obtains dynamic risk entropy values and alarm data through image recognition, and encapsulates them into emergency data packets, which are sent to the main relay node; The specific steps of obtaining dynamic risk entropy values and alarm data through image recognition and encapsulating them into emergency data packets are as follows, The continuously captured environmental video streams are decoded and frame extracted; Each frame is analyzed using a target detection model based on deep learning to identify dangerous objects and abnormal behaviors of a preset category; According to the evolution of dangerous objects and abnormal behaviors in time series, dynamic risk entropy values are calculated and alarm data are extracted; The dynamic risk entropy values, alarm data, and timestamps and device identifiers of the main sensing node are collectively encapsulated into emergency data packets; The main relay node transmits the emergency data packets to the command center, and when it detects that the performance state of the main sensing node or the main relay node deteriorates, it triggers a dynamic election process and re-elects the corresponding node.

2. The method of claim 1, wherein: The close communication range of each other refers to a physical distance range determined through short-range wireless communication that can maintain stable data exchange; The formation of the dynamic cooperative communication group includes identity authentication of each law enforcement recorder through the exchange of pre-set digital certificates, negotiation to generate a unique group identifier, and establishment of a shared key for communication within the dynamic cooperative communication group.

3. The method of claim 1, wherein: The mutual broadcast of real-time resource state vectors refers to the sending of a set of current working state parameters of each law enforcement recorder to all other law enforcement recorders within the dynamic cooperative communication group when the resource state changes.

4. The method of claim 1, wherein: During the process of continuously capturing environmental video streams by the main sensing node, the image sensor picture quality and computational load of the main sensing node are dynamically evaluated at a preset time interval, and the evaluation results are updated to the real-time resource state vector of the main sensing node itself.

5. The method of claim 1, wherein: When the main relay node transmits the emergency data packets to the command center, it marks the emergency data packets with a differentiated services code point.

6. The method of claim 1, wherein: The performance state deterioration refers to that the current load rate of the computing unit in the real-time resource state vector of the master perception node continuously exceeds a first threshold for a predetermined time, or that the signal quality single index of the background communication link with the command center in the real-time resource state vector of the master relay node continuously is lower than a second threshold for a predetermined time.

7. The method of claim 1, wherein: After triggering the dynamic election process and reelecting the corresponding node, the newly elected master perception node or master relay node broadcasts a role switching completion notification to the dynamic cooperative communication group. After receiving the role switching completion notification, the original master perception node or the original master relay node stops performing the corresponding original role task; the newly elected master perception node starts to take over the capture and analysis task of the environmental video stream, or the newly elected master relay node starts to take over the transmission task of the emergency data packet to the command center.

8. An emergency state communication system of a law enforcement recording apparatus, based on the emergency state communication method of any one of claims 1 to 7, characterized in that: The network authentication module, the dynamic election module, the risk perception module and the relay control module are included. The network authentication module is configured to perform mutual authentication through a wireless ad hoc network protocol when the multiple law enforcement recorders enter the close communication range of each other, to form a dynamic cooperative communication group; and each law enforcement recorder of the dynamic cooperative communication group broadcasts its real-time resource state vector. The dynamic election module is configured to comprehensively consider the real-time resource state vectors received by each law enforcement recorder, and dynamically elect a master perception node and a master relay node by running a lightweight distributed consensus algorithm. The risk perception module is configured to continuously capture an environmental video stream by the master perception node, to obtain a dynamic risk entropy value and alarm data through image recognition, and to encapsulate the dynamic risk entropy value and the alarm data into an emergency data packet and send the emergency data packet to the master relay node. The relay control module is configured to transmit the emergency data packet to the command center by the master relay node, and to trigger a dynamic election process and reelect a corresponding node when a performance state deterioration of the master perception node or the master relay node is monitored.

Images