A multi-modal internet of things communication module and dynamic resource optimization method

Abstract

The application discloses a kind of multimode Internet of Things communication module and dynamic resource optimization method, it is related to communication technical field, receives sensing layer monitoring data, communication link operating state, node operating parameter and energy supply type;QoS classification is carried out to monitoring data, and the priority of each communication link is determined in combination with link state;Dynamic host election is executed based on node parameter, and communication task and energy consumption mode are allocated for node according to energy type;Multi-communication link dynamic switching and transmission scheduling are executed, transit signal monitoring and transmission window reservation are carried out for satellite link, and data transmission is completed in combination with QoS classification.The scheme uses national chip architecture, integrates double energy adaptive strategy and satellite resource protection engine, effectively solves the problems of resource waste and transmission interruption caused by static scheduling, improves the reliability of data transmission and the utilization rate of communication resources, prolongs the endurance of battery nodes, and ensures the autonomous controllability of communication in extreme environment, adapts to low data volume quasi-real-time communication demand.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a multimodal Internet of Things (IoT) communication module and a dynamic resource optimization method. Background Technology

[0002] During cargo ship transportation, in order to achieve real-time monitoring of the status of goods inside containers (such as temperature, humidity, location, etc.), it is necessary to collect monitoring data such as temperature, humidity, and location through IoT modules deployed on the containers, and transmit the data to a remote control platform through multiple communication links such as satellite and LoRa, so as to achieve real-time traceability of the status of goods.

[0003] In existing technologies, IoT communication solutions for container monitoring generally adopt a static resource scheduling mode. Under this mode, there is a lack of collaborative optimization mechanism between data transmission, host node operation, and node energy consumption configuration. Specifically, the collected data is not differentiated in terms of priority, the host nodes are pre-set and cannot be adjusted according to real-time operating parameters, the node communication tasks and energy consumption modes cannot be dynamically configured to adapt to the energy supply type, and the use of satellite communication links lacks a targeted transit window monitoring and reservation mechanism. Ultimately, this results in low data transmission efficiency, serious node energy waste, and low utilization of satellite link resources. Summary of the Invention

[0004] To address the technical problems existing in the prior art, this invention provides a multimodal Internet of Things (IoT) communication module and a dynamic resource optimization method.

[0005] The technical solution adopted in this invention is:

[0006] A multimodal IoT communication module and dynamic resource optimization method, comprising the following steps:

[0007] Step 1: Receive monitoring data sent by the perception layer, obtain the operating status information of each communication link, obtain the operating parameters of each node in the network, and obtain the energy supply type of each node in the network.

[0008] Step 2: Perform QoS classification processing on the monitoring data to generate QoS classification results;

[0009] Step 3: Based on the QoS classification results and communication link operation status information, determine the priority of each communication link;

[0010] Step 4: Based on the network node operating parameters, execute the dynamic host election process to generate host nodes;

[0011] Step 5: Based on the energy supply type and host node, assign communication tasks and corresponding energy consumption modes to each node to obtain the node task and energy consumption configuration scheme.

[0012] Step 6: Based on the priority of each communication link, host node, node task and energy consumption configuration scheme, perform dynamic switching of multiple communication links and data transmission scheduling;

[0013] Step 7: For the satellite communication link, perform overpass signal monitoring and transmission window reservation, and complete data transmission based on QoS classification results.

[0014] Preferably, in step 1, while acquiring the operating parameters of each node in the network, the location information of each node is acquired simultaneously, the relay area is divided based on the number of container stacking layers, and a preset number of nodes are selected from the constantly powered nodes in the network to serve as the regional repeaters for each relay area.

[0015] Preferably, in step 4, the scoring model used in the dynamic host election process is as follows: the host score is calculated based on four dimensions: signal strength, remaining power, task priority, and relay coverage capability, according to preset weight ratios. Among them, relay coverage capability is a newly added scoring dimension, and the weight ratios of signal strength, remaining power, and task priority are adjusted according to the corresponding proportions. Based on the scoring model, the score of each node is calculated, and the node with the highest score is determined as the host node.

[0016] Preferably, during the dynamic switching of multiple communication links and data transmission scheduling in step 6, the following control operations are performed on the underlying nodes in the relay area: the underlying nodes are controlled to perform real-time signal quality monitoring; if the underlying nodes fail to establish a communication connection with the repeater in the same layer area for a preset number of consecutive times, the LoRa communication channel is automatically switched and the communication rate is adjusted to improve the anti-blocking capability.

[0017] Preferably, if the monitoring data collected by the bottom node in a single instance reaches a preset data volume threshold, the corresponding bottom node is controlled to divide the monitoring data into multiple data packets, which are then transmitted to the repeater in the same layer area through different communication channels. The host node receives each data packet and completes the reassembly.

[0018] Preferably, the following operations are also included: monitoring the dynamic offset status of the container stacking position through the accelerometer data of each node; when the position offset of a node reaches a preset threshold, triggering the node to initiate signal strength detection to all repeaters in the same layer; selecting the target repeater with the best signal strength based on the detection results and establishing a new communication connection; updating the repeater coverage relationship corresponding to the node and synchronizing it to the global host and other nodes in the same layer to complete the dynamic replacement of repeater coverage.

[0019] Preferably, the following operations are also included:

[0020] The sensing and monitoring data is divided into time slices according to preset rules to generate independent data slices;

[0021] For each data slice, feature anchors are extracted. These feature anchors contain key traceability information of the data slice. The feature anchors are then distributed and redundantly stored via LoRa relay links.

[0022] When the module's local cache reaches a preset threshold, delete the full-scale perception and monitoring data cached on each node, while retaining the feature anchor points of the distributed storage;

[0023] Once the satellite communication link is restored, all feature anchors are recalled through the global host and spliced ​​together according to timestamps to form a traceability anchor chain.

[0024] Based on the traceability anchor chain, the target data slice corresponding to the anomaly is identified, and the corresponding node is instructed to send back the full data of the target data slice to complete the cargo status traceability.

[0025] Preferably, before performing the deletion of all cached sensing and monitoring data from each node, the following operations are also performed:

[0026] After each feature anchor point completes a preset number of distributed redundant storages, the corresponding storage node sends a storage completion signal back to the global host.

[0027] After receiving completion signals from all storage nodes, the global host generates a unique deletion permission token, which is bound to the integrity check code of the corresponding feature anchor.

[0028] The global host will distribute deletion permission tokens to each node. The node will only perform the deletion operation of the full-scale perception and monitoring data after receiving the deletion permission token bound to its own feature anchor point.

[0029] If a node does not receive a deletion permission token after a momentary power outage is restored, it will retain all sensing and monitoring data until it completes redundant anchor storage and obtains a token again.

[0030] The beneficial effects of the present invention are at least one of the following: by performing QoS classification processing on monitoring data and matching the corresponding communication link priority, it helps to realize differentiated allocation of communication resources, reduce resource contention between critical and non-critical data, and improve the rationality of communication resource utilization; by performing dynamic host election based on the real-time operating parameters of the nodes, the rationality of host node selection can be optimized, the risk of communication interruption caused by host node power depletion or poor signal can be reduced, and the overall operational stability of the communication system can be improved.

