Dynamic sensing Mesh network equipment upgrading method, medium and terminal

Abstract

The invention is suitable for the technical field of wireless communication, and relates to a dynamic sensing Mesh network equipment upgrading method, a medium and a terminal.The dynamic sensing Mesh network equipment upgrading method comprises the steps that firstly, a boundary router inquires all wireless terminal node equipment lists in a network by broadcasting an inquiry instruction, and a complete network topology structure is constructed; dividing nodes into a plurality of groups according to the network topology based on the hop count and the signal strength threshold; the boundary router only pushes a complete software update package to the seed node of each group, and the seed node locally caches the update package and triggers to send a polling notification to the adjacent node; after receiving the notice, the adjacent node actively pulls the firmware update packet from the seed node, the updated node becomes a new source node, diffusion and update are continued to the next hop node, and if node pulling fails, the update packet is retransmitted from the superior node or the server; and the node sends a state report to the border router after finishing firmware downloading and updating. According to the invention, the upgrading efficiency and reliability of the Mesh network equipment are effectively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communication, and particularly relates to a method, medium and terminal for upgrading Mesh network devices with dynamic perception. Background Art

[0002] In the fields of smart cities, industrial Internet of Things, and advanced metering infrastructure (AMI) of smart meters, wireless Mesh networks are widely used. Such networks consist of a large number of devices (such as smart meters, sensors, controllers), and the devices are interconnected through wireless signals to form a Mesh network, transmitting data like a "net". When firmware needs to be upgraded for devices (such as fixing vulnerabilities or adding new functions), three major challenges are faced: dynamically changing network environment: devices may move, signals may be interfered with, or the external environment may change, resulting in the sudden interruption of an originally stable communication link; complex dependency relationships: when upgrading firmware, some devices need to rely on adjacent devices to transmit upgrade packages (for example, edge devices rely on central nodes), but the fixed dependency relationships preset by traditional methods (similar to fixed routes) cannot adapt to dynamic changes; conflict between efficiency and reliability: it is necessary to quickly complete the upgrade of a large area of devices while avoiding repeated retransmissions due to dependency failures.

[0003] In the existing Mesh network firmware upgrade technologies, the following main defects exist: mismatch between static dependency detection and dynamic network: device dependency relationships are based on preset fixed topologies and cannot respond in real time to fluctuations in network link quality (such as signal attenuation, temporary interference), resulting in frequent failures of the transmission paths of firmware upgrade packages and the need to repeatedly initiate the upgrade process; vulnerability of single-path propagation: relying only on the relay transmission of a single path, when the link is interrupted, the central controller needs to intervene again for distribution, greatly prolonging the upgrade time; risk of cascading failures: without anticipating the dynamic changes in device dependency relationships, downstream nodes enter a waiting state due to the failure of the main path upgrade, resulting in the stagnation of the upgrade or even network fragmentation. The above reasons lead to low efficiency and reliability in upgrading existing Mesh network devices.

[0004] Therefore, how to improve the efficiency and reliability of Mesh network device upgrades is an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a method for upgrading Mesh network devices with dynamic perception to solve the problem of low efficiency and reliability in upgrading Mesh network devices in the prior art; in addition, the present invention also provides a medium and a terminal for upgrading Mesh network devices with dynamic perception.

[0006] To solve the above technical problems, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a method for upgrading Mesh network devices with dynamic perception, including the following steps:

[0008] S10. First, the border router queries the list of all wireless terminal node devices in the network by broadcasting a query instruction. The nodes respond to the query and return their own information and the list of adjacent nodes. According to the information returned by the nodes, a complete network topology structure is constructed by using a dynamic link quality evaluation method and a dependency topology generation method.

[0009] S20. Based on the hop count and signal strength threshold, the nodes are divided into multiple groups according to the network topology. In each group, the border router selects seed nodes, requiring that the seed nodes cover most of the adjacent nodes within the group.

[0010] S30. The border router only pushes the complete software update package to the seed nodes of each group. The seed nodes cache the update package locally and trigger the sending of polling notifications to adjacent nodes.

