A Dynamic Topology Optimization Method for Fire and Rescue On-site Command and Control Network Based on Mesh Node State Awareness

Abstract

This invention discloses a dynamic topology optimization method for fire and rescue command and control networks based on mesh node state awareness. It relates to the fields of emergency communication and mobile ad hoc network technology for fire and rescue teams. The method includes mesh node state self-identification, collecting mesh node motion data and determining whether a node is stationary or moving using a lightweight sensor fusion algorithm; constructing static direct links to form a static backbone subnet; and adopting a hierarchical heterogeneous routing strategy: for each static backbone subnet, internally, a logical subnet coordinator is elected, a shortest path algorithm is run to construct static routes, and data forwarding is implemented. This invention fundamentally solves the problem of surging routing overhead and network congestion when traditional mesh networks expand, ensuring stable network performance in large-scale rescue scenarios with multiple nodes and wide coverage.

Description

Technical Field

[0001] This invention belongs to the field of emergency communication and mobile ad hoc network technology for fire and rescue teams. Specifically, it relates to a method for dynamic topology segmentation and routing strategy optimization of a wireless mesh-based command and control network in fire disaster rescue scenarios, in order to solve the performance collapse problem caused by the uncontrollable growth of the network size of the on-site command and control network as the rescue progresses. Background Technology

[0002] Currently, the command and combat network of my country's fire and rescue teams has transformed from narrowband voice to broadband multimedia convergence technology. The command and combat network based on broadband applications has initially built a hybrid architecture with wireless mesh self-organizing network as the core and satellite and public network as backhaul, realizing the basic carrying capacity of on-site high-definition video, individual soldier area network data, environmental perception data and voice trunking communication.

[0003] However, in urban mega-complexes, complex underground spaces, and natural disaster rescue sites, as the on-site command and control network continues to expand, its scale becomes increasingly massive. This creates a fundamental and irreconcilable conflict between the network's size and the physical limits of its overall performance. For example... Figure 2 The diagram shows the existing network communication structure.

[0004] The core issue lies in the mismatch between the design paradigm of existing Mesh networks based on general standards (such as IEEE 802.11s) and the dynamic topology of disaster sites. At rescue sites, approximately 70% of the nodes, including backpack relays deployed at passageways and entrances, form a relatively stable backbone network after deployment. The remaining 30%, such as individual attackers, move rapidly at the ends of the backbone. Current protocols treat all nodes as homogeneous, equal, and dynamic entities, forcing both stable and mobile links to be involved in continuous network-wide route detection and updates. This control mechanism causes network control overhead to increase quadratically with the number of nodes. When the number of nodes exceeds a critical threshold of 30 to 50, effective bandwidth decreases sharply, leading to video stuttering, command delays, and even network-wide paralysis. Essentially, radio resources are wasted on unnecessary detection of static topologies.

[0005] At a deeper level, this bottleneck exposes the fact that field communication equipment has hardware but lacks intelligence: existing Mesh nodes can only mechanically execute protocols, lacking any awareness of the network equipment's own role, link status, and overall network situation. The network cannot distinguish between critical internal attack video streams and ordinary status data streams, lacks an intelligent scheduling mechanism based on services and topology, and its performance is highly dependent on the experience-based deployment of field communicators, making it fragile and unpredictable in complex environments.

[0006] The current technical problems are as follows:

[0007] Theoretical model defects: Classic Mesh routing protocols (such as HWMP and AODV) are designed based on the premise that the network topology is completely dynamic or completely static. In a completely dynamic model, the protocol continuously calculates the optimal path for all nodes, resulting in massive control overhead. In a completely static model, the flexibility to deal with local changes is lost.

[0008] Semi-dynamic characteristics of rescue site topology: In actual fire rescue sites, network topology exhibits highly regular semi-dynamic or hierarchical dynamic characteristics. Approximately 70% of nodes (such as Mesh relay nodes deployed at building entrances, stairwells, and command centers) remain fixed in their positions after deployment, forming a relatively stable backbone topology. The remaining 30% of nodes (individual soldier nodes) move rapidly at the "periphery" of the backbone topology. General protocols ignore this characteristic and perform unnecessary periodic route probing on stable links, resulting in a huge waste of protocol overhead.