[0031] Dynamically allocating communication tasks and energy consumption modes based on the node's energy supply type can adapt to the energy characteristics of different power supply nodes, reduce unnecessary energy consumption of battery-powered nodes, and help extend the battery life of battery-powered nodes; performing overpass signal monitoring and transmission window reservation for satellite communication links can improve the matching degree between data transmission timing and satellite overpass time periods, and optimize the utilization efficiency of satellite link resources. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the method flow of Embodiment 1 of the present invention. Detailed Implementation

[0033] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0034] The multimodal IoT communication module and dynamic resource optimization method described in the various embodiments of the present invention are executed by the main control layer of the multimodal IoT module, supplemented by the collaborative work of each node in the network, the sensing layer sensors, and the communication link module.

[0035] The main control layer is built on a fully domestically produced chip architecture. Its core components include a main control chip, a baseband chip, and a radio frequency chip. It is responsible for core operations such as data reception, command issuance, strategy decision-making, and resource scheduling, and can be integrated into the host nodes within the network. Each node in the network includes normally powered nodes and battery-powered nodes. The nodes have built-in sensor interfaces, multi-mode communication modules, and energy detection units, and can perform operations such as data acquisition, data transmission, and signal monitoring. Some normally powered nodes can be selected as regional repeaters to undertake data relay functions.

[0036] The sensing layer sensors include GPS modules, gyroscopes, accelerometers, temperature and humidity sensors, etc., used to collect various data such as position, motion status, and environment in container monitoring scenarios.

[0037] The communication link module includes a satellite communication module, a LoRa communication module, a cellular communication module (4G / 5G), an Ethernet module, and a WiFi module, which are responsible for enabling data transmission between different nodes and between nodes and remote management platforms.

[0038] Example 1

[0039] A multimodal IoT communication module and dynamic resource optimization method, comprising the following steps:

[0040] Step 1: Receive monitoring data sent by the perception layer, obtain the operating status information of each communication link, obtain the operating parameters of each node in the network, and obtain the energy supply type of each node in the network.

[0041] It should be noted that the perception layer refers to the sensing network composed of various sensors deployed on the container, including GPS modules, gyroscopes, accelerometers, temperature and humidity sensors, etc., which transmit monitoring data such as the container's position, movement status, and surrounding environment; the communication link operation status information refers to the real-time operation data such as connection stability, transmission rate, latency, and signal strength of satellite communication links, LoRa communication links, cellular communication links, Ethernet links, and WiFi communication links; each node in the network refers to the IoT module nodes deployed on different containers, and the nodes are built based on domestically produced chips; the node operation parameters refer to the node's real-time signal strength, remaining power, current task load, chip operating temperature, etc.; the energy supply type refers to the power supply method of the nodes, including battery power, mains power, solar power, etc.; the domestic chip architecture refers to the hardware architecture in which the main control chip uses the domestic ESP32-H2, the baseband chip uses the domestic Unisoc V510, and the RF chip uses the domestic SAW filter from the 24th Research Institute of China Electronics Technology Group Corporation.

[0042] In the specific implementation process, the main control layer receives monitoring data uploaded by various sensors in the perception layer through the UARTTTL protocol. Among them, the GPS module, gyroscope, accelerometer, and temperature and humidity sensor all establish connections with the main control chip ESP32-H2 through this protocol. The link monitoring module collects the operating status information of various communication links such as satellite, LoRa, cellular, Ethernet, and WiFi in real time, focusing on monitoring key parameters such as the signal strength of the satellite link, the transmission rate of the cellular link, and the bit error rate of the LoRa link. Through the node interaction protocol within the network, parameter acquisition instructions are sent to each node, and operating parameters such as signal strength, remaining power, and task load are received from each node. The energy detection unit of each node collects the energy supply type information of each node. The main control layer operates based on a domestic chip architecture, and the SAW filter of the radio frequency chip simultaneously optimizes the anti-interference capability of satellite and LoRa signals, and summarizes and stores all acquired information.

[0043] Step 2: Perform QoS classification processing on the monitoring data to generate QoS classification results.

[0044] It should be noted that QoS classification processing refers to the process of prioritizing monitoring data based on service quality requirements; the QoS classification result refers to the priority level of each monitoring data after classification, specifically divided into 5 levels, where key data (such as location data and alarm data) is marked as QoS=5, non-key data (such as device log data) is marked as QoS=1, and the remaining data are marked as QoS=2, 3, and 4 in order of importance.

[0045] Different monitoring data have different requirements for timeliness and reliability. For example, data on excessive temperature and humidity of goods is directly related to cargo safety and needs to be transmitted with priority, while normal temperature and humidity data has relatively lower timeliness requirements. QoS classification can achieve differentiated allocation of communication resources and improve the reliability of critical data transmission.

[0046] In the specific implementation process, a QoS classification rule is preset to clarify the priority judgment standard corresponding to different types of monitoring data. That is, location data, temperature and humidity exceeding alarm data, and location offset alarm data are QoS=5, normal temperature and humidity data and normal location data are QoS=3, node normal operation status data are QoS=2, and device log data are QoS=1. The monitoring data received in step 1 is matched with the preset classification rule to determine the priority level of each data. The priority levels of all data are sorted into a QoS classification result and associated with the corresponding original monitoring data identifier.

[0047] Step 3: Based on the QoS classification results and communication link operation status information, determine the priority of each communication link.

[0048] It should be noted that the priority of the communication link refers to the transmission priority assigned to the five communication links: satellite, LoRa, cellular, Ethernet, and WiFi. This priority is used to determine the order in which the transmission links for monitoring data of different QoS levels are selected. The default basic priority rule is Ethernet > WiFi > cellular > satellite. The LoRa link, as a relay link in the self-organizing network, is adapted to all-priority data transmission.

[0049] Different communication links have different transmission characteristics. Ethernet links offer stable and high-speed transmission, while satellite links have long transmission distances but are resource-scarce. LoRa links are suitable for short-range, low-power relay transmission. By combining QoS classification results and link operating status to determine link priorities, optimal matching of data and links can be achieved, improving transmission efficiency while avoiding non-critical data consuming satellite resources.

[0050] In the specific implementation process, a preset link priority matching rule is established, which clarifies that critical data with QoS=5 can select the link with the best operating status among all links, while non-critical data with QoS=1-2 prioritizes Ethernet, WiFi, or cellular links to avoid occupying satellite links. Combining the QoS classification results generated in step 2 and the communication link operating status information obtained in step 1, the real-time priority ranking of each communication link is performed. For monitoring data of different QoS levels, the corresponding optimal communication link priority sequence is determined to form the priority allocation result of each communication link.