[0011] S40. After receiving the notification, the adjacent nodes actively pull the firmware update package from the seed nodes. The nodes that complete the update become new source nodes and continue to spread the update to the next-hop nodes. If a node fails to pull, it retransmits the update package from the upper-level node or the server.

[0012] S50. After the node completes the firmware download and update, it sends a status report to the border router. If the node does not respond or the update fails, the border router directly retransmits through unicast. After all the nodes complete the firmware download and update, the border router sends a global synchronization instruction to trigger the nodes to switch to the new version.

[0013] Further, in the step S10, in the dynamic link quality evaluation method, the sampling parameters include the physical layer metrics RSSI, PER, and the network layer metrics ETX and the remaining energy ED of the node. Among them, RSSI and PER are calculated through the PHY layer FCS, ETX is dynamically updated, and the remaining energy ED of the node is estimated by integrating the cell voltage / current.

[0014] Further, the dynamic score model in the dynamic link quality evaluation method is as follows:

[0015] Score = w1·((RSSI + 100) / 30) + w2·(1 / ETX) + w3·ED (∑wi = 1);

[0016] Among them, the signal quality w1 is 0.4, the transmission efficiency w2 is 0.4, and the energy efficiency w3 is 0.2.

[0017] Further, in the dynamic link quality evaluation method, the link status is periodically reported to the BR through DAOs, triggering a global DODAG reconstruction.

[0018] Furthermore, in the dependency topology generation method, the specific steps of DODAG reconstruction are as follows:

[0019] BR broadcasts the new DODAG version number, triggering a network-wide routing update;

[0020] Each RN re-elects a parent node based on the dual-weight criterion;

[0021] Generates a multi-path dependency topology.

[0022] Furthermore, in the dependent topology generation method, the number of path hops is limited to a maximum of 6 hops, and the energy balancing strategy is that if a node is selected as a parent node more than a threshold number of times, its priority is actively downgraded to extend the network life.

[0023] Furthermore, in step S40, a reinforcement learning switching engine mechanism is adopted to trigger the pre-switching backup path through the historical link status and energy consumption decay rate of data packet transmission. The reinforcement learning switching engine specifically includes state space definition, action space definition, reward function design, training and reasoning.

[0024] Furthermore, the state space definition includes the primary path implementation indicators {Score, remaining bandwidth, number of consecutive failures}, the backup path preload status {cache hit rate, channel interference level};

[0025] The action space definition includes action 1: maintain transmission on the primary path, action 2: switch to the backup path, action 3: maintain transmission on the backup path, and action 4: request BR local rerouting;

[0026] The reward function is as follows:

[0027] r = number of successfully transmitted blocks / total number of blocks-λ·switching energy consumption;

[0028] Among them, the penalty term λ = 0.05;

[0029] Training and reasoning include offline training: generating Q tables based on historical network topology data and deploying them to each RN after convergence; online updating: triggering incremental learning when the network environment changes significantly.

[0030] In a second aspect, the present invention further provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0031] In a third aspect, the present invention further provides an electronic terminal, comprising: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal executes the method as described above.

[0032] The method, medium, and terminal for upgrading a dynamically perceiving Mesh network device provided by the present invention have at least the following beneficial effects compared with the prior art:

[0033] The upgrading efficiency and reliability of existing Mesh network devices are relatively low. The process of the present invention is simple and convenient to operate. Through methods such as dynamic evaluation of link quality, generation method of dependency topology, and enhanced learning switching engine mechanism, in a Mesh environment where the network link state changes dynamically, by real-time perceiving and predicting changes in dependency relationships, redundant upgrade paths are constructed to achieve highly reliable and low-latency progressive firmware distribution, improving the upgrading efficiency and reliability of large-scale nodes. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the solution of the present invention, the following will briefly introduce the drawings required for description in the embodiments. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0035] Figure 1 It is a flowchart of a method for upgrading a dynamically perceiving Mesh network device provided by an embodiment of the present invention;

[0036] Figure 2 It is a system architecture diagram of a method for upgrading a dynamically perceiving Mesh network device provided by an embodiment of the present invention;

[0037] Figure 3 It is a flowchart of a method for dynamically evaluating link quality of a method for upgrading a dynamically perceiving Mesh network device provided by an embodiment of the present invention;

[0038] Figure 4 It is a schematic diagram of constructing a multi-path DODAG of a method for upgrading a dynamically perceiving Mesh network device provided by an embodiment of the present invention;

[0039] Figure 5 It is a schematic diagram of a state transition matrix of a Q-Learning decision model of a method for upgrading a dynamically perceiving Mesh network device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] To facilitate understanding of the present invention, the following will describe the present invention more comprehensively with reference to the relevant drawings. The preferred embodiments of the present invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive.

