A unified management method and device of a multi-protocol internet of things device

Abstract

The application discloses a kind of unified management method and device of multi-protocol internet of things equipment, it is related to internet of things communication and protocol management technical field, including, based on structured gene-semantic mapping table, adopt graph neural network to construct cross-protocol semantic knowledge graph, and generate dynamic semantic relationship matrix by node embedding method;According to dynamic semantic relationship matrix, deploy genetic algorithm engine on edge node to dynamically recombine and optimize protocol gene, through evolutionary sandbox stress test to screen the protocol variant adapted to network environment, and generate optimized protocol gene library;Based on the optimized protocol gene library, in the protocol conversion gateway, combine multi-thread parallel analysis architecture to carry out real-time conversion to heterogeneous protocol message, generate standardized semantic data stream.The application realizes the unified representation and management of complex heterogeneous protocol, simplifies the protocol conversion process, improves cross-protocol semantic understanding and compatibility, improves the flexibility and analysis efficiency of device.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) communication and protocol management technology, and in particular to a unified management method and apparatus for multi-protocol IoT devices. Background Technology

[0002] With the rapid development of IoT technology, more and more devices are connected through networks, enabling data sharing and interaction. However, devices from different manufacturers often follow different communication protocols, making multi-protocol compatibility and interoperability a significant challenge in practical applications.

[0003] While some existing technologies attempt to optimize protocol conversion processes using machine learning algorithms, they typically focus only on improving the efficiency or accuracy of individual steps, neglecting the dynamic adaptability and scalability of the entire system. Particularly in handling real-time conversion of heterogeneous protocol packets, existing solutions fail to adequately consider how to efficiently utilize edge computing resources and how to reduce latency while maintaining service quality. Furthermore, current technologies are significantly inadequate in monitoring abnormal packets and assessing health, failing to provide accurate quantitative analysis of service quality indicators, thus impacting the overall stability and reliability of the network. Summary of the Invention

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

[0005] Therefore, this invention provides a unified management method for multi-protocol IoT devices to address the shortcomings of existing technologies in terms of dynamic adaptability, edge computing resource utilization, and abnormal message monitoring.

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

[0007] In a first aspect, the present invention provides a unified management method for multi-protocol IoT devices, comprising: mapping protocol field structures to artificial gene sequences using a protocol gene encoder based on historical device protocol data, and generating a structured gene-semantic mapping table; constructing a cross-protocol semantic knowledge graph using a graph neural network based on the structured gene-semantic mapping table, and generating a dynamic semantic relationship matrix using a node embedding method; deploying a genetic algorithm engine on edge nodes to dynamically recombine and optimize protocol genes according to the dynamic semantic relationship matrix, screening protocol variants adapted to the network environment through evolutionary sandbox stress testing, and generating an optimized protocol gene library; performing real-time conversion of heterogeneous protocol packets in a protocol conversion gateway using a multi-threaded parallel parsing architecture based on the optimized protocol gene library, generating a standardized semantic data stream; and processing abnormal packets and generating a health assessment report through distributed caching based on the standardized semantic data stream, triggering dynamic version iteration of the protocol library, and achieving unified scheduling of cross-protocol resources through the global configuration interface of the protocol conversion gateway.

[0008] As a preferred embodiment of the unified management method for multi-protocol IoT devices described in this invention, the historical device protocol data includes device performance indicators, protocol structure metadata, protocol semantic ontology library, and protocol interaction logs.

[0009] As a preferred embodiment of the unified management method for multi-protocol IoT devices described in this invention, the specific steps of mapping protocol field structures to artificial gene sequences using a protocol gene encoder based on historical device protocol data and generating a structured gene-semantic mapping table are as follows.

[0010] Based on historical device protocol data, the topological relationships of protocol fields are encoded into gene sequences using a protocol gene encoder, and an artificial gene sequence containing the hierarchical structure of protocol fields and communication interactions is generated.

[0011] By using standardized semantic tags in the protocol semantic ontology library, artificial gene sequences are semantically associated with device functions to generate a structured gene-semantic mapping table.

[0012] As a preferred embodiment of the unified management method for multi-protocol IoT devices described in this invention, the method involves constructing a cross-protocol semantic knowledge graph based on a structured gene-semantic mapping table using a graph neural network, and generating a dynamic semantic relationship matrix through node embedding. The specific steps are as follows.

[0013] Based on the topological relationship features of protocol fields in the structured gene-semantic mapping table, cross-protocol semantic association rules are extracted through graph neural networks, and connection weights between nodes of cross-protocol semantic knowledge graph are generated.

[0014] Based on protocol version compatibility rules and time-series data in protocol interaction logs, the connection weights between nodes in the cross-protocol semantic knowledge graph are dynamically adjusted, and a dynamic semantic relationship matrix is ​​generated through node embedding methods.

[0015] As a preferred embodiment of the unified management method for multi-protocol IoT devices described in this invention, the steps of deploying a genetic algorithm engine at edge nodes to dynamically recombine and optimize protocol genes based on a dynamic semantic relationship matrix, screening protocol variants adapted to the network environment through evolutionary sandbox stress testing, and generating an optimized protocol gene library are as follows:

[0016] Based on the node weight distribution of the dynamic semantic relation matrix, the protocol gene is dynamically recombined through the genetic algorithm engine, and a protocol gene variant population containing mutation probability parameters and crossover probability parameters is generated.