[0051] For example: The default basic priority is Ethernet > WiFi > Cellular > Satellite, with LoRa as a relay supplement; if the current Ethernet link is stable, temperature and humidity alarm data with QoS=5 will be transmitted via the Ethernet link first; if the Ethernet link is interrupted, it will automatically switch to the WiFi link, and if the WiFi link is abnormal, it will switch to the cellular link. Satellite link transmission of critical data with QoS=5 will only be enabled when all terrestrial links are interrupted; device log data with QoS=1 will only be transmitted via Ethernet or WiFi link and will not occupy satellite resources.

[0052] This achieves precise matching between monitoring data and communication links, clarifies the boundaries of satellite resource usage, improves the rationality of communication resource utilization, ensures priority transmission of high-priority data, and reduces the ineffective occupation of satellite links.

[0053] Step 4: Based on the network node operating parameters, execute the dynamic host election process to generate host nodes.

[0054] It should be noted that the dynamic host election process refers to the process of selecting the core control node in the network based on the real-time operating parameters of the node and according to the preset scoring model. The election cycle is once every hour. The host node refers to the node elected to be responsible for core control functions such as data aggregation, link scheduling, and command issuance in the network. The scoring model refers to the quantitative calculation model of host score = 0.4 × signal strength + 0.3 × remaining power + 0.3 × task priority.

[0055] Static host node configuration is prone to network paralysis due to node failure, power depletion, or poor signal. Dynamically electing host nodes based on real-time operating parameters and a quantitative scoring model, with re-election every hour, ensures stable host node operation, improves overall network reliability, and reduces election failure rate.

[0056] In the specific implementation process, a preset quantitative scoring model is adopted, in which signal strength accounts for 40% of the weight, remaining power accounts for 30%, and task priority accounts for 30%. The operating parameters of each node obtained in step 1 are collected, and the parameter values ​​corresponding to signal strength, remaining power, and task priority are extracted. Signal strength is quantified into a score of 0-100, remaining power is converted into a score of 0-100 based on the actual percentage, and task priority is converted into a score of 0-100 based on the QoS level of the corresponding data. The parameter values ​​of each node are comprehensively calculated based on the scoring model to obtain the comprehensive score of each node. The node with the highest comprehensive score is determined as the host node, the host node election result is generated and synchronized to all nodes in the network, and the state transition process is set as listening, election, hosting, and abdication. A new round of election is automatically triggered when the network fails.

[0057] For example: If a node has a signal strength parameter of 85 points, a remaining battery of 80% (equivalent to 80 points), and is currently undertaking a data transmission task with QoS=5 (task priority equivalent to 90 points), then the node's comprehensive score is 85×0.4+80×0.3+90×0.3=85 points. After comparing the comprehensive scores of all nodes, the node with the highest score is determined as the host node, and the host node identification information is sent to all nodes. The election process is re-executed every hour. If the remaining battery of the original host node drops to 30%, it will be removed from the position in the new round of election due to the lower score.

[0058] By employing a quantitative scoring model and a periodic election mechanism, we ensure that host nodes are in good operational condition, reduce the risk of communication interruptions caused by host node anomalies, improve the overall stability of the network, and significantly reduce the host election failure rate.

[0059] Step 5: Based on the energy supply type and host node, assign communication tasks and corresponding energy consumption modes to each node to obtain the node task and energy consumption configuration scheme.

[0060] It should be noted that communication tasks refer to tasks that nodes need to perform, such as data acquisition, data transmission, and link relay; energy consumption mode refers to the energy consumption mode set by the node to adapt to different task requirements, including high power consumption mode, normal power consumption mode, and low power consumption mode, where the node standby current is ≤5μA in low power consumption mode; node task and energy consumption configuration scheme refers to the configuration file containing the communication tasks and corresponding energy consumption modes assigned to each node; dual energy adaptive protocol refers to the protocol that dynamically allocates tasks and energy consumption modes according to the node's energy supply type, that is, nodes with constant power supply undertake high-frequency communication tasks, while battery-powered nodes enter low power consumption mode.

[0061] Nodes with different energy supply types have varying endurance. Nodes powered by constant power (mains / solar) can handle more energy-intensive tasks, while battery-powered nodes require strict energy consumption control. By allocating tasks and energy consumption modes based on a dual-energy adaptive protocol, energy can be used more efficiently, extending the endurance of battery-powered nodes. Simultaneously, nodes powered by constant power, acting as relay nodes, can improve data transmission coverage.

[0062] In the specific implementation process, the host node, based on the energy supply type of each node obtained in step 1, follows the dual-energy adaptive protocol and presets task allocation preferences and energy consumption mode adaptation rules for nodes with different energy supply types. Nodes with constant power supply are preset to undertake high-energy-consuming tasks such as high-frequency data acquisition, data relay, and large-volume data transmission, and are adapted to the normal power consumption mode. Solar-powered nodes simultaneously enable the MPPT algorithm to improve charging efficiency. Battery-powered nodes are preset to undertake only basic data acquisition and a small number of emergency data transmission tasks, and are adapted to the low-power mode, only waking up when the host is summoned or in case of an emergency. Based on the total number of communication tasks in the network and the operating status of each node, specific communication tasks are assigned to each node in combination with preset rules. For the tasks assigned to each node, the corresponding energy consumption mode is matched. The task allocation results and energy consumption modes of each node are organized into a node task and energy consumption configuration scheme and distributed to each node.

[0063] For example: Mains-powered nodes are assigned data relay tasks and high-frequency data acquisition tasks every 10 seconds, adapting to the normal power consumption mode; solar-powered nodes are assigned data forwarding tasks to battery nodes within the area, enabling the MPPT algorithm to improve charging efficiency, also adapting to the normal power consumption mode; battery-powered nodes are only assigned alarm data acquisition tasks when temperature and humidity are abnormal, adapting to the low power consumption mode, with standby current controlled within 5μA, and are normally in sleep mode, only waking up when temperature and humidity exceed the standard or when a host call signal is received; the host node organizes the above allocation results into a configuration scheme and distributes it to each node.

[0064] The dual-energy adaptive protocol enables precise matching between communication tasks and node energy supply capabilities, reducing unnecessary energy consumption of battery-powered nodes, significantly extending their battery life, and ensuring the smooth execution of high-energy-consuming tasks. The charging efficiency of solar-powered nodes is further optimized.

[0065] Step 6: Based on the priority of each communication link, host node, node task and energy consumption configuration scheme, perform dynamic switching of multiple communication links and data transmission scheduling.

[0066] It should be noted that dynamic switching of multiple communication links refers to the operation of switching data transmission channels between satellite, LoRa, cellular, Ethernet, and WiFi links according to the link operation status and task requirements, with a switching time of ≤500ms; data transmission scheduling refers to the operation of arranging the transmission tasks of each node in a specific time sequence and allocating resources; the LoRa self-organizing network module supports star / mesh hybrid topology, with a maximum number of 16 hops and a coverage radius of 5km in open environments.

[0067] Single communication links are susceptible to environmental factors and may experience anomalies. Dynamic switching between multiple links improves transmission reliability, and optimized switching time ensures continuous data transmission. Reasonable transmission scheduling avoids link congestion caused by multiple nodes transmitting simultaneously, improving transmission efficiency. The topology design of the LoRa self-organizing network module expands transmission coverage and adapts to the multi-node communication needs of container stacking scenarios. By combining link priority, host control, and node configuration schemes to perform switching and scheduling, end-to-end collaborative optimization can be achieved.