[0009] In summary, the current construction of broadband command and control networks for fire and rescue teams lacks a solution that can automatically identify the inherent stability characteristics of the network topology and manage the routing mechanism in a refined and differentiated manner, making it difficult to guarantee the reliability of Mesh command and control networks in large-scale and complex rescue operations. Summary of the Invention

[0010] To address the shortcomings of existing technologies, this invention provides a dynamic topology optimization method for fire and rescue command and control networks based on Mesh node status awareness. By reconstructing the network into a heterogeneous hierarchical structure combining static backbone subnets and dynamic access edges, the effective node size of the dynamic routing plane is reduced from the total number of network nodes to the sum of the number of mobile nodes and the number of static subnets. This significantly reduces the proportion of routing control overhead, fundamentally solving the problems of soaring routing overhead and network congestion when traditional Mesh networks expand in scale, and ensuring stable network performance in large-scale rescue scenarios with multiple nodes and wide coverage.

[0011] To achieve the above objectives, this invention provides the following technical solution: a dynamic topology optimization method for fire and rescue on-site command and control network based on Mesh node state awareness, comprising the following steps:

[0012] Step 1, Mesh Node Status Self-Identification: A nine-axis inertial measurement unit is integrated into each Mesh networking device to collect the motion data of the Mesh nodes and determine whether the nodes are stationary or moving through a lightweight sensor fusion algorithm. The status flag is encapsulated into the topology status notification message and the status is broadcast periodically.

[0013] Step 2: Construct static through links and form static backbone subnets: Each Mesh node maintains an enhanced neighbor table for topology discovery and link classification. Static through links are filtered based on node status and link quality thresholds. Then, through the distributed connected component discovery algorithm, the undirected graph composed of all static through links is identified as multiple unconnected static backbone subnets.

[0014] Step 3: Adopt a hierarchical heterogeneous routing strategy: For each static backbone subnet, a logical subnet coordinator is elected internally, and the shortest path algorithm is run to construct static routes and realize data forwarding; for communication between subnets and between static backbone subnets and mobile nodes, each static backbone subnet is defined as a logical super node, and the propagation scope of routing requests is limited to all mobile nodes and the gateway nodes of each static backbone subnet, so as to suppress the explosive growth of routing control overhead with the network size and ensure the efficiency, reliability and stability of communication of the fire rescue on-site command and control network.

[0015] In step 1, the nine-axis inertial measurement unit includes a three-axis gyroscope, a three-axis accelerometer, and a three-axis magnetometer, which are used to collect motion data of the Mesh nodes.

[0016] The lightweight sensor fusion algorithm described in step 1 is a complementary filtering algorithm, which processes the collected motion data.

[0017] If a Mesh node satisfies the following conditions: its angular velocity vector value is continuously less than the angular velocity threshold for a certain period of time, and the variance of the accelerometer readings within the past window is less than the acceleration variance threshold, then the Mesh node is determined to be in a stationary state. If these conditions are not met, the Mesh node is determined to be in a moving state.

[0018] The period of status broadcast in step 1 is dynamically adjusted according to the network environment at the fire and rescue site.

[0019] In step 2, the enhanced neighbor table records the neighbor ID, signal strength, neighbor status, and link quality to complete neighbor topology discovery and link classification.

[0020] For any two Mesh nodes, determine whether they meet the preset conditions within the time window. If they do, the nodes negotiate to upgrade the link to a static pass-through link.

[0021] The preset conditions must be met simultaneously:

[0022] Both ends of the link are stationary.

[0023] The average link quality is higher than a preset threshold for the average link quality.

[0024] The link quality variance is less than the preset link quality variance threshold.

[0025] All static direct links form an undirected graph. A distributed connected component discovery algorithm identifies several unconnected connected subgraphs, each of which is a static backbone subnet.

[0026] Step 3 specifically involves:

[0027] The network routing space of the fire command and control network is divided into two independent layers: the static routing plane within the subnet and the dynamic routing plane between subnets and dynamic nodes.