[0017] Based on the protocol gene variant population, a network environment stress test scenario with high packet loss rate and low latency is simulated in an evolutionary sandbox. By using device performance indicators and protocol interaction logs, protocol variants that perform well and have stable interactions in the simulated scenario and are adapted to the network environment are selected.

[0018] Based on the fitness score of protocol variants adapted to the network environment, the protocol gene library version iteration is triggered, and an optimized protocol gene library is generated.

[0019] As a preferred embodiment of the unified management method for multi-protocol IoT devices described in this invention, the method involves: based on an optimized protocol gene library, using a multi-threaded parallel parsing architecture in the protocol conversion gateway to perform real-time conversion of heterogeneous protocol messages and generate standardized semantic data streams. The specific steps are as follows:

[0020] Three protocol parsing threads are initialized based on the optimized protocol gene library, and the features of the received heterogeneous protocol messages are extracted through a multi-threaded parallel architecture to generate raw data packets carrying protocol type identifiers.

[0021] The corresponding semantic mapping table is called according to the protocol type identifier to perform field-level semantic parsing on the original data packet, and redundant fields are removed and data format is unified by predefined protocol field mapping rules to generate intermediate structured data.

[0022] After sorting the intermediate structured data by timestamp, a standardized semantic data stream is generated through the TCP / IP protocol.

[0023] As a preferred embodiment of the unified management method for multi-protocol IoT devices described in this invention, the steps include: processing abnormal messages and generating health assessment reports through distributed caching based on standardized semantic data streams, triggering dynamic version iterations of the protocol library, and achieving unified scheduling of cross-protocol resources through the global configuration interface of the protocol conversion gateway. The specific steps are as follows:

[0024] Based on standardized semantic data streams, abnormal packets are filtered through the packet feature matching mechanism of a distributed cache cluster. The filtered standardized semantic data streams are then subjected to quantitative analysis of service quality indicators to generate a health assessment report that includes network latency and packet loss rate.

[0025] When the broadband utilization rate index in the health assessment report exceeds the broadband warning threshold, the protocol library version is triggered to generate a new protocol version adapted to the current network status, and is synchronously updated to all access nodes through the global configuration interface of the protocol conversion gateway.

[0026] Based on the updated protocol library parameters, unified path scheduling is implemented through heterogeneous protocol resources of MQTT and CoAP to achieve dynamic optimization configuration of transmission paths.

[0027] Secondly, this invention provides a unified management device for multi-protocol IoT devices, comprising a protocol mapping module, a graph generation module, an algorithm optimization module, a protocol conversion module, and a message management module. The protocol mapping module is used to map protocol field structures to artificial gene sequences using a protocol gene encoder based on historical device protocol data, and generate a structured gene-semantic mapping table. The graph generation module is used to construct a cross-protocol semantic knowledge graph using a graph neural network based on the structured gene-semantic mapping table, and generate a dynamic semantic relationship matrix using a node embedding method. The algorithm optimization module is used to deploy a genetic algorithm engine on edge nodes to dynamically recombine and optimize protocol genes according to the dynamic semantic relationship matrix, screen protocol variants adapted to the network environment through evolutionary sandbox stress testing, and generate an optimized protocol gene library. The protocol conversion module is used to perform real-time conversion of heterogeneous protocol messages in a protocol conversion gateway based on the optimized protocol gene library, combining a multi-threaded parallel parsing architecture to generate a standardized semantic data stream. The message management module is used to process abnormal messages through distributed caching and generate a health assessment report based on the standardized semantic data stream, trigger dynamic version iteration of the protocol library, and achieve unified scheduling of cross-protocol resources through the global configuration interface of the protocol conversion gateway.

[0028] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the unified management method for multi-protocol Internet of Things devices as described in the first aspect of the present invention.

[0029] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the unified management method for multi-protocol Internet of Things devices as described in the first aspect of the present invention.

[0030] The beneficial effects of this invention are as follows: By mapping historical device protocol data to artificial gene sequences and generating a structured gene-semantic mapping table, a unified representation and management of complex heterogeneous protocols is achieved, simplifying the protocol conversion process, improving cross-protocol semantic understanding and compatibility, and enhancing device flexibility and parsing efficiency. Furthermore, by utilizing a dynamic semantic relationship matrix, a genetic algorithm engine is deployed at edge nodes to dynamically recombine and optimize protocol genes, and the best protocol variant adapted to the network environment is selected through evolutionary sandbox stress testing, thus achieving automated adjustment of protocol configuration. Attached Figure Description

[0031] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a flowchart of the unified management method for multi-protocol IoT devices in Example 1.

[0033] Figure 2 This is a flowchart of the protocol gene encoding and cross-protocol semantic knowledge graph construction of the unified management method for multi-protocol IoT devices in Example 1.

[0034] Figure 3 This is a flowchart of the genetic algorithm optimization and protocol variant screening of the unified management method for multi-protocol IoT devices in Example 1.

[0035] Figure 4 This is a flowchart of the protocol conversion and health management of the unified management method for multi-protocol IoT devices in Example 1. Detailed Implementation

[0036] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0037] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0038] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0039] Example 1, referring to Figures 1-4 This embodiment provides a unified management method for multi-protocol IoT devices, including the following steps:

[0040] S1: Based on historical device protocol data, the protocol field structure is mapped to artificial gene sequences through a protocol gene encoder, and a structured gene-semantic mapping table is generated.

[0041] S1.1: Historical device protocol data includes device performance indicators, protocol structure metadata, protocol semantic ontology library, and protocol interaction logs.