[0068] In the specific implementation process, the host node monitors the operating status of each communication link in real time and compares it with the link priority determined in step 3; according to the task arrangement and energy consumption mode of each node in the node task and energy consumption configuration scheme, a transmission timing plan is formulated. Nodes with constant power supply transmit data according to the high-frequency timing, while battery-powered nodes only transmit data when the wake-up condition is triggered; according to the transmission timing plan and link priority, each node is controlled to execute data transmission. The LoRa self-organizing network module forwards data in a star / mesh hybrid topology, supporting a maximum of 16 hops; when the operating status of a link is abnormal, a dynamic switching mechanism is triggered, and data transmission is switched to the next highest priority link within 500ms; the transmission process is monitored in real time, and the scheduling plan is dynamically adjusted according to the transmission situation. The LoRa module optimizes and reduces the bit error rate through preamble detection.

[0069] For example: The host node detects that the Ethernet link is currently operating stably, and controls the normal temperature and humidity data with QoS=3 to be transmitted through the Ethernet link according to the link priority; according to the transmission timing plan, three mains-powered nodes are arranged to transmit relay data in sequence to avoid link congestion caused by simultaneous transmission; after the battery-powered node is woken up, it transmits temperature and humidity alarm data through the LoRa link, and the LoRa module forwards it to the host node through 2 hops according to the mesh topology; if the Ethernet link is suddenly interrupted, the dynamic switching mechanism is immediately triggered, and the data transmission is switched to the WiFi link within 500ms. If the WiFi link is abnormal, it is further switched to the cellular link.

[0070] By achieving efficient utilization of multiple communication links and orderly scheduling of data transmission, the optimization of switching time ensures the continuity of data transmission. The LoRa self-organizing network topology expands the transmission coverage, improves the reliability and efficiency of data transmission, and adapts to the energy consumption mode requirements of each node, thereby reducing overall energy consumption.

[0071] Step 7: For the satellite communication link, perform overpass signal monitoring and transmission window reservation, and complete data transmission based on QoS classification results.

[0072] It should be noted that transit signal monitoring refers to the operation of real-time monitoring of signals in the area where the satellite passes over the cargo ship; transmission window reservation refers to the operation of applying for transmission resources in advance during the satellite transit period and determining the data transmission window; the transmission window refers to the time period suitable for data transmission during the satellite transit; and the satellite resource protection engine refers to a comprehensive optimization engine that integrates data caching, transit signal detection, transmission window reservation, and QoS priority scheduling.

[0073] Satellite communication links are affected by satellite orbits, enabling effective transmission only during satellite transit, and transmission resources are scarce. By using a satellite resource protection engine to monitor transit signals and reserve transmission windows, transmission resources can be locked in advance. Combined with QoS classification results, critical data can be prioritized for transmission, improving satellite link utilization efficiency and the success rate of critical data transmission. At the same time, a data caching mechanism can prevent data backlog within the transit window.

[0074] In the specific implementation process, satellite orbit parameters and a transit prediction model are preset. Based on the real-time position of the cargo ship and the satellite orbit parameters, the transit period of the satellite is predicted with a prediction error of ≤30 seconds. Five minutes before the predicted transit period, the transit signal monitoring module is activated to monitor the satellite signal strength and stability in real time. At the same time, the satellite communication module uses frequency hopping spread spectrum (FHSS) technology to improve anti-interference capability. When the satellite signal is detected to meet the transmission requirements, a transmission window reservation request is sent to the satellite communication system to determine the transmission window. Within the reserved transmission window, based on the QoS classification results in step 2, key data with QoS=5 is transmitted first, and other data are transmitted in sequence according to QoS level. During non-transit periods, the satellite communication module buffers untransmitted data. The link is established in advance before data transmission, and the handshake time is optimized from the usual 2.3 seconds to 0.8 seconds. After the data transmission is completed, the transmission result is fed back to the host node.

[0075] For example: Based on the cargo ship's current location in the open ocean and the orbital parameters of the Tianqi constellation, it is predicted that there will be three satellite transit periods within the next 24 hours, namely 08:00-08:15, 14:30-14:45, and 21:10-21:25, with the prediction error controlled within 30 seconds; Signal monitoring is started 5 minutes before 08:00. After detecting that the satellite signal strength meets the transmission requirements and optimizing the anti-interference capability through FHSS technology, a transmission window of 08:00-08:15 is reserved with the satellite system; Within this transmission window, temperature and humidity alarm data and positioning data with QoS=5 are transmitted first, followed by normal environmental data with QoS=3, and finally equipment log data with QoS=1. The link is established in advance before transmission, and the handshake time is optimized to 0.8 seconds; After the transmission is completed, the host node is informed that all QoS=5 data transmissions were successful and other data transmissions were partially completed.

[0076] The satellite resource protection engine improves the utilization efficiency of satellite link transmission resources, ensuring that high-priority data is transmitted smoothly during satellite transit periods, avoiding data backlog caused by mismatched transmission timing. Frequency hopping spread spectrum technology and handshake time optimization further enhance the anti-interference capability and transmission efficiency of satellite communication.

[0077] Example 2

[0078] Considering that in the actual application of the scheme in Embodiment 1, the signal of the bottom node is blocked due to the stacking of metal containers, making it difficult to transmit data to the host node, and thus causing the data transmission interruption, in order to solve this technical problem, in a possible implementation, in step 1, when obtaining the operating parameters of each node in the network, the location information of each node is obtained simultaneously, the relay area is divided based on the number of stacked container layers, and a preset number of nodes are selected from the constantly powered nodes in the network as the area repeaters of each relay area.

[0079] It should be noted that node location information refers to spatial coordinates such as the number of stacking layers and lateral arrangement of the container where the node is located; relay area refers to the area divided based on the number of container stacking layers, used to realize data relay transmission; continuously powered node refers to node that uses continuous power supply methods such as mains power or solar power, which is different from battery powered node; regional repeater refers to the node selected in each relay area that is responsible for the data relay transmission of the underlying nodes in that area, and realizes data forwarding based on the star / mesh hybrid topology of LoRa self-organizing network module.

[0080] Metal containers significantly obstruct wireless signals, causing severe signal attenuation when lower-level nodes communicate directly with the host node. By dividing the network into relay zones and setting up regional repeaters, the topological advantages of LoRa ad hoc networks can be leveraged to achieve tiered relay transmission of data from lower-level nodes, avoiding signal attenuation caused by metal obstruction and improving data transmission success rate.