[0028] Static routing implementation within a subnet:

[0029] For any static backbone subnet, pre-computed optimized tree or mesh static routing is used internally. Specifically, a logical subnet coordinator is elected within the subnet. This coordinator runs a lightweight shortest path algorithm to calculate the optimal path for all node pairs within the subnet and distributes the routing table entries to each node within the subnet. Communication within the subnet bypasses dynamic routing protocols.

[0030] Implementation of dynamic routing between subnets and dynamic nodes:

[0031] Each static backbone subnet is abstracted as a logical supernode. The location of this logical supernode is represented by the subnet's gateway node, i.e., the node connecting the dynamic link. When cross-subnet communication or communication between a static backbone subnet and a mobile node occurs, the propagation scope of the routing request is limited to all mobile nodes and the gateway nodes of each static backbone subnet, reducing the effective node size of the dynamic routing plane from the total number of network nodes to the sum of the number of mobile nodes and the number of static subnets.

[0032] This invention provides the following beneficial effects:

[0033] 1. This invention reconstructs the network into a heterogeneous hierarchical structure that combines a static backbone subnet with a dynamic access edge. This reduces the effective node size of the dynamic routing plane from the total number of network nodes to the sum of the number of mobile nodes and the number of static subnets, thereby significantly reducing the routing control overhead ratio. This fundamentally solves the problem of increased routing overhead and network lag when the scale of traditional Mesh networks expands, ensuring stable network performance in large-scale rescue scenarios with multiple nodes and wide coverage.

[0034] 2. The static backbone subnet is built based on static nodes and stable links. Pre-calculated static routes are used within the subnet to efficiently handle critical data. Dynamic routes are limited to mobile nodes and subnet gateway nodes, reducing invalid route interactions, lowering command latency, ensuring the real-time and accuracy of command and dispatch at the rescue site, and avoiding rescue risks caused by communication interruptions or delays.

[0035] 3. Node status self-identification adopts a low-power nine-axis inertial measurement unit and a lightweight sensor fusion algorithm, which does not require additional complex hardware deployment and is adapted to the power consumption requirements of Mesh networking devices; link screening, subnet formation and routing strategies are all dynamically adjusted based on the on-site node status and link quality, without manual intervention. It can adapt to the complex environment of fire rescue sites with changing terrain, frequent node movement and large fluctuations in link quality, and is flexible in deployment and low in maintenance cost. Attached Figure Description

[0036] Figure 1 This is a flowchart of the dynamic topology optimization method for fire rescue on-site command and control network based on Mesh node status awareness, as described in this invention.

[0037] Figure 2 This is a schematic diagram of the network topology before optimization implemented in this invention;

[0038] Figure 3 This is a schematic diagram of the network topology optimization implemented in this invention. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] This embodiment addresses the actual needs of command and control networks at fire and rescue sites (such as high-rise building fires and large chemical industrial park fires). The on-site Mesh nodes include fixed Mesh base stations mounted on fire command vehicles, Mesh terminals worn by firefighters, and temporarily deployed fixed Mesh relay nodes. The total number of nodes N is 100-200, of which M are mobile nodes (firefighters' personal terminals) of 60-120 and static nodes (command vehicle base stations and temporary relay nodes) of 40-80. It needs to achieve high-definition video transmission, real-time issuance of command instructions, and multi-node collaborative communication, solving the problems of increased routing overhead and communication lag when the scale of traditional Mesh networks expands.

[0041] like Figure 1 As shown, this invention provides a dynamic topology optimization method for fire and rescue on-site command and control network based on Mesh node status awareness, including the following steps:

[0042] Step 1: Mesh Node Status Self-Identification: In each Mesh networking device, such as a fixed base station, a personal terminal, or a relay node, a nine-axis inertial measurement unit is integrated. In this embodiment, the MPU-9250 model is selected. This unit includes a three-axis gyroscope, a three-axis accelerometer, and a three-axis magnetometer. It adopts a low-power mode with an operating current ≤5mA, adapting to the power consumption requirements of personal terminals and relay nodes, and is used to collect motion data such as angular velocity and acceleration of the nodes.