[0042] It should be noted that historical device protocol data includes device performance metrics, protocol structure metadata, a protocol semantic ontology, and protocol interaction logs. This data provides a comprehensive understanding of the performance of different devices in actual operation and the specific details of their communication protocols. Device performance metrics provide key information about the device's operating status, helping to assess its health and performance level. Protocol structure metadata details the field structure and hierarchical relationships of each protocol, providing a basic framework for subsequent parsing and conversion. The protocol semantic ontology contains standardized semantic tags used to associate artificial gene sequences with device functional semantics, ensuring accurate semantic mapping between different protocols. Protocol interaction logs record the communication processes and interaction behaviors between devices.

[0043] S1.2: Based on historical device protocol data, the topological relationship of the protocol fields is encoded into gene sequences using a protocol gene encoder, and an artificial gene sequence containing the hierarchical structure of the protocol fields and communication interactions is generated.

[0044] The specific process involves first identifying and extracting the hierarchical structure and communication interaction patterns of each protocol field using information from device performance metrics, protocol structure metadata, protocol semantic ontology, and protocol interaction logs. During this process, the protocol gene encoder treats each protocol field as a gene unit and determines its position in the artificial gene sequence based on its location within the protocol and its relationship to other fields.

[0045] Furthermore, the protocol structure metadata provides specific structural information about the protocol fields, enabling the protocol gene encoder to accurately construct artificial gene sequences containing the hierarchical structure of the protocol fields. Simultaneously, the protocol interaction log records the actual interactions between different protocol fields, helping the protocol gene encoder understand and encode the communication and interaction relationships between them. Ultimately, the artificial gene sequence generated based on the above process not only contains the hierarchical structure information of the protocol fields but also covers the complex communication and interaction patterns between fields, thus achieving accurate gene sequence encoding of the topological relationships of the protocol fields. This lays the foundation for generating a structured gene-semantic mapping table in subsequent steps.

[0046] S1.3: Using standardized semantic tags in the protocol semantic ontology library, the artificial gene sequence is semantically associated with the device function to generate a structured gene-semantic mapping table.

[0047] Furthermore, the protocol fields corresponding to each gene unit in the artificial gene sequence are first identified. Then, using standardized semantic tags from the protocol semantic ontology library, corresponding semantic information is assigned to each gene unit, ensuring that each protocol field not only contains its structural information but is also closely related to its functional semantics.

[0048] Specifically, for each gene unit, based on its position in the artificial gene sequence and the protocol field it represents, the most suitable standardized semantic tag is searched and matched from the protocol semantic ontology. This process enables the hierarchical structure of the protocol fields and the communication interaction patterns to be accurately transformed into information fragments with clear semantic meaning. In this way, each gene unit is endowed with clear functional semantics, thereby achieving a precise mapping between the artificial gene sequence and the device's functional semantics. Finally, a structured gene-semantic mapping table is generated based on the precise mapping relationship. This structured gene-semantic mapping table records in detail the association between each gene unit and its corresponding semantic tag.

[0049] S2: Based on the structured gene-semantic mapping table, a cross-protocol semantic knowledge graph is constructed using a graph neural network, and a dynamic semantic relationship matrix is ​​generated through node embedding.

[0050] S2.1: Based on the topological relationship features of protocol fields in the structured gene-semantic mapping table, cross-protocol semantic association rules are extracted through graph neural networks, and connection weights between nodes of cross-protocol semantic knowledge graph are generated.

[0051] The specific process involves extracting cross-protocol semantic association rules using a graph neural network based on the topological relationship features of protocol fields in a structured gene-semantic mapping table. First, the protocol field corresponding to each gene unit is treated as a node, and connections between nodes are established according to the hierarchical structure and communication interaction mode. The graph neural network updates the node feature vectors through a message passing mechanism, thereby determining the similarity and correlation between nodes, and ultimately extracting cross-protocol semantic association rules.

[0052] Furthermore, when generating connection weights between nodes in a cross-protocol semantic knowledge graph, the graph neural network adjusts the connection strength between nodes based on the similarity of node feature vectors, protocol version compatibility rules, and time-series data in protocol interaction logs to ensure that the actual interaction and dependency relationships between protocol fields are reflected.

[0053] S2.2: Based on protocol version compatibility rules and time-series data in protocol interaction logs, dynamically adjust the connection weights between nodes in the cross-protocol semantic knowledge graph, and generate a dynamic semantic relationship matrix through node embedding methods.

[0054] It should be noted that protocol version compatibility rules refer to the criteria for adjusting the connection weights between nodes based on the compatibility between different protocol versions, ensuring interoperability and consistency between different versions. Time-series data refers to time-series information extracted from protocol interaction logs, reflecting the temporal order and frequency of interactions between protocol fields, used to dynamically adjust connection weights to reflect actual interactions. First, the initial connection weights between nodes are corrected according to the protocol version compatibility rules to ensure interoperability and consistency between different protocol versions. Specifically, for each connection weight between cross-protocol semantic knowledge graph nodes, the protocol version compatibility rules are applied to determine the increase or decrease of connection weights based on the protocol fields represented by the two endpoints and their respective protocol versions, reflecting the compatibility of protocol fields between different protocol versions. Next, the connection weights are further refined using time-series data from the protocol interaction logs. By analyzing the actual interaction frequency and temporal order between different protocol fields, the connection weights between nodes with frequent interactions and strong time correlation are strengthened, while the connection weights of nodes with fewer interactions or no interaction are weakened.