[0081] In the specific implementation process, while obtaining the node operating parameters in step 1, the location information of each node, such as the number of container stacking layers and the lateral arrangement position, is obtained by combining the node's GPS module and accelerometer. The relay area is divided based on the number of stacking layers. Usually, each layer is an independent relay area. If there are many containers on a certain layer, it can be further divided into sub-areas according to the lateral arrangement. The number of normally powered nodes in the network is counted. Based on the location information of each normally powered node and the LoRa signal coverage range (≤5km in open environment), a preset number of normally powered nodes are selected as regional repeaters for each relay area. The relay area division results and regional repeater identification information are synchronized to all nodes and host nodes. The regional repeaters undertake the data forwarding task of the underlying nodes in the area according to the LoRa self-organizing network topology.

[0082] For example: Containers on a cargo ship are stacked in 4 layers and 5 rows. Based on the number of stacking layers, they are divided into 4 relay areas, corresponding to layers 1-4 respectively. There are 8 normally powered nodes in the network, of which 2 are located on layer 1, 3 on layer 2, 2 on layer 3, and 1 on layer 4. Two normally powered nodes are selected as regional repeaters for each relay area (1 is selected for layer 4 due to the limitation on the number of normally powered nodes). Nodes A and B on layer 1 are determined as repeaters for that layer, nodes C and D on layer 2 are as repeaters for that layer, and so on. The regional repeaters rely on the mesh topology of the LoRa self-organizing network to receive the transmitted data from the battery-powered nodes on that layer and forward it to the host node, and synchronize the information of each regional repeater to all nodes.

[0083] By dividing relay areas and setting up regional repeaters, a hierarchical communication transmission network was constructed. Relying on the advantages of LoRa self-organizing network topology, the signal obstruction caused by stacked metal containers was avoided, improving the success rate and coverage of data transmission at the lower-level nodes.

[0084] Example 3

[0085] Considering the technical problem that the dynamic host election in the second embodiment does not take into account the coverage requirements of the regional repeaters in actual application, resulting in the elected host node being unable to effectively cover all relay areas and thus affecting the relay transmission efficiency, in order to solve this technical problem, in a possible implementation, the scoring model adopted in step 4 of the dynamic host election process is as follows: the host score is calculated by four dimensions: signal strength, remaining power, task priority, and relay coverage capability, according to a preset weight ratio. Among them, relay coverage capability is a newly added scoring dimension, and the weight ratios of signal strength, remaining power, and task priority are adjusted according to the corresponding proportions. Based on the scoring model, the score of each node is calculated, and the node with the highest score is determined as the host node.

[0086] It should be noted that relay coverage capability refers to the signal coverage range and coverage strength of a node in each relay area, characterizing the node's ability to effectively receive data transmitted by repeaters in each relay area. When quantifying, it takes into account the coverage radius and signal attenuation of the LoRa self-organizing network. The preset weight ratio refers to the pre-set ratio coefficients for the four scoring dimensions used to calculate the comprehensive score.

[0087] Example 2 sets up regional repeaters, requiring host nodes to have strong relay coverage capabilities to efficiently receive relay data from each relay region. Adding a relay coverage capability dimension to the host election scoring model ensures that the elected host nodes can adapt to the relay transmission architecture, improving the transmission efficiency of relay data and avoiding data transmission delays or losses due to insufficient coverage.

[0088] In the specific implementation process, the weighting of four scoring dimensions is preset, and the respective weighting coefficients of signal strength, remaining power, task priority, and relay coverage capability are clearly defined. For each node, the parameter values ​​of the four dimensions are obtained. The relay coverage capability parameter value is calculated by the node based on the signal coverage range and strength of each relay area, and quantified by combining the 5km coverage radius of the LoRa self-organizing network and the signal attenuation coefficient in a metallic environment. The comprehensive score of each node is calculated based on the scoring model. The calculation formula of the scoring model is: S=a×S1+b×S2+c×S3+d×S4, where S is the comprehensive score of the node, a is the weighting of signal strength, S1 is the signal strength parameter value, b is the weighting of remaining power, S2 is the remaining power parameter value, c is the weighting of task priority, S3 is the task priority parameter value, d is the weighting of relay coverage capability, and S4 is the relay coverage capability parameter value. The comprehensive scores of all nodes are compared, and the node with the highest score is determined as the host node.

[0089] For example: With preset weight percentages of a=35%, b=25%, c=20%, and d=20%, and a node's S1=85, S2=90, S3=80, and S4=88 (quantitative score of relay coverage capability, indicating that it can effectively cover all 4 relay areas), then the node's comprehensive score S=35%×85+25%×90+20%×80+20%×88=85.6 points; after comparing the comprehensive scores of all nodes, the node with the highest score is determined as the host node. This node has good coverage capability for each relay area and can efficiently receive data forwarded by the repeaters in each area.

[0090] By adding a new relay coverage capability scoring dimension, we ensure that host nodes have good relay coverage capabilities, improve the overall collaborative efficiency of the relay transmission network, and make the solution more suitable for the needs of hierarchical relay transmission.

[0091] Example 4

[0092] Considering the technical problem that in the actual application of the scheme in Embodiment 3, when the bottom node in the relay area communicates with the repeater in the same layer area, the signal may still be interrupted due to metal obstruction or environmental interference, and there is no emergency communication adjustment mechanism, in one possible implementation, during the dynamic switching of multiple communication links and data transmission scheduling in step 6, the following control operations are performed on the bottom node in the relay area: control the bottom node to perform real-time signal quality monitoring, and if the bottom node fails to establish a communication connection with the repeater in the same layer area for a preset number of consecutive times, automatically switch the LoRa communication channel and adjust the communication rate to improve the anti-obstruction capability.

[0093] It should be noted that signal quality monitoring refers to the operation of real-time monitoring of parameters such as signal strength, signal-to-noise ratio, and bit error rate of the LoRa communication link between the underlying node and the regional repeater; preset number of attempts refers to the number of consecutive attempts to determine if the communication connection has failed; LoRa communication channel refers to the wireless communication channel based on LoRa technology, and different channels correspond to different frequencies; communication rate refers to the data transmission rate, which is usually negatively correlated with the signal coverage; preamble detection optimization refers to the technical means of reducing the bit error rate by optimizing the preamble detection algorithm of the LoRa module.

[0094] Even with regional repeaters in place, signal interruptions between the underlying nodes and the repeaters can still occur due to metal obstructions or environmental interference. By monitoring signal quality in real time and dynamically switching channels and adjusting rates when a connection fails, interference from specific channels can be avoided. Combined with preamble detection optimization, this further enhances the communication link's resistance to obstruction and its stability, while reducing the bit error rate.

[0095] In the specific implementation process, during the transmission scheduling in step 6, the host node controls the bottom-level nodes in each relay area to start the signal quality monitoring module, which monitors parameters such as signal strength, signal-to-noise ratio, and bit error rate of the LoRa communication link with the repeater in the same layer area in real time; a preset threshold for the number of communication connection attempts is set. When the number of consecutive attempts by the bottom-level node to establish a communication connection reaches the preset threshold and all attempts fail, a channel switching and rate adjustment mechanism is triggered; the bottom-level node automatically scans the currently available LoRa communication channels and selects the channel with the least signal interference as the new communication channel; at the same time, based on the current signal environment, the communication rate is reduced to improve signal coverage and anti-interference capability; after the switching is completed, the communication connection with the repeater in the same layer area is re-attempted, and the switching result is fed back to the host node; the LoRa module simultaneously enables preamble detection optimization to further reduce the bit error rate.