[0041] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this invention belongs. The terms used in the description of the invention herein are for the purpose of describing specific embodiments only and are not intended to limit the invention.

[0042] The present invention provides a method for upgrading a dynamically perceiving Mesh network device, which is applied to the process of upgrading the Mesh network firmware. The method for upgrading a dynamically perceiving Mesh network device includes the following steps:

[0043] S10. First, the border router queries the list of all wireless terminal node devices in the network by broadcasting a query instruction. The nodes respond to the query and return their own information and the list of adjacent nodes. According to the information returned by the nodes, a complete network topology structure is constructed by using a link quality dynamic evaluation method and a dependency topology generation method; S20. Based on the hop count and signal strength threshold, the nodes are divided into multiple groups according to the network topology. In each group, the border router selects seed nodes, and it is required that the seed nodes cover most of the adjacent nodes in the group; S30. The border router only pushes the complete software update package to the seed nodes of each group. The seed nodes cache the update package locally and trigger the sending of a polling notice to the adjacent nodes; S40. After receiving the notice, the adjacent nodes actively pull the firmware update package from the seed nodes. The nodes that complete the update become new source nodes and continue to spread the update to the next-hop nodes. If the node fails to pull, it requests the retransmission of the update package from the upper-level node or the server; S50. After the node completes the firmware download and update, it sends a status report to the border router. If the node does not respond or the update fails, the border router directly retransmits through unicast. After all the nodes complete the firmware download and update, the border router sends a global synchronization instruction to trigger the nodes to switch to the new version.

[0044] The present invention effectively improves the efficiency and reliability of upgrading Mesh network devices.

[0045] In order to enable those skilled in the technical field to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0046] The present invention provides a method for upgrading a dynamically perceiving Mesh network device, which is applied to the process of upgrading the Mesh network firmware. As Figure 2 shown, the system architecture includes a Wi-SUN gateway, a border router (BR), a wireless terminal node (routing node, RN), and an enhanced parent selection module. The Wi-SUN gateway includes a border router and an uplink 4G communication module. Its main function is to communicate with the wireless terminal nodes in the Wi-SUN Mesh network through the border router and perform data interaction with the master station system through the uplink 4G communication module, playing a role of connecting the upper and lower levels.

[0047] The core functions of the border router include global coordination and task distribution: As the central controller for firmware upgrade, it multicasts upgrade instructions through the IPv6 protocol to allocate the initial path policies for each node. Data convergence and relay: Receive firmware package chunks from the server side and forward them to the terminal nodes, and at the same time aggregate the upgrade status feedback to the cloud. Support Wi-SUN and IPv6 gateway protocols, taking into account low power consumption and wide area coverage (the visible distance of point-to-point coverage is more than 1 km). The dynamic topology adaptation generates a global path view through the DODAG structure of the RPL (Routing Protocol for Low-Power and Lossy Networks) protocol.

[0048] The protocol features of the wireless terminal node include RPL route maintenance: Periodically broadcast its own topology information through DIO (DODAG Information Object) messages to construct a directed acyclic graph (DODAG) rooted at the BR. Parent selection optimization: The default RPL parent node selection is based on the principle of minimizing ETX (Expected Transmission Count), and the present invention extends it to a dual-weight criterion (energy consumption + link quality). Enhanced parent selection module: MAC layer integration: Embed a dual-path selection algorithm in the MAC protocol stack to continuously monitor the broadcast signals (such as Beacon frames) of the primary / backup parent nodes. Redundant link binding: Preset the primary path as the traditional RPL parent node, and select a neighbor node with sub-optimal ETX and orthogonal channel hopping as the backup path.