[0055] Furthermore, when generating the dynamic semantic relationship matrix, a matrix reflecting the current semantic association strength between all nodes is constructed based on the adjusted connection weights between nodes in the cross-protocol semantic knowledge graph. In this process, the connection weights of each node with other nodes are organized into a matrix, where each element represents the semantic association strength between a pair of nodes. The combined effect of protocol version compatibility rules and time-series data from protocol interaction logs ensures that this matrix not only contains static structural information but also dynamically reflects the changing relationships between protocol fields in the network environment over time. Ultimately, the generated dynamic semantic relationship matrix provides foundational data support for subsequent steps, enabling the optimization of artificial gene sequences and adaptation to the ever-changing network environment requirements.

[0056] S3: Based on the dynamic semantic relationship matrix, deploy a genetic algorithm engine at the edge nodes to dynamically recombine and optimize the protocol genes, screen protocol variants that adapt to the network environment through evolutionary sandbox stress testing, and generate an optimized protocol gene library.

[0057] S3.1: Based on the node weight distribution of the dynamic semantic relation matrix, a genetic algorithm engine is used to perform dynamic recombination on the protocol genes, generating a protocol gene variant population containing mutation probability parameters and crossover probability parameters. The expressions for the mutation probability parameters and crossover probability parameters are as follows:

[0058]

[0059] Where P represents the mutation probability parameter, P1 represents the basic mutation probability, n represents the number of nodes participating in the calculation in the dynamic semantic relation matrix, i represents the node number currently being calculated, and w i γ represents the weight of the i-th node in the dynamic semantic relation matrix. i γ represents the topology sensitivity of the i-th node. max Y represents the maximum value of the topology sensitivity of all nodes, Y represents the crossover probability parameter, Y1 represents the initial crossover probability baseline value, δ represents the exponential function, λ represents the smoothing coefficient (ranging from 0 to 1), W represents the cumulative value of the weights of all nodes, and F represents the average aggregate value of the feature vectors of all nodes.

[0060] The specific process involves first determining mutation probability parameters and crossover probability parameters based on the weight and topology sensitivity of each node in the dynamic semantic relation matrix. For the mutation probability parameter, a base mutation probability (e.g., ranging from 0.01 to 0.2) is used, combined with the number of nodes participating in the calculation in the dynamic semantic relation matrix, the currently calculated node number, and the node's weight and topology sensitivity, to adjust the mutation probability of each node. This ensures that nodes with higher weights and greater topology sensitivity have higher mutation probabilities, promoting the generation of more adaptive protocol variants. Next, for the crossover probability parameter, an initial crossover probability baseline value (e.g., ranging from 0.4 to 0.9) is adjusted using an exponential function, a smoothing coefficient, and the accumulated weights of all nodes.

[0061] Furthermore, based on the adjusted mutation and crossover probability parameters, the genetic algorithm engine performs dynamic recombination on the protocol genes, generating a protocol gene variant population containing various mutation and crossover combinations. Each protocol gene variant reflects a different potential optimization scheme, ensuring that the final generated protocol gene variant population can cover a wide range of possible configurations, thereby providing a rich selection basis for subsequent steps.

[0062] S3.2: Based on the protocol gene variant population, simulate network environment stress test scenarios with high packet loss rate and low latency in the evolutionary sandbox. Through device performance indicators and protocol interaction logs, select protocol variants that perform well and have stable interaction in the simulated scenario and are adapted to the network environment.

[0063] The specific process involves simulating high packet loss rate and low latency network environment stress test scenarios in an evolutionary sandbox based on the protocol variant population. First, each protocol variant in the population is tested in a defined network environment. The evolutionary sandbox evaluates the performance of each variant under these harsh network conditions by simulating different network conditions, such as setting a packet loss rate of 10% (example value) and a latency of 20 milliseconds (example value). During the simulation, each variant runs according to its configuration and records device performance metrics and protocol interaction logs. Device performance metrics include, but are not limited to, throughput, response time, and error rate, reflecting the actual performance of the protocol variant in operation. Protocol interaction logs record in detail the interaction behavior and time sequence between various protocol fields, providing in-depth insights into protocol stability and compatibility.

[0064] Furthermore, when selecting protocol variants adapted to the network environment by combining device performance metrics and protocol interaction logs, a preliminary screening of all protocol variants is first conducted based on device performance metrics. For example, protocol variants with high throughput, low response time, and low error rate are selected. Next, the protocol interaction logs are further analyzed to identify protocol variants that can maintain stable interaction and have good compatibility. In this way, it is ensured that the finally selected protocol variants not only perform well in terms of performance but also demonstrate high reliability and stability in actual interaction. Through the above steps, the most suitable protocol variants for the current network environment can be effectively selected from the protocol variant group, providing a solid foundation for the subsequent generation of an optimized protocol gene library.

[0065] S3.3: Based on the fitness score of protocol variants adapted to the network environment, trigger the iteration of the protocol gene library version and generate an optimized protocol gene library, the expression is:

[0066]

[0067] Where S represents the fitness score of the protocol variant, α represents the throughput weighting coefficient, and T represents the throughput. max Represents maximum throughput, β represents the response time weighting coefficient, and R represents the response time. max τ represents the maximum permissible response time, E represents the error rate weighting coefficient, and E represents the error rate. max Indicates the maximum permissible error rate. C represents the stability weighting coefficient, and C represents the stability score in the protocol interaction log. max This represents the highest stability score in the protocol interaction log.