[0096] For example: The preset threshold for the number of communication connection attempts is 3. If a bottom-level node fails to establish a communication connection with the repeater in the same area for 3 consecutive times, the adjustment mechanism is triggered. The node scans and finds that the current LoRa channel 1 has high interference and a high bit error rate, while channel 5 has less interference, so it switches to channel 5. At the same time, the communication rate is adjusted from the original 1200bps to 600bps to improve the anti-blocking capability. After the switch is completed, the connection is re-attempted and communication with the repeater in the area is successfully established. The LoRa module optimizes through preamble detection and controls the bit error rate to within 0.3%.

[0097] This embodiment enables dynamic adjustment of the communication link between the underlying node and the regional repeater, effectively avoiding communication interruptions caused by signal interference and metal obstruction. Combined with preamble detection optimization, it further reduces the bit error rate and improves the stability and anti-interference capability of the relay transmission link.

[0098] Example 5

[0099] Considering that in the actual application of the scheme in Embodiment 4, there is still a technical problem that when the amount of monitoring data collected by the bottom node in a single instance is too large, it is easy to cause transmission failure or occupy too much communication resources, affecting the transmission of other data. In one possible implementation, if the monitoring data collected by the bottom node in a single instance reaches a preset data volume threshold, the corresponding bottom node is controlled to divide the monitoring data into multiple data packets and transmit them to the repeater in the same layer area through different communication channels. The host node receives each data packet and completes the reassembly.

[0100] It should be noted that the preset data volume threshold refers to the upper limit of the single data collection volume that is pre-set and determined to require data fragmentation; data fragmentation refers to the operation of dividing a large amount of monitoring data into multiple small data blocks; data packets refer to the small data blocks formed after fragmentation, and each data packet contains information such as data identifier, fragment number, and data content; data reassembly refers to the operation of the host node to splice and integrate multiple received data packets according to the fragment number to restore the original complete data.

[0101] Transmitting large amounts of data in a single transmission can easily lead to timeouts and failures, and consumes a significant amount of communication resources, affecting the transmission of other data. By transmitting data in segments, the amount of data transmitted in a single transmission can be reduced, improving the success rate of transmission. At the same time, the communication resources are distributed to reduce the impact on the transmission of other priority data, which is especially suitable for the low data volume transmission characteristics of LoRa links.

[0102] In the specific implementation process, a preset threshold for the amount of data collected in a single instance is established. During the transmission scheduling process in step 6, the host node monitors the amount of monitoring data collected in a single instance by each underlying node in real time. When the amount of data collected in a single instance by a certain underlying node reaches the preset threshold, a data fragmentation transmission instruction is sent to that node. After receiving the instruction, the underlying node divides the monitoring data into multiple data packets according to the preset fragment size, and adds a unique data identifier and fragment sequence number to each data packet. Through different LoRa communication channels optimized in step 4, the multiple data packets are transmitted sequentially to the repeater in the same layer area, and then the repeater in the area aggregates them and transmits them to the host node. After receiving all the data packets, the host node sorts and splices the data packets according to the data identifier and fragment sequence number to complete the data reassembly and restore the original complete data.

[0103] For example: The preset threshold for the amount of data collected in a single session is 1024 bytes. The amount of historical temperature and humidity data and abnormal event records collected in a single session by a certain bottom node is 2560 bytes, reaching the threshold. The host node issues a fragmentation instruction, and the node divides the data into 5 small data packets with a fragment size of 512 bytes, respectively labeled with the data identifier Data-001 and fragment sequence numbers 1-5. The 5 small data packets are transmitted sequentially to the repeater in the same layer area through two different LoRa channels, No. 5 and No. 6. After receiving all the small packets, the host node splices them together according to the sequence numbers 1-5 to restore the original data of 2560 bytes. During the entire transmission process, the bit error rate of the LoRa module is controlled within 0.3%.

[0104] This embodiment reduces the risk of failure in a single transmission of large amounts of data, and improves the success rate of large data transmission; at the same time, it disperses the occupation of communication resources, reduces the impact on the transmission of other priority data, adapts to the transmission characteristics of LoRa links, and improves the overall transmission efficiency.

[0105] Example 6

[0106] Considering the technical problem that in actual application of the above embodiments, the turbulence and waves during cargo ship navigation can cause dynamic positional shifts in containers, leading to the failure of the originally covered relay links, forming dynamic blind spots in relay coverage, and consequently causing data transmission interruptions in some lower-level nodes, one possible implementation also includes the following operations: monitoring the dynamic shift status of container stacking positions through accelerometer data from each node; when the positional shift of a node reaches a preset threshold, triggering the node to initiate signal strength detection to all repeaters in the same layer; selecting the target repeater with the best signal strength based on the detection results and establishing a new communication connection; updating the relay coverage relationship corresponding to the node and synchronizing it to the global host and other nodes in the same layer to complete dynamic relay coverage replacement.

[0107] It should be noted that accelerometer data refers to the acceleration data collected by the node's built-in accelerometer, which characterizes the container's motion state; dynamic offset state refers to the dynamic change in the container's position caused by factors such as ship turbulence and wave impact; position offset refers to the deviation distance between the node's current position and its initial position; signal strength detection refers to the operation of the node sending detection signals to all repeaters in the same layer and obtaining the signal strength data fed back by each repeater; target repeater refers to the repeater in the area with the best signal strength selected after detection; and repeater coverage relationship refers to the correspondence between the bottom-level node and the repeater in the area responsible for covering it.

[0108] Examples 2 through 5 assume a fixed container stacking position and do not consider dynamic offsets during navigation. Container offsets can cause the original relay coverage relationship to fail, creating dynamic blind spots. By monitoring the offset status with accelerometers and dynamically adjusting the relay coverage relationship based on the detection results, dynamic blind spots can be eliminated in real time, ensuring the continuity of data transmission and adapting to dynamic scenarios during ocean voyages.

[0109] In the specific implementation process, each node activates its built-in accelerometer to collect acceleration data in real time, calculates the container's position offset based on the acceleration data, and monitors the dynamic offset status. A preset position offset threshold is set. When the position offset calculated by a node reaches the preset threshold, a signal strength detection process is triggered. The node sends a signal detection command to all repeaters in the same layer and receives the signal strength data fed back by each repeater. The fed-back signal strength data is sorted, and the repeater in the area with the best signal strength is selected as the target repeater. The communication connection with the original repeater in the area is disconnected, and a new communication connection is established with the target repeater. The host node updates the repeater coverage relationship corresponding to the node and synchronizes the new coverage relationship to all nodes in the same layer and other relevant nodes to complete dynamic replacement.