[0049] The enhanced parent selection module includes an input end: RSSI (dBm), LQI (Link Quality Indicator), and channel noise level (Noise Floor) reported by the PHY layer. Processing core: Path scoring engine: Synthesize ETX, ED (Energy Depletion), and link encryption status (such as AES-128 encryption flag) to generate the dual-path priority weights; Handover decision maker: Trigger path handover based on a dynamic threshold (for example, the packet loss rate of the primary path > 15%). Output end: Control the switching of the primary / backup transceiver channels of the radio frequency module.

[0050] Combined with Figures 1 to 5 , in this embodiment, the method for upgrading a dynamically aware Mesh network device includes the following steps:

[0051] S10. First, the border router queries the device list of all wireless terminal nodes in the network by broadcasting a query instruction. The nodes respond to the query and return their own information and the list of adjacent nodes. According to the information returned by the nodes, a complete network topology structure is constructed by using the link quality dynamic evaluation method and the dependency topology generation method.

[0052] Specifically, in this embodiment, the link quality dynamic evaluation (link quality scoring mechanism) is as follows: Figure 3 As shown, including:

[0053] Sampling parameters: Physical layer indicators: RSSI (based on IEEE 802.15.4g received signal strength), PER (Packet Error Rate) calculated through PHY layer FCS (Frame Check Sequence);

[0054] Network layer indicators: ETX (dynamically updated), node remaining energy ED (estimated by cell voltage / current integration).

[0055] Dynamic Scoring Model:

[0056] Score=w1·((RSSI+100) / 30)+w2·(1 / ETX)+w3·ED(∑wi=1);

[0057] Among them, the weight configuration is signal quality w1=0.4, transmission efficiency w2=0.4, and energy efficiency w3=0.2.

[0058] Feedback loop: Periodically report link status to BR through DAOs (Destination Advertisement Objects) to trigger global DODAG reconstruction.

[0059] Further, in this embodiment, the topology generation method (DODAG dynamic multi-path extension) is relied upon as follows Figure 4 As shown, including:

[0060] The specific steps of DODAG reconstruction are as follows:

[0061] BR broadcasts the new DODAG version number (via DIO message), triggering a network-wide routing update;

[0062] Each RN re-elects a parent node based on the dual-weight criteria (primary path score and backup path score);

[0063] Generate a multi-path dependency topology, such as Figure 4 As shown:

[0064] Node A (BR): The primary path points to B, and the backup path points to C;

[0065] Node B: The primary path accesses D, and the backup path crosses the layer to access E;

[0066] Failure recovery: When the link from B to D is interrupted, B takes a detour through the backup path E to F to D.

[0067] The topology optimization constraints include path hop count limit and energy balance strategy. The maximum path hop count limit is 6 hops (to balance delay and energy consumption). If the number of times a node is selected as a parent node exceeds the threshold, its priority is actively downgraded to extend the network lifespan.

[0068] S20: Based on the hop count and signal strength threshold, divide the nodes into multiple groups according to the network topology. Within each group, the border router selects seed nodes, requiring the seed nodes to cover most of the adjacent nodes within the group (preferably select high-stability and high-bandwidth devices).

[0069] S30: The border router only pushes the complete software update package to the seed nodes of each group. The seed nodes cache the update package locally and trigger the sending of polling notifications to adjacent nodes.

[0070] S40: After receiving the notification, adjacent nodes actively pull the firmware update package from the seed nodes. The nodes that complete the update become new source nodes and continue to spread the update to the next-hop nodes. If a node fails to pull, it requests a retransmission of the update package from the upper-level node or the server. During the distributed transmission and upgrade of the package, an enhanced learning switching engine mechanism is adopted, and the pre-switch standby path is triggered based on the historical link state and energy consumption attenuation rate of the data packet transmission.

[0071] Specifically, in this embodiment, the enhanced learning switching engine, as Figure 5 shown, includes:

[0072] State space definition (State, S): The implementation metrics of the main path {Score, remaining bandwidth, consecutive failure count} and the pre-loaded state of the standby path {cache hit rate, channel interference level}.