[0068] The specific process involves evaluating each protocol variant that passes the evolutionary sandbox stress test based on its fitness score adapted to the network environment. In the simulated environment, each variant receives a fitness score based on its performance under network conditions (e.g., throughput, response time, error rate, and stability in protocol interaction logs). This score reflects the overall performance of the variant in the network environment. Next, based on the fitness scores of all variants, the top-scoring variants are selected as candidates to trigger protocol genealogy version iterations. For example, if the selection is set to the top 10% of variants (example value), only those variants with fitness scores in the top 10% will be selected for further processing. This process ensures that only the best-performing variants proceed to subsequent steps.

[0069] Furthermore, after identifying protocol variants with high fitness scores, the optimized protocol genealogy is generated. First, the best characteristics of the selected protocol variants are integrated to form a comprehensive and optimized protocol configuration scheme. Specifically, for each selected protocol variant, its optimal features are extracted and combined with the advantages of other high-fitness protocol variants to ensure that the newly generated protocol genealogy covers a variety of superior characteristics. For example, one protocol variant might excel in network latency, while another variant might have an advantage in network packet loss rate; these advantages will be merged into the new protocol genealogy.

[0070] Next, the integrated features are used to update the protocol genealogy, creating a new version. This process involves encoding newly discovered optimization features into the protocol genealogy and adjusting the corresponding parameter settings to reflect the latest optimization results. For example, if a protocol variant performs well under specific network conditions, its relevant parameter settings will be incorporated into the new protocol genealogy version to ensure optimal performance under similar conditions.

[0071] S4: Based on the optimized protocol gene library, the protocol conversion gateway combines a multi-threaded parallel parsing architecture to perform real-time conversion of heterogeneous protocol messages and generate standardized semantic data streams.

[0072] S4.1: Initialize three protocol parsing threads based on the optimized protocol gene library, and use a multi-threaded parallel architecture to extract features from the received heterogeneous protocol messages to generate raw data packets carrying protocol type identifiers.

[0073] The specific process involves initializing three protocol parsing threads based on the optimized protocol gene library. First, configuration information applicable to different protocol types is extracted from the optimized protocol gene library. Each protocol parsing thread loads the corresponding configuration information according to its protocol type, ensuring correct parsing of heterogeneous protocol messages of a specific type. For example, if there are three main protocol types (such as HTTP, MQTT, and CoAP), an independent protocol parsing thread is allocated for each protocol type and initialized according to the parameters in the optimized protocol gene library.

[0074] Furthermore, when performing feature extraction on received heterogeneous protocol messages using a multi-threaded parallel architecture, each protocol parsing thread is responsible for processing the received message of the corresponding protocol type. Specifically, when a heterogeneous protocol message arrives, its protocol type is quickly identified based on the message header information, and it is assigned to the corresponding protocol parsing thread. For example, if an MQTT protocol message is received, it will be assigned to a dedicated MQTT protocol parsing thread. Within each protocol parsing thread, detailed feature extraction is performed on the assigned message. This includes parsing the message header information, extracting payload data, and identifying key fields in the message. For example, for an HTTP protocol message, the feature extraction process may involve parsing the URL, HTTP method (GET, POST, etc.), header fields (such as Content-Type), and request body content.

[0075] Specifically, after feature extraction, each protocol parsing thread combines the extracted feature information with the original message data and appends a specific protocol type identifier. This protocol type identifier identifies the specific protocol type to which the message belongs, so that subsequent processing steps can take appropriate processing logic based on the protocol type. For example, for a parsed CoAP protocol message, the generated data packet not only contains the original message data and the extracted feature information, but also includes a protocol type identifier indicating "CoAP".

[0076] S4.2: Based on the protocol type identifier, call the corresponding semantic mapping table to perform field-level semantic parsing on the original data packet, and remove redundant fields and unify the data format through predefined protocol field mapping rules to generate intermediate structured data.

[0077] The specific process involves first identifying the protocol type identifier in the original data packet. For example, if the original data packet contains a protocol type identifier indicating "MQTT", the semantic mapping table corresponding to the MQTT protocol is looked up. Each semantic mapping table predefines fields for a specific protocol type and their corresponding semantic tags, ensuring accurate parsing of messages of different protocol types. When performing field-level semantic parsing on the original data packet, each field in the original data packet is parsed one by one based on the information in the invoked semantic mapping table, and mapped to the corresponding semantic tags. For example, when processing MQTT protocol data packets, the parsing process may include identifying and parsing various fields (such as Client Identifier, Topic Name, etc.) of control messages such as CONNECT and PUBLISH, and associating the protocol fields with predefined semantic tags.

[0078] Specifically, the predefined protocol field mapping rules are based on a structured gene-semantic mapping table and historical device protocol data. First, historical device protocol data (including device performance metrics, protocol structure metadata, protocol semantic ontology, and protocol interaction logs) is parsed to obtain field definitions, relationships between fields, and actual usage. Next, a protocol gene encoder maps the protocol field structure to artificial gene sequences and associates them with standardized semantic tags in the protocol semantic ontology, generating a structured gene-semantic mapping table. This table not only contains field structure information but also clarifies the functional semantics of each field. Based on this table, detailed field mapping rules are developed for each protocol type, specifying the role, format, and relationship of each field with other fields. For example, in the MQTT protocol, the ClientIdentifier is used to uniquely identify the client, and in the HTTP protocol, Content-Length should be an integer representing the message body length. Redundant fields are identified and removed. Simultaneously, a unified data format standard is established; for example, all timestamp fields use the ISO8601 format.