[0110] For example: The preset position offset threshold is 10 cm. A certain bottom-level node calculates a position offset of 12 cm based on acceleration data, which reaches the threshold. The node sends a detection command to two repeaters in the same layer and receives a signal strength of 75 dBm from repeater A and 82 dBm from repeater B. Repeater B is selected as the target repeater. The connection with the original repeater A is disconnected, and a new connection is established with repeater B. The host node updates the repeater coverage relationship of the node to node X-repeater B and synchronizes it to all nodes in the same layer to ensure that subsequent data is transmitted through the new repeater link.

[0111] This embodiment uses accelerometers to monitor the dynamic offset status of containers in real time. By dynamically adjusting the relay coverage relationship, it quickly eliminates the dynamic blind spots in relay coverage caused by offset, ensuring the continuity and stability of data transmission at the underlying nodes, making the solution more suitable for dynamic monitoring scenarios of ocean voyages.

[0112] Example 7

[0113] Considering that the solution in Implementation Example 1 still has a technical problem in practical application: when multiple consecutive satellite transit windows are interrupted due to weather or other factors, the large amount of historical monitoring data cached by the module can easily exceed the cache capacity, causing early abnormal data to be overwritten and making it impossible to trace the time point of cargo spoilage, a possible implementation also includes the following operations:

[0114] The sensing and monitoring data is divided into time slices according to preset rules to generate independent data slices. Feature anchors are extracted for each data slice, and the feature anchors contain key traceability information of the data slice. The feature anchors are distributed and redundantly stored through LoRa relay links. When the local cache of the module reaches a preset threshold, the full amount of sensing and monitoring data cached by each node is deleted, and the distributed storage of feature anchors is retained. When the satellite communication link is restored, all feature anchors are recalled through the global host and spliced ​​according to timestamps to form a traceability anchor chain. Based on the traceability anchor chain, the target data slice corresponding to the anomaly is identified, and the corresponding node is instructed to send back the full amount of data of the target data slice to complete the cargo status traceability.

[0115] It should be noted that: preset rules refer to pre-defined time slice division rules, usually fixed time intervals; time slices refer to independent data segments formed by dividing continuous monitoring data according to time intervals; feature anchors refer to key data extracted from each time slice that can characterize the core information of that slice; key traceability information refers to core data used to trace the status of goods, including slice timestamps, key values ​​of environmental parameters, QoS levels, data integrity check codes, etc.; distributed redundant storage refers to a storage method that stores feature anchors on multiple different nodes to achieve data redundancy backup; preset thresholds refer to pre-defined capacity thresholds for determining whether the module cache is full; traceability anchor chain refers to a complete set of time series anchors formed by splicing all feature anchors in timestamp order; target data slices refer to time slices containing abnormal information identified in the traceability anchor chain.

[0116] Conventional solutions address data overflow by increasing cache capacity. This embodiment forgoes storing the entire dataset, storing only extremely small feature anchors. The extremely small anchor size avoids cache overflow, and the anchor chain allows for precise location of abnormal data slices. Full data is then transmitted back as needed, resolving the overflow issue while ensuring traceability. Furthermore, distributed storage via LoRa relay links enhances the security of anchor data.

[0117] In the specific implementation process, a preset time-slicing rule is used to divide the sensing and monitoring data into multiple independent data slices at fixed time intervals. For each data slice, feature anchors containing slice timestamps, key environmental parameter values ​​(maximum, minimum, and average), QoS level, and data integrity check codes are extracted. The host node controls each node to store each feature anchor on multiple different nodes via LoRa relay links, achieving distributed redundant storage and ensuring that the failure of a single node does not affect the anchor data. The local cache capacity of each node is monitored in real time, and when the cache capacity reaches a preset threshold, the control of each node is activated. The cached full-scale sensing and monitoring data is deleted, retaining only the feature anchors in the distributed storage. Once the satellite communication link is restored, the host node sends an anchor recall command to all nodes, aggregating all feature anchors. The feature anchors are then assembled in timestamp order to form a complete traceability anchor chain. The traceability anchor chain is analyzed to identify feature anchors containing QoS=5 abnormal alarm information, and their corresponding target data slices are determined. A return command is sent to the node storing the target data slice, instructing it to return the full data of the target data slice. After receiving the full data, the host node completes the traceability of the cargo's deterioration time node and other statuses.

[0118] For example: The preset time slicing rule is one slice every 10 minutes, dividing continuous temperature and humidity monitoring data into multiple 10-minute data slices; extracting the timestamp, maximum / minimum / average temperature and humidity values, QoS level, and checksum of each slice as feature anchors; storing each anchor point to 3 different nodes in the same layer to achieve distributed redundant storage; when the cache capacity of a node reaches 90% (preset threshold), deleting all temperature and humidity data cached by that node, and retaining only the anchor points; if the satellite link is interrupted for 3 transit windows due to weather reasons, after restoration, the host node recalls all anchor points and splices them according to the timestamp to form an anchor point chain, and finds that the anchor point of the slice from 10:00 to 10:10 is marked as QoS=5 (temperature and humidity exceed the standard), instructing the corresponding node to send back the full data of that slice, and tracing back to find that the goods showed signs of deterioration at 10:05.

[0119] This embodiment solves the module cache overflow problem, preventing early abnormal data from being overwritten; it achieves accurate positioning of abnormal data slices through feature anchor chain, and the on-demand back transmission of full data saves satellite link bandwidth and ensures the accuracy of cargo status traceability; distributed redundant storage improves the security of anchor data and avoids traceability failure due to single node failure.

[0120] Example 8

[0121] Considering that the solution in Example 7 may have technical problems in practical applications, such as the instantaneous power outage of the cargo ship's power supply node, which may lead to the deletion of all data of the feature anchor points before the preset number of distributed redundant storage is completed, resulting in missing anchor points and affecting the accuracy of traceability, in one possible implementation, before performing the deletion of all the cached sensing and monitoring data of each node, the following operations are also performed:

[0122] After each feature anchor point completes a preset number of distributed redundant storage operations, the corresponding storage node sends a storage completion signal to the global host. Upon receiving completion signals from all storage nodes, the global host generates a unique deletion permission token, which is bound to the integrity verification code of the corresponding feature anchor point. The global host distributes the deletion permission token to each node. Nodes only perform the deletion operation on all sensing and monitoring data after receiving the deletion permission token bound to their own feature anchor point. After a momentary energy outage is restored, nodes that have not received a deletion permission token retain all sensing and monitoring data until the anchor point redundant storage is completed again and a token is obtained.

[0123] It should be noted that the preset quantity refers to the number of redundant storage nodes set in advance to ensure the security of anchor point storage; the storage completion signal refers to the confirmation signal sent by the storage node to the host node after successfully storing the feature anchor point; the deletion permission token refers to the authorization credential generated by the host node that allows the node to delete all data; the integrity check code refers to the code used to verify the integrity of the feature anchor point data, which is usually generated by a hash algorithm; and the power interruption refers to the brief power outage caused by the constant power supply node due to bumps, poor contact, or short-term power supply failure.