[0073] Action space definition (Action, A): Action 1: Maintain the main path transmission; Action 2: Switch to the standby path; Action 3: Maintain the standby path transmission; Action 4: Request local re-routing by the BR (in extreme cases).

[0074] Reward function design:

[0075] r = number of successfully transmitted blocks / total number of blocks - λ · switching energy consumption;

[0076] The penalty term λ = 0.05.

[0077] Training and inference: Offline training: Generate the Q-table based on historical network topology data and deploy it to each RN after convergence; Online update: When the network environment changes significantly (such as when new nodes are added), trigger incremental learning.

[0078] After the node completes the firmware download and update, it sends a status report to the border router. If the node does not respond or the update fails, the border router directly retransmits via unicast. After all node firmware downloads and updates are completed, the border router sends a global synchronization instruction to trigger the node to switch to the new version.

[0079] Furthermore, in this embodiment, the specific steps of the dynamic link quality detection process (optimized by combining a smoothing function) in step S10 are as follows:

[0080] Step 1: Original parameter collection: Physical layer: RSSI (Received Signal Strength), SNR (Signal-to-Noise Ratio), PER (Packet Error Rate); Network layer: ETX (Expected Transmission Count), ED (Energy Consumption Level); Security layer: Link encryption identifier. Collection frequency: Fixed period (default 1s) + event trigger (e.g., immediately sample when ETX jump ≥ 20%).

[0081] Step 2: Dynamic index smoothing process to suppress instantaneous fluctuations (such as occasional signal interference) and make the link quality assessment more stable.

[0082] Step 3: Constraint condition injection to filter out unavailable links: Screening conditions: Encryption link identifier = TRUE & ED ≥ 30% (node remaining energy threshold); Eliminate low-energy nodes or non-encrypted links to ensure the path is secure and reliable.

[0083] Dynamic weight adjustment:

[0084] Furthermore, in this embodiment, the hierarchical path decision tree is as follows:

[0085] EMA t = α · current sampling value + (1 - α) · EMA t-1 (α = 0.3, damping factor);

[0086] Input of the reinforcement learning optimization model: Current network state (Score, historical handover times, cache pressure);

[0087] Q-Learning action space: Action 1: Select the top 3 candidate nodes for both the primary and backup paths; Action 2: The primary path is the top 1, and the backup path is the node with the lowest interference across clusters.

[0088] Reward function correction:

[0089] Furthermore, in this embodiment, the dual-path maintenance and handover mechanism is as follows:

[0090] Hot backup mechanism: Primary path transmission:

[0091] Send data chunks and receive ACK / NACK feedback (per-chunk timeout = 100 ms);

[0092] Maintain standby path heartbeat detection (period = 200 ms) and monitor its availability.

[0093] Handover trigger conditions (execute when any of the following is met):

[0094] Three consecutive NACKs or timeouts;

[0095] The main path Score drops below 120% of the standby path Score;

[0096] Node remaining energy alert (ED < 10%).

[0097] Lossless handover design:

[0098] Data synchronization: The standby path pre-caches the next 2 chunks in the background;

[0099] Status inheritance: Carry the chunk sequence number and CRC check value during handover to ensure resume from breakpoint.

[0100] The following takes the example of deploying a smart meter network based on the Wi-SUN protocol in an urban smart grid when facing firmware upgrade and solving the problem of link instability caused by dynamic interference (such as electromagnetic interference from the subway power system and building blockage):

[0101] Hardware composition and material selection: FG25 Wi-SUN module: Supports IEEE 802.15.4g, integrates a mesh network protocol stack, and operates in the frequency band of 902 - 928 MHz (CH26 is the main path, CH40 is the standby).

[0102] Main control unit: STM32H743 MCU, running a lightweight RTOS (FreeRTOS).

[0103] Link quality sensing: Integrate an RSSI dynamic sampling circuit (accuracy of ±0.5 dBm), and output in real time based on the FG25 RF front end.

[0104] Dynamic link quality monitoring: Parameter acquisition: Sample RSSI once every 1 second (calibrated for both hot and cold data channels). Calculate the ETX value every 5 seconds (based on the ACK / NACK rate).