[0043] Each Mesh node runs a lightweight complementary filtering algorithm to fuse the angular velocity and acceleration data acquired by the nine-axis inertial measurement unit, filter out environmental interference, and calculate... t angular velocity vector value at time t ,past =Variance of accelerometer readings within a 5-second window ;

[0044] When the condition is met: within the steady time... The interior angular velocity vector value remains below the angular velocity threshold. And in the past The variance of the accelerometer readings within the window is less than the acceleration variance threshold. .

[0045] Then the Mesh node is determined to be in a static state, and the state flag is set. If not both conditions are met simultaneously, the Mesh node is determined to be in a moving state; status flags ;

[0046] Each Mesh node will display its own status flag. (0 for moving, 1 for stationary) The node ID is encapsulated in the network topology status notification message. The broadcast period is dynamically adjusted according to the on-site network environment to ensure that neighboring nodes can obtain their own status information in a timely manner, while avoiding excessively frequent broadcasts that would increase network overhead.

[0047] Step 2: Construct static through links and form static backbone subnets: Each Mesh node maintains an enhanced neighbor table for topology discovery and link classification. Static through links are filtered based on node status and link quality thresholds. Then, through the distributed connected component discovery algorithm, the undirected graph composed of all static through links is identified as multiple unconnected static backbone subnets.

[0048] Specifically, after each Mesh node starts up, it automatically scans its neighboring nodes, establishes and maintains an enhanced neighbor table, which records the neighbor node ID, received signal strength (RSSI), and neighbor node status. Instantaneous link quality Among them, instantaneous link quality The value of LQ_ij(t) is calculated by weighting RSSI, signal-to-noise ratio (SNR), and packet reception rate (PRR) with weights of 0.4, 0.3, and 0.3 respectively. The value of LQ_ij(t) ranges from 0 to 100, and the larger the value, the better the link quality.

[0049] The preset time window T_verify = 15s, the average link quality threshold LQ_th = 70, and the link quality variance threshold LQ_var_th = 5. For any link L_ij between two Mesh nodes i and j, nodes i and j statistically analyze the link parameters within the 15s time window in real time to determine whether they meet the preset conditions. The preset conditions are that both ends of the link must be stationary, the average link quality must be higher than the preset average link quality threshold, and the link quality variance must be lower than the preset link quality variance threshold. If these conditions are met, nodes i and j upgrade link L_ij to a static direct link through a negotiation mechanism, mark it as a backbone link, and include it in static link management. If the link parameters no longer meet the above conditions in subsequent operation, it will automatically be downgraded to a normal dynamic link to ensure the stability and reliability of the static direct link. The time window T_verify can be adjusted according to the on-site link stability requirements. When the on-site terrain is complex and the link fluctuates greatly, it can be adjusted to 20s.

[0050] All upgraded static direct links form an undirected graph. ,in Let S_i(t) be the set of all static state nodes (S_i(t)=1). This is the set of all statically connected links; each Mesh node runs a flood-based label propagation algorithm, sending its own label information to neighboring nodes via flooding to identify the undirected graph. Connected components in.

[0051] Ultimately, several unconnected connected subgraphs were automatically identified. Each connected subgraph SC_k is a static backbone subnet, and the set of nodes inside the subnet is denoted as V_k; thus achieving zonal coverage of the rescue site.

[0052] Step 3: Adopt a hierarchical heterogeneous routing strategy: For each static backbone subnet, a logical subnet coordinator is elected internally, and the shortest path algorithm is run to construct static routes and realize data forwarding; for communication between subnets and between static backbone subnets and mobile nodes, each static backbone subnet is defined as a logical super node, and the propagation scope of routing requests is limited to all mobile nodes and the gateway nodes of each static backbone subnet, so as to suppress the explosive growth of routing control overhead with the network size and ensure the efficiency, reliability and stability of communication of the fire rescue on-site command and control network.