[0124] Example 7 does not guarantee the atomicity of anchor point storage; momentary power outages may lead to incomplete anchor point storage. By setting a deletion permission token mechanism, the completion of anchor point storage is strongly bound to the deletion of all data. Only when the anchor point completes preset redundant storage and obtains a token can all data be deleted, thus ensuring the integrity of anchor point storage from a mechanism perspective and avoiding anchor point loss due to momentary power outages.

[0125] In the specific implementation process, the number of distributed redundant storage nodes for each feature anchor is preset. After each feature anchor is stored on a node, the node verifies the integrity of the anchor data. If the data is complete, it sends a storage completion signal to the global host, carrying the anchor's integrity verification code. After receiving storage completion signals from a preset number of different nodes for the same feature anchor, the global host confirms that the anchor has completed redundant storage. Based on the anchor's integrity verification code, a unique deletion permission token is generated, and the token is bound to the verification code. The global host distributes the deletion permission token to all nodes. After receiving the token, each node verifies whether the verification code bound to the token matches the verification code of its own stored feature anchor. If they match, the node performs a deletion operation on all sensing and monitoring data. If a power outage occurs and power is restored, the node checks whether it has received the deletion permission token for the corresponding anchor. If not, it retains all data and sends an anchor replenishment request to the host node. The host node instructs the relevant nodes to re-complete the redundant storage of the anchor. After storage is completed and a token is generated, the node performs the deletion operation.

[0126] For example: Each feature anchor point is pre-stored to 3 nodes, and the integrity check code of a certain feature anchor point is "MD5-123456". After the anchor point is successfully stored to node A, node B, and node C, the three nodes send storage completion signals to the host respectively. After receiving the three signals, the host generates a deletion permission token bound to MD5-123456 and distributes it to all nodes. Nodes A, B, and C verify that the token matches the anchor point check code stored in them and perform full data deletion. If node D does not receive the token due to a momentary power outage, after power is restored, it retains all the data, requests the host to restore the anchor point, obtains the token after restoring it to 3 nodes, and then performs full data deletion.

[0127] This embodiment achieves a strong binding between feature anchor storage and full data deletion through a deletion permission token mechanism, ensuring the atomicity and integrity of anchor storage; it effectively avoids the anchor loss problem caused by power outages, ensures the integrity of the traceability anchor chain and the accuracy of the traceability results, and further improves the reliability of the solution.

[0128] The embodiments described above are merely illustrative of specific implementations of the present invention, and while the descriptions are detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A multimodal Internet of Things (IoT) communication module and a dynamic resource optimization method, characterized in that, Includes the following steps: Step 1: Receive monitoring data sent by the perception layer, obtain the operating status information of each communication link, obtain the operating parameters of each node in the network, and obtain the energy supply type of each node in the network. Step 2: Perform QoS classification processing on the monitoring data to generate QoS classification results; Step 3: Based on the QoS classification results and communication link operation status information, determine the priority of each communication link; Step 4: Based on the network node operating parameters, execute the dynamic host election process to generate host nodes; Step 5: Based on the energy supply type and host node, assign communication tasks and corresponding energy consumption modes to each node to obtain the node task and energy consumption configuration scheme. Step 6: Based on the priority of each communication link, host node, node task and energy consumption configuration scheme, perform dynamic switching of multiple communication links and data transmission scheduling; Step 7: For the satellite communication link, perform overpass signal monitoring and transmission window reservation, and complete data transmission based on QoS classification results.

2. The multimodal IoT communication module and dynamic resource optimization method according to claim 1, characterized in that, In step 1, while obtaining the operating parameters of each node in the network, the location information of each node is obtained simultaneously. The relay area is divided based on the number of container stacking layers, and a preset number of nodes are selected from the constantly powered nodes in the network to serve as the regional repeaters of each relay area.

3. The multimodal IoT communication module and dynamic resource optimization method according to claim 2, characterized in that, In step 4, the scoring model used in the dynamic host election process is as follows: the host score is calculated based on four dimensions: signal strength, remaining power, task priority, and relay coverage capability, according to preset weight ratios. Among them, relay coverage capability is a newly added scoring dimension. The weight ratios of signal strength, remaining power, and task priority are adjusted according to the corresponding proportions. Based on the scoring model, the score of each node is calculated, and the node with the highest score is determined as the host node.

4. The multimodal IoT communication module and dynamic resource optimization method according to claim 2 or 3, characterized in that, During the dynamic switching of multiple communication links and data transmission scheduling in step 6, the following control operations are performed on the underlying nodes in the relay area: the underlying nodes are controlled to perform real-time signal quality monitoring. If the underlying node fails to establish a communication connection with the repeater in the same layer area for a preset number of consecutive times, the LoRa communication channel is automatically switched and the communication rate is adjusted to improve the anti-blocking capability.

5. The multimodal IoT communication module and dynamic resource optimization method according to claim 4, characterized in that, If the monitoring data collected by the bottom node in a single instance reaches the preset data volume threshold, the corresponding bottom node will be controlled to divide the monitoring data into multiple data packets and transmit them to the repeater in the same layer area through different communication channels. The host node will then receive each data packet and reassemble it.

6. The multimodal IoT communication module and dynamic resource optimization method according to claim 5, characterized in that, It also includes the following operations: monitoring the dynamic offset of the container stacking position through accelerometer data of each node; when the position offset of a node reaches a preset threshold, triggering the node to initiate signal strength detection to all repeaters in the same layer; Based on the detection results, the target repeater with the best signal strength is selected, a new communication connection is established, the relay coverage relationship corresponding to the node is updated, and synchronized to the global host and other nodes in the same layer to complete the dynamic replacement of relay coverage.

7. The multimodal IoT communication module and dynamic resource optimization method according to claim 1, characterized in that, This also includes the following operations: The sensing and monitoring data is divided into time slices according to preset rules to generate independent data slices; Extract feature anchor points for each data slice. The feature anchor points contain key traceability information of the data slices; each feature anchor point is stored in a distributed redundant manner through a LoRa relay link. When the module's local cache reaches a preset threshold, delete the full-scale perception and monitoring data cached on each node, while retaining the feature anchor points of the distributed storage; Once the satellite communication link is restored, all feature anchors are recalled through the global host and spliced ​​together according to timestamps to form a traceability anchor chain. Based on the traceability anchor chain, the target data slice corresponding to the anomaly is identified, and the corresponding node is instructed to send back the full data of the target data slice to complete the cargo status traceability.

8. The multimodal IoT communication module and dynamic resource optimization method according to claim 7, characterized in that, Before performing the deletion of all cached sensing and monitoring data from each node, the following operations are also performed: After each feature anchor point completes a preset number of distributed redundant storages, the corresponding storage node sends a storage completion signal back to the global host. After receiving completion signals from all storage nodes, the global host generates a unique deletion permission token, which is bound to the integrity check code of the corresponding feature anchor. The global host will distribute deletion permission tokens to each node. The node will only perform the deletion operation of the full-scale perception and monitoring data after receiving the deletion permission token bound to its own feature anchor point. If a node does not receive a deletion permission token after a momentary power outage is restored, it will retain all sensing and monitoring data until it completes redundant anchor storage and obtains a token again.

Images