[0105] Dynamic smoothing processing: Use an EMA filter

[0106]

[0107] Dynamic Dependency Topology Generation: DODAG Construction, Parent Node Selection Criteria: RSSI ≥ -82 dBm (meeting the minimum communication requirements of Wi-SUN). ETX ≤ 3 (allowing an average of 1 retransmission).

[0108] Multipath Rule: Primary Path: Based on the highest comprehensive score (Score = 0.6·RSSI + 0.4·(1 / ETX)). Backup Path: Sub-optimal Score and the channel is frequency-hopping orthogonal to the primary path.

[0109] Topology Update Trigger: When the primary path Score drops by ≥ 15%, or when there are 3 consecutive packet losses, dynamic reconstruction is triggered.

[0110] Reinforcement Learning Prediction Model (Q-Learning): State Space Definition (State, S): Real-time metrics of the primary path: {Score, remaining bandwidth, consecutive failure count}; Preloaded state of the backup path: {cache hit rate, channel interference level}.

[0111] Action Space Definition (Action, A): Action 1: Maintain primary path transmission; Action 2: Switch to the backup path; Action 3: Maintain backup path transmission; Action 4: Request BR local rerouting (in extreme cases).

[0112] Reward Function Optimization:

[0113] Model Training: Initial Training Dataset: Historical link data (100,000 sets of samples). Online Update: Incremental learning once every 24 hours.

[0114] Dual-Path Redundant Switching Design: Primary Path Transmission Process: The gateway distributes firmware blocks (block size 512 bytes) through the BR (Border Router).

[0115] The primary path is distributed hop-by-hop to child nodes according to the RPL multicast protocol.

[0116] Backup Path Preloading: During the primary path transmission, the backup path caches the next 2 blocks in the background (occupying 128KB SRAM of STM32H743).

[0117] Switching Trigger Logic: Condition: The primary path has 2 consecutive NACKs or timeouts (threshold = 200 ms). Hardware Response: FG25 triggers dual-channel switching through SPI (response time ≤ 2 ms).

[0118] Specific effects are as follows (actual test data compared with traditional tree-like upgrade data):

[0119]

[0120] From the above test results, it can be seen that through the dynamic citrus Mesh network device upgrade method of the embodiments of the present invention, the reliability and efficiency of online firmware upgrade are improved.

[0121] The embodiments of the present invention also provide a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, any one of the methods in this embodiment is implemented.

[0122] The embodiments of the present invention also provide an electronic terminal, including: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory so that the terminal executes any one of the methods in this embodiment.

[0123] For the computer-readable storage medium in this embodiment, those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to the computer program. The foregoing computer program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above method embodiments; and the foregoing storage medium includes: various media such as ROM, RAM, magnetic disk or optical disk that can store program codes.

[0124] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication with each other. The memory is used to store a computer program, the communication interface is used for communication, and the processor and the transceiver are used to run the computer program so that the electronic terminal executes each step of the above method.

[0125] For the dynamic perception Mesh network device upgrade method, medium and terminal described in the above embodiments, compared with the prior art, the upgrade efficiency and reliability of the existing Mesh network devices are relatively low. The process of the present invention is simple and the operation is convenient. Through methods such as link quality dynamic evaluation method, dependency topology generation method, and reinforcement learning switching engine mechanism, in a Mesh environment where the network link state changes dynamically, by real-time perceiving and predicting the change of dependency relationship, a redundant upgrade path is constructed to achieve high-reliability and low-latency progressive firmware distribution, and improve the update and upgrade efficiency and reliability of large-scale nodes.

[0126] Obviously, the embodiments described above are only the preferred embodiments of the present invention, rather than all embodiments. The preferred embodiments of the present invention are given in the accompanying drawings, but do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosed content of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions recorded in the foregoing specific embodiments, or perform equivalent replacements for some of the technical features. Any equivalent structure made by using the content of the specification and drawings of the present invention, directly or indirectly applied in other related technical fields, is similarly within the scope of the patent protection of the present invention.