[0053] Furthermore, the network routing space of the fire command and control network is divided into two independent layers: a static routing plane within subnets and a dynamic routing plane between subnets and dynamic nodes.

[0054] Static routing implementation within a subnet:

[0055] For any static backbone subnet, pre-computed optimized tree or mesh static routing is used internally; specifically, a logical subnet coordinator is elected within the subnet, which runs a lightweight shortest path algorithm to calculate the optimal path for all node pairs within the subnet and distributes the routing table entries to each node within the subnet, and communication within the subnet bypasses dynamic routing protocols for forwarding.

[0056] Implementation of dynamic routing between subnets and dynamic nodes:

[0057] Each static backbone subnet is abstracted as a logical supernode. The location of this logical supernode is represented by the subnet's gateway node, i.e., the node connecting the dynamic link. When cross-subnet communication or communication between a static backbone subnet and a mobile node occurs, the propagation scope of the routing request is limited to all mobile nodes and the gateway nodes of each static backbone subnet. This reduces the effective node size of the dynamic routing plane from the total number of network nodes to the sum of the number of mobile nodes and the number of static subnets, thereby reducing the routing control overhead by orders of magnitude.

[0058] Specifically, the network routing space of the fire command and control network is divided into two independent planes: the static routing plane within the subnet (responsible for communication within the static backbone subnet) and the dynamic routing plane between subnets and dynamic nodes (responsible for communication across subnets and communication between static subnets and mobile nodes). The two planes operate independently and do not interfere with each other, ensuring the orderliness of routing management.

[0059] For each static backbone subnet SC_k, a node election algorithm is used to select the static node with the best link quality and the strongest computing power within the subnet, and a logical subnet coordinator is elected. The coordinator runs a lightweight Dijkstra algorithm, which calculates the optimal path for all node pairs (u,v) (u,v∈ V_k) within the subnet based on the link quality of the nodes within the subnet, generates routing table entries, and distributes them to all nodes within the subnet via broadcast.

[0060] Once the routing table entries are distributed, communication between nodes within the subnet completely bypasses the dynamic routing protocol and directly forwards data according to the pre-calculated static routes, achieving zero routing control overhead and deterministic forwarding to meet the communication needs of the rescue site.

[0061] Each static backbone subnet SC_k is abstracted to the outside as a logical super node. The location of the logical super node is determined by the gateway node of the subnet, that is, the static node with the most links between the subnet and the external nodes is selected. Each logical super node only exposes the communication address of the gateway node to the outside, hiding the details of the internal nodes of the subnet.

[0062] When cross-subnet communication (such as communication between SC_1 subnet and SC_2 subnet) or communication between a static backbone subnet and a mobile node (S_i(t)=0) is required, the propagation scope of the routing request is strictly limited to all mobile nodes or the gateway nodes of each static backbone subnet.

[0063] like Figure 3 The diagram shows the optimized network communication. M1-M9 are Mesh networking nodes. Dashed lines represent dynamic routing links, and solid lines represent static direct links. After optimization, M1, M4, and M7 automatically aggregate into a static subnet A, which contains dynamic links and connects to the static node M5 externally. This fundamentally solves the problem of soaring routing overhead as traditional networks expand. Simultaneously, when mobile nodes access the network, they only need to establish a connection with the nearest static backbone subnet gateway node, enabling rapid access and adapting to scenarios where firefighters move frequently.

[0064] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A dynamic topology optimization method for fire and rescue on-site command and control network based on Mesh node state awareness, characterized in that, Includes the following steps: Step 1, Mesh Node Status Self-Identification: A nine-axis inertial measurement unit is integrated into each Mesh networking device to collect the motion data of the Mesh nodes and determine whether the nodes are stationary or moving through a lightweight sensor fusion algorithm. The status flag is encapsulated into the topology status notification message and the status is broadcast periodically. Step 2: Construct static through links and form static backbone subnets: Each Mesh node maintains an enhanced neighbor table for topology discovery and link classification. Static through links are filtered based on node status and link quality thresholds. Then, through the distributed connected component discovery algorithm, the undirected graph composed of all static through links is identified as multiple unconnected static backbone subnets. Step 3: Adopt a hierarchical heterogeneous routing strategy: For each static backbone subnet, internally, a logical subnet coordinator is elected, the shortest path algorithm is run to construct static routes and implement data forwarding; for inter-subnet communication and communication between static backbone subnets and mobile nodes, each static backbone subnet is defined as a logical super node, and the propagation scope of routing requests is limited to all mobile nodes and the gateway nodes of each static backbone subnet, so as to suppress the explosive growth of routing control overhead with the network size; The location of the logical super node is determined by the gateway node of the subnet. The static node with the most links between the subnet and external nodes is selected. Each logical super node only exposes the communication address of the gateway node to the outside world and hides the internal nodes of the subnet.