[0079] Furthermore, based on the field mapping rules in the semantic mapping table, unnecessary or duplicate fields in the original data packets are identified. For example, when processing an HTTP request message, if certain custom header fields do not affect business logic, they can be considered redundant and removed. Next, the remaining valid fields are standardized according to predefined format requirements to ensure that all fields follow a consistent data format. For example, the representation of timestamps is standardized, or numeric fields are ensured to use the same precision and units. When finally generating intermediate structured data, the processed fields are organized into a structured form, which facilitates subsequent processing and analysis. For example, for a parsed and standardized CoAP protocol message, the generated intermediate structured data not only contains the key fields and semantic information of the original message, but is also organized according to a unified data format, enabling messages of different protocol types to be compared and processed within the same framework.

[0080] S4.3: Based on the intermediate structured data sorted by timestamp, a standardized semantic data stream is generated through the TCP / IP protocol.

[0081] The specific process involves first sorting all records in the intermediate structured data by timestamp to ensure the data is arranged in chronological order; for example, all records with ISO 8601 timestamps are sorted in ascending order. After sorting, each record is converted into a data packet conforming to the TCP / IP protocol specification based on its timestamp and content. Specifically, each record is encapsulated in a TCP segment, with necessary TCP header information such as source port, destination port, and sequence number, as well as IP header information such as source IP address and destination IP address, to ensure correct data transmission.

[0082] Next, the encapsulated data packets are sent sequentially in the ordered order, forming a continuous, standardized semantic data stream. During this process, the reliable transmission mechanism provided by the TCP / IP protocol ensures the orderly arrival and integrity of the data packets. Ultimately, the generated standardized semantic data stream not only contains the semantic information of the original data but also conforms to the standard format of the TCP / IP protocol, enabling seamless exchange of original data between different devices and equipment.

[0083] S5: Based on standardized semantic data streams, it processes abnormal messages and generates health assessment reports through distributed caching, triggers dynamic version iteration of the protocol library, and achieves unified scheduling of cross-protocol resources through the global configuration interface of the protocol conversion gateway.

[0084] S5.1: Based on standardized semantic data streams, abnormal packets are filtered through the packet feature matching mechanism of a distributed cache cluster. The filtered standardized semantic data streams are then subjected to quantitative analysis of service quality indicators to generate a health assessment report that includes network latency and packet loss rate.

[0085] The specific process involves first inputting standardized semantic data streams into a distributed caching cluster. Within the distributed caching cluster, each standardized semantic data stream is matched against predefined message feature patterns. These predefined message feature patterns are determined by analyzing device performance metrics, protocol structure metadata, protocol semantic ontology, and protocol interaction logs from historical device protocol data. For example, based on historical data, characteristic parameters such as the packet size range (e.g., 100 to 1500 bytes), frequency (e.g., 10 packets per second), and time interval (e.g., the time interval between each packet does not exceed 50 milliseconds) for normal messages are set. When a packet in the standardized semantic data stream does not conform to the predefined feature pattern, it is considered an abnormal message and is marked or filtered out. This process ensures that only standardized semantic data streams conforming to the predefined feature patterns can proceed to the next step of processing, thereby effectively eliminating potentially abnormal or malicious messages.

[0086] Furthermore, when performing service quality indicator (SMI) quantitative analysis on the filtered standardized semantic data stream, key SMI indicators, such as network latency and packet loss rate, are analyzed based on the remaining normal standardized semantic data stream. Specifically, for network latency, the timestamp difference between the sender and receiver for each data packet is recorded, along with the average latency value; for example, if a data packet is sent at 10:00:00.000 and received at 10:00:00.123, the latency is 123 milliseconds. For packet loss rate, the difference between the total number of data packets sent and the actual number of data packets received within a certain time period is calculated, and the loss ratio is recorded; for example, if 1000 data packets are sent during the test but only 980 are received, the packet loss rate is 2%. Based on the test results, a health assessment report including network latency and packet loss rate is generated, detailing the specific values ​​of each SMI indicator and the corresponding evaluation criteria.

[0087] Ultimately, by matching the message features of standardized semantic data streams and conducting subsequent quantitative analysis of service quality indicators, we can not only effectively filter out abnormal messages, but also generate detailed health assessment reports, providing solid data support for network performance monitoring and optimization.

[0088] S5.2: When the broadband utilization rate index in the health assessment report exceeds the broadband warning threshold, the protocol library version is triggered to generate a new protocol version adapted to the current network status, and is synchronously updated to all access nodes through the global configuration interface of the protocol conversion gateway.

[0089] The specific process includes the following steps: First, once the broadband utilization rate is detected to exceed a preset broadband warning threshold (e.g., set at 80%), it indicates that the current network load is too high and may affect service quality. At this point, based on the dynamic semantic relationship matrix and the optimized protocol gene library, the protocol library version iteration process is initiated. The existing protocol genes are recombined and optimized using a genetic algorithm engine to generate a new protocol version adapted to the current network state.

[0090] Furthermore, the broadband warning threshold is determined based on equipment performance indicators from historical equipment protocol data and network operation optimization experience. For example, by analyzing past network packet patterns, the maximum bandwidth utilization level the network can withstand without affecting service quality is identified. The specific function of the broadband warning threshold is to provide a clear standard for determining when to adjust or update protocol configurations to cope with changes in network load. For example, if bandwidth utilization remains between 70% and 80% for a long period without causing a decline in service quality, the broadband warning threshold can be set to 80% so that timely measures can be taken when the network load approaches this level.