Claims

1. A method for upgrading a Mesh network device with dynamic perception, characterized in that The following steps are involved: S10, the border router first queries the list of all wireless terminal node devices in the network by broadcasting query instructions, and the node responds to the query and returns its own information and a list of adjacent nodes. According to the information returned by the node, a dynamic link quality evaluation method and a dependent topology generation method are used to build a complete network topology structure; S20, based on the hop count and signal strength threshold, the nodes are divided into multiple groups according to the network topology, and the border router in each group selects a seed node, which is required to cover most of the adjacent nodes in the group; S30, the border router pushes the complete software update package only to the seed node of each group, the seed node caches the update package locally, and triggers the sending of polling notifications to adjacent nodes; S40, after receiving the notification, the adjacent node actively pulls the firmware update package from the seed node, and the node that completes the update becomes the new source node, and continues to spread the update to the next hop node. If the node fails to pull, the update package is retransmitted from the upper node or server; S50, after the node completes the firmware download and update, it sends a status report to the border router. If the node does not respond or the update fails, the border router directly retransmits via unicast. After all nodes complete the firmware download and update, the border router sends a global synchronization instruction to trigger the node to switch to the new version.

2. The method for upgrading a dynamically perceiving Mesh network device according to claim 1, wherein In the step S10, in the link quality dynamic evaluation method, the sampling parameters include the physical layer indicators RSSI and PER and the network layer indicators ETX and the node residual energy ED, wherein RSSI and PER are calculated by the PHY layer FCS, ETX is dynamically updated, and the node residual capacity ED is estimated by the cell voltage / current integration.

3. The method for upgrading a dynamically perceiving Mesh network device according to claim 2, wherein The dynamic scoring model in the link quality dynamic evaluation method is as follows: Score=w1·((RSSI+100) / 30)+w2·(1 / ETX)+w3·ED(∑wi=1); Among them, the signal quality w1 is 0.4, the transmission efficiency w2 is 0.4, and the energy efficiency w3 is 0.

2.

4. A method for upgrading a dynamically perceiving Mesh network device according to claim 3, characterized in that, In the dynamic link quality assessment method, the link status is periodically reported to the BR through DAOs to trigger the global DODAG reconstruction.

5. The method for upgrading a dynamically sensed Mesh network device according to claim 4, wherein In the dependency topology generation method, the specific steps of DODAG reconstruction are as follows: BR broadcasts the new DODAG version number, triggering a network-wide routing update; Each RN re-elects a parent node based on the dual-weight criterion; Generates a multi-path dependency topology.

6. The method for upgrading a dynamically perceiving Mesh network device according to claim 5, wherein In the dependent topology generation method, the number of path hops is limited to a maximum of 6 hops, and the energy balancing strategy is that if a node is selected as a parent node more than the threshold, its priority is actively downgraded to extend the network life.

7. A method for upgrading a dynamically perceiving Mesh network device according to claim 1, characterized in that, In the step S40, a reinforcement learning switching engine mechanism is adopted to trigger the pre-switching backup path through the historical link status and energy consumption decay rate of data packet transmission. The reinforcement learning switching engine specifically includes state space definition, action space definition, reward function design, training and reasoning.

8. The method for upgrading a dynamically perceiving Mesh network device according to claim 7, wherein, The state space definition includes the implementation indicators of the primary path {Score, remaining bandwidth, number of consecutive failures}, the preload status of the backup path {cache hit rate, channel interference level}; The action space definition includes Action 1: maintaining the primary path transmission, Action 2: switching to the backup path, Action 3: maintaining the backup path transmission, and Action 4: requesting BR local rerouting; The reward function is as follows: r = the number of successfully transmitted blocks / the total number of blocks - λ · switching energy consumption; Among them, the penalty term λ = 0.05; Training and inference include offline training: generating a Q-table based on historical network topology data and deploying it to each RN after convergence; online update: triggering incremental learning when the network environment changes significantly.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any one of claims 1 to 8 is implemented.

10. An electronic terminal, characterized in that, It includes: a processor and a memory; The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory so that the terminal executes the method described in any one of claims 1 to 8.

Images