2. The method for dynamic topology optimization of a fire rescue on-site command and control network based on Mesh node state awareness as described in claim 1, characterized in that, In step 1, the nine-axis inertial measurement unit includes a three-axis gyroscope, a three-axis accelerometer, and a three-axis magnetometer, which are used to collect motion data of the Mesh nodes.

3. The method for dynamic topology optimization of fire rescue on-site command and control network based on Mesh node state awareness as described in claim 1, characterized in that, The lightweight sensor fusion algorithm described in step 1 is a complementary filtering algorithm, which processes the collected motion data. If a Mesh node satisfies the following conditions: its angular velocity vector value is continuously less than the angular velocity threshold for a certain period of time, and the variance of the accelerometer readings within the past window is less than the acceleration variance threshold, then the Mesh node is determined to be in a stationary state. If these conditions are not met, the Mesh node is determined to be in a moving state.

4. The method for dynamic topology optimization of fire rescue on-site command and control network based on Mesh node state awareness as described in claim 1, characterized in that, The period of status broadcast in step 1 is dynamically adjusted according to the network environment at the fire and rescue site.

5. The method for dynamic topology optimization of fire rescue on-site command and control network based on Mesh node state awareness as described in claim 1, characterized in that, In step 2, the enhanced neighbor table records the neighbor ID, signal strength, neighbor status, and link quality to complete neighbor topology discovery and link classification. For any two Mesh nodes, determine whether they meet the preset conditions within the time window. If they do, the nodes negotiate to upgrade the link to a static pass-through link.

6. The method for dynamic topology optimization of a fire rescue on-site command and control network based on Mesh node state awareness as described in claim 5, characterized in that, The preset conditions must be met simultaneously: Both ends of the link are stationary; The average link quality is higher than a preset threshold for the average link quality. The link quality variance is less than the preset link quality variance threshold.

7. A dynamic topology optimization method for fire and rescue on-site command and control network based on Mesh node state awareness, as described in claim 5, is characterized in that... All static direct links form an undirected graph. A distributed connected component discovery algorithm identifies several unconnected connected subgraphs, each of which is a static backbone subnet.

8. The method for dynamic topology optimization of fire rescue on-site command and control network based on Mesh node state awareness as described in claim 1, characterized in that, Step 3 specifically involves: The network routing space of the fire command and control network is divided into two independent layers: the static routing plane within the subnet and the dynamic routing plane between subnets and dynamic nodes. Static routing implementation within a subnet: For any static backbone subnet, pre-computed optimized tree or mesh static routes are used internally; Specifically, a logical subnet coordinator is elected within the subnet. This coordinator runs a lightweight shortest path algorithm to calculate the optimal path for all node pairs within the subnet and distributes routing table entries to each node within the subnet. Communication within the subnet bypasses dynamic routing protocols for forwarding. Implementation of dynamic routing between subnets and dynamic nodes: Each static backbone subnet is abstracted to the outside as a logical super node. The location of this logical super node is represented by the subnet's gateway node, which is the node connecting the dynamic links. When cross-subnet communication or communication between a static backbone subnet and a mobile node is performed, the propagation scope of the routing request is limited to all mobile nodes and the gateway nodes of each static backbone subnet, reducing the effective node size of the dynamic routing plane from the total number of network nodes to the sum of the number of mobile nodes and the number of static subnets.

Images