[0091] Next, when updating all access nodes synchronously via the global configuration interface of the protocol conversion gateway, the generated new protocol version and its related configuration information are first uploaded to the protocol conversion gateway. The protocol conversion gateway then broadcasts the update notification and the details of the new protocol version to all access nodes in the network through its global configuration interface. Upon receiving the update notification, each access node adjusts its protocol parsing and processing logic according to the requirements of the new protocol version. For example, for access nodes using the MQTT protocol, the update process may involve reconfiguring parameters such as connection timeout and heartbeat interval to adapt to the requirements of the new protocol version. All access nodes can complete the synchronous update of the protocol version in a short time, ensuring protocol consistency and efficient operation throughout the network environment.

[0092] S5.3: Based on the updated protocol library parameters, unified path scheduling is implemented through heterogeneous protocol resources of MQTT and CoAP to achieve dynamic optimization configuration of transmission paths.

[0093] The specific process involves first dynamically adjusting the transmission paths for MQTT and CoAP protocols using configuration information from the optimized protocol gene library. During this process, the optimal transmission path is determined based on the characteristics of each protocol and the current network status. For example, for MQTT, considering the subscription model and low bandwidth requirements, a path with lower latency and stability is prioritized; while for CoAP, the focus may be more on optimizing connection establishment speed and packet size.

[0094] Furthermore, by analyzing the service quality indicators (SMIs) such as network latency and packet loss rate included in the updated protocol library parameters, and combining this with real-time network conditions, the most suitable transmission path for each protocol is determined. Next, the routing mechanisms of both MQTT and CoAP protocols are used to forward data packets along the calculated optimal paths. For example, the MQTT broker can adjust its message routing strategy based on the latest protocol library parameters to ensure messages are delivered to the target node in the most efficient way; similarly, the CoAP server can optimize resource discovery and data transmission paths based on the updated parameters. Ultimately, this method achieves dynamic optimization of transmission paths, enabling data transmission under different protocols to achieve optimal performance, while ensuring efficient resource utilization and consistent management throughout the network environment.

[0095] This embodiment also provides a unified management device for multi-protocol IoT devices, including: a protocol mapping module, a graph generation module, an algorithm optimization module, a protocol conversion module, and a message management module; the protocol mapping module is used to map protocol field structures to artificial gene sequences based on historical device protocol data through a protocol gene encoder, and generate a structured gene-semantic mapping table; the semantic graph generation module is used to construct a cross-protocol semantic knowledge graph based on the structured gene-semantic mapping table using a graph neural network, and generate a dynamic semantic relationship matrix through a node embedding method; the genetic algorithm optimization module is used to optimize the dynamic semantic relationship matrix at the edge... The node deploys a genetic algorithm engine to dynamically recombine and optimize protocol genes, screens protocol variants adapted to the network environment through evolutionary sandbox stress testing, and generates an optimized protocol gene library. The protocol conversion and standardization module is used to perform real-time conversion of heterogeneous protocol packets in the protocol conversion gateway based on the optimized protocol gene library, combined with a multi-threaded parallel parsing architecture, to generate standardized semantic data streams. The packet management module is used to process abnormal packets and generate health assessment reports through distributed caching based on standardized semantic data streams, trigger dynamic version iteration of the protocol library, and achieve unified scheduling of cross-protocol resources through the global configuration interface of the protocol conversion gateway.

[0096] This embodiment also provides a computer device applicable to a unified management method for multi-protocol IoT devices, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the unified management method for multi-protocol IoT devices as proposed in the above embodiment.

[0097] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0098] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the unified management method for multi-protocol IoT devices as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0099] In summary, this invention achieves unified representation and management of complex heterogeneous protocols by mapping historical device protocol data to artificial gene sequences and generating a structured gene-semantic mapping table. This simplifies the protocol conversion process, improves cross-protocol semantic understanding and compatibility, and enhances device flexibility and parsing efficiency. Furthermore, by utilizing a dynamic semantic relationship matrix, a genetic algorithm engine is deployed at edge nodes to dynamically recombine and optimize protocol genes. Evolutionary sandbox stress testing is then used to screen for the optimal protocol variant adapted to the network environment, thus achieving automated adjustment of protocol configuration.

[0100] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A unified management method for multi-protocol Internet of Things (IoT) devices, characterized in that: The application comprises the following steps: Based on historical device protocol data, the protocol field structure is mapped to artificial gene sequence by a protocol gene encoder, and a structured gene-semantic mapping table is generated; Based on the structured gene-semantic mapping table, a graph neural network is used to construct a cross-protocol semantic knowledge graph, and a dynamic semantic relationship matrix is generated by a node embedding method; According to the dynamic semantic relationship matrix, a genetic algorithm engine is deployed at the edge node to dynamically recombine and optimize the protocol genes, the protocol variants that adapt to the network environment are screened through evolutionary sandbox stress testing, and an optimized protocol gene library is generated, the specific steps are as follows, Based on the node weight distribution of the dynamic semantic relationship matrix, the protocol genes are executed by the genetic algorithm engine for dynamic recombination, and a protocol gene variant group containing mutation probability parameters and crossover probability parameters is generated; Based on the protocol gene variant group, high packet loss rate and low delay network environment stress test scenarios are simulated in the evolutionary sandbox, and through device performance indicators and protocol interaction logs, protocol variants that adapt to the network environment are screened out in the simulation scenario and interact stably; Based on the fitness score of the protocol variant that adapts to the network environment, trigger the protocol gene library version iteration, and generate the optimized protocol gene library; Based on the optimized protocol gene library, the heterogeneous protocol packets are converted in real time in the protocol conversion gateway combined with the multi-thread parallel parsing architecture, and the standardized semantic data stream is generated; Based on the standardized semantic data stream, abnormal packets are processed through distributed caching and a health assessment report is generated, triggering dynamic version iteration of the protocol library, and realizing unified scheduling of cross-protocol resources through the global configuration interface of the protocol conversion gateway. 2.The method of claim 1, wherein: The historical device protocol data includes device performance indicators, protocol structure metadata, protocol semantic ontology library and protocol interaction logs. 3.The method of claim 2, wherein: The specific steps are as follows, Based on the historical device protocol data, the protocol field topology relationship is encoded by the protocol gene encoder, and an artificial gene sequence containing protocol field hierarchical structure and communication interaction is generated; Through the standardized semantic tags in the protocol semantic ontology library, the artificial gene sequence is semantically associated with the device function to generate a structured gene-semantic mapping table. 4.The method of claim 3, wherein: The specific steps are as follows, Based on the protocol field topology relationship features in the structured gene-semantic mapping table, cross-protocol semantic association rules are extracted by the graph neural network, and the connection weights between the nodes of the cross-protocol semantic knowledge graph are generated; Based on the protocol version compatibility rules and the time sequence data in the protocol interaction log, the connection weights between the nodes of the cross-protocol semantic knowledge graph are dynamically adjusted, and a dynamic semantic relationship matrix is generated by a node embedding method. 5.The method of claim 4, wherein: The specific steps are as follows, Three protocol analysis threads are initialized based on the optimized protocol gene library, and feature extraction is performed on the received heterogeneous protocol packets through a multi-thread parallel architecture to generate raw data packets carrying protocol type identifiers; According to the protocol type identifier, the corresponding semantic mapping table is called to perform field-level semantic analysis on the raw data packet, and through the pre-defined protocol field mapping rule, the redundant fields are removed and the data format is unified to generate intermediate structured data; Based on the intermediate structured data, the standardized semantic data stream is generated through TCP / IP protocol. 6.The method of claim 5, wherein: Based on the standardized semantic data stream, the abnormal packets are processed through the distributed cache and the health evaluation report is generated, triggering the protocol library dynamic version iteration, and through the global configuration interface of the protocol conversion gateway, the cross-protocol resource unified scheduling is realized, the specific steps are as follows, Based on the standardized semantic data stream, the abnormal packets are filtered through the message feature matching mechanism of the distributed cache cluster, and the quality of service index of the filtered standardized semantic data stream is quantitatively analyzed to generate a health evaluation report containing network delay and packet loss rate; When the broadband utilization rate index in the health evaluation report exceeds the broadband warning threshold, the protocol library version iteration is triggered to generate a new protocol version adapted to the current network state, and through the global configuration interface of the protocol conversion gateway, it is updated to all access nodes; Based on the updated protocol library parameters, the unified path scheduling is implemented through the MQTT and CoAP heterogeneous protocol resources, and the dynamic optimization configuration of the transmission path is completed.

7. A device for unified management of multi-protocol Internet of Things, based on the method for unified management of multi-protocol Internet of Things according to any one of claims 1 to 6, characterized in that: It includes a protocol mapping module, a graph generation module, an algorithm optimization module, a protocol conversion module and a message management module; The protocol mapping module is used for mapping the protocol field structure to the artificial gene sequence based on the historical device protocol data through the protocol gene encoder, and generating a structured gene-semantic mapping table; The graph generation module is used for constructing a cross-protocol semantic knowledge graph based on the structured gene-semantic mapping table, and generating a dynamic semantic relationship matrix through a node embedding method; The algorithm optimization module is used for dynamically recombining and optimizing the protocol gene based on the dynamic semantic relationship matrix, and selecting the protocol variant adapted to the network environment through the evolution sandbox stress test, and generating an optimized protocol gene library, the specific steps are as follows, Based on the node weight distribution of the dynamic semantic relationship matrix, the protocol gene is executed through the genetic algorithm engine, and a protocol gene variant group containing mutation probability parameters and crossover probability parameters is generated; Based on the protocol gene variant group, the high packet loss rate and low delay network environment stress test scenario is simulated in the evolution sandbox, and through the device performance index and protocol interaction log, the protocol variant adapted to the network environment is selected which performs well in the simulation scenario and has stable interaction; Based on the fitness score of the protocol variant adapted to the network environment, the protocol gene library version iteration is triggered, and the optimized protocol gene library is generated; The protocol conversion module is used for real-time conversion of heterogeneous protocol packets based on the optimized protocol gene library in the protocol conversion gateway combined with a multi-thread parallel analysis architecture to generate a standardized semantic data stream. A message management module is configured to process abnormal messages through a distributed cache based on standardized semantic data flow, generate a health degree evaluation report, trigger dynamic version iteration of a protocol library, and realize unified scheduling of cross-protocol resources through a global configuration interface of a protocol conversion gateway.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that: The computer program, when executed by the processor, implements the steps of the unified management method of the multi-protocol Internet of Things device according to any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program, when executed by the processor, implements the steps of the unified management method of the multi-protocol Internet of Things device according to any one of claims 1-6.

Images