Internet of Things protocol management scheme supporting hot plug extension

Abstract

The invention provides an Internet of Things protocol management scheme supporting hot plug extension, and relates to the technical field of Internet of Things, the scheme comprises a protocol processing module hot plug unit, an intelligent protocol identification unit and a distributed protocol management unit, the protocol processing module hot plug unit encapsulates an Internet of Things protocol into an independent plug-in, and the intelligent protocol identification unit is connected with the distributed protocol management unit; dynamic loading and unloading during operation are supported, hot plug extension is realized, an intelligent protocol identification unit adopts a layered acquisition strategy, deep learning and a traditional statistical feature extraction method are combined, a protocol is accurately identified, a distributed protocol management unit comprehensively considers a geographic position, network delay and node load, an optimal agent node is selected, and the intelligent protocol identification unit is used for identifying the protocol. And when a new protocol or equipment is accessed, the whole system does not need to be modified on a large scale, and only corresponding protocol plug-ins need to be added, so that the modification cost of the system is reduced, the development period is shortened, the system can quickly adapt to the continuously changing environment of the Internet of Things, and the flexibility and maintainability of the system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to an Internet of Things protocol management solution supporting hot plug and play expansion. Background Art

[0002] With the rapid development of IoT technology, more and more devices are connected to the network. These devices may use different communication protocols, such as MQTT, CoAP, HTTP, etc. Traditional IoT protocol management methods are often designed for specific protocols, lacking flexibility and scalability. When new protocols or devices need to be connected, the entire system often needs to be modified on a large scale, which not only increases development costs, but also reduces system stability and reliability. In addition, the scale and complexity of IoT systems are constantly increasing, which also puts higher requirements on the efficiency and performance of protocol management. On the one hand, traditional systems have difficulty supporting the simultaneous access and management of multiple IoT protocols, and the protocol compatibility is poor, so devices with different protocols cannot achieve efficient and smooth communication. On the other hand, when new protocols or devices need to be added, large-scale modifications to the entire system are inevitable due to the rigidity of the system architecture, which not only greatly increases the cost of system modification, but also lengthens the development cycle. Therefore, a IoT protocol management scheme supporting hot-swap expansion is proposed. Summary of the invention

[0003] In view of this, the present invention provides an Internet of Things protocol management solution that supports hot-plug expansion to solve or alleviate the technical problems existing in the prior art and at least provide a beneficial option.

[0004] The technical solution of the present invention is implemented as follows: an IoT protocol management solution supporting hot-plug expansion, the management solution comprising a protocol processing module hot-plug unit, an intelligent protocol identification unit, a distributed protocol management unit, a software-defined network traffic scheduling unit and a blockchain protocol security management unit; The protocol processing module hot-swap unit, intelligent protocol identification unit, distributed protocol management unit, software-defined network traffic scheduling unit and blockchain protocol security management unit cooperate with each other, wherein: The hot-swap unit of the protocol processing module provides plug-in support for device communication, encapsulates the IoT protocol as an independent plug-in, and supports dynamic loading and unloading at runtime to achieve hot-swap expansion; The intelligent protocol identification unit selects a suitable plug-in according to the device communication data, uses machine learning technology to analyze the device communication data in real time, identifies the protocol type and selects a suitable plug-in for communication; The distributed protocol management unit is responsible for selecting device access proxy nodes, constructing a distributed proxy node network, and connecting to appropriate nodes according to geographical location and network topology when devices access; The software-defined network traffic scheduling unit dynamically adjusts protocol traffic paths and policies. With the help of SDN technology, it collects information, formulates policies through the controller, and distributes them to switches; The blockchain protocol security management unit ensures the secure storage and access control of protocol information. Using blockchain characteristics, it constructs a blockchain network to store protocol information and realizes protocol access control, authorization, and settlement through smart contracts.

[0005] Further preferably, the protocol processing module hot-swap unit includes a data receiving module, a data processing module, and a data sending module; The data receiving module is used to receive communication data from IoT devices. It includes a data cache for temporary storage during peak data reception. The cache size is dynamically adjusted according to device processing capabilities and network traffic, and a basic cache size is set in advance according to device processing capabilities , and the network traffic is monitored in real-time , when the network traffic exceeds the preset high traffic threshold , the cache size is increased according to the proportionality coefficient , that is, the cache size , where , , when the cache is full, the first-in, first-out (FIFO) strategy is used for data replacement; The data receiving module is also involved in data transmission mode, encryption method, and error retransmission. Among them, the data transmission mode is selected as synchronous transmission or asynchronous transmission according to device requirements. When asynchronous transmission is selected, the data transmission efficiency is expressed by the formula:

[0006] where is the total amount of transmitted data, is the total transmission time, is the transmission error rate; the encryption method uses the AES symmetric encryption algorithm, and the encryption strength is set according to security requirements. In the error retransmission, the number of retransmissions is set according to network conditions and device reliability, and the value range is 1 - 5 times; The data processing module has dedicated parsing and encapsulation logic libraries for different protocol types to identify and process various IoT protocol data; The data sending module includes data compression, and it selects whether to compress the data before sending according to network conditions and device requirements. The compression ratio is expressed by the formula:

[0007] in is the original data volume, The amount of data after compression.

[0008] Further preferably, the intelligent protocol identification unit includes a data acquisition module, a feature extraction module and a machine learning model training module; The data acquisition module uses a hierarchical acquisition strategy to collect the communication data of the device; the collected data is divided into Layer, each layer of data The acquisition frequency It is set according to the importance and change frequency of the data, and the collection frequency of all layers of data meets is the preset total acquisition frequency; in is the total acquisition frequency; The feature extraction module uses a combined feature extraction algorithm, which includes a convolutional neural network algorithm based on deep learning and a traditional statistical feature extraction method; Among them, the traditional statistical feature extraction method is based on the collected data. Calculate the mean of the data according to the formula:

[0009] Calculate the variance of the data according to the formula:

[0010] in, is the number of data samples, For the data samples; The machine learning model training module uses cross-validation and regularization during the training process. The cross-validation uses The fold cross validation method divides the data set into Each time you select a subset The remaining 1 subset is used for validation and repeated times the process; The regularization adopts L2 regularization, and the regularization term is introduced into the loss function. ; in, is the regularization parameter, is the model weight.

[0011] Further preferably, the distributed protocol management unit includes a routing algorithm module, an agent node communication module and an agent node monitoring module; The routing algorithm module is used to select a suitable proxy node for the device. , obtain the geographical location distance from the device to the node , network latency and node load , calculate the comprehensive evaluation index according to the formula :

[0012] Among them, , , are weight coefficients and satisfy + + = 1; , , respectively represent the maximum values of geographical location distance, network latency, and node load among all proxy nodes. According to the calculation results, select the proxy node with the smallest value as the target node for the device to connect; The node communication security module is used to ensure the security of information interaction and collaborative work between proxy nodes. The AES symmetric encryption algorithm is used to encrypt communication data, and the key length is set to 128 bits or 256 bits. The encryption process is as follows: the sender uses the key to encrypt the plaintext data to generate ciphertext, and the receiver uses the same key to decrypt the ciphertext to obtain the plaintext. At the same time, digital certificate authentication is used for identity verification; The node status monitoring module is used for the management center to comprehensively monitor and manage proxy nodes. The proxy nodes report their own operating status and the protocol traffic information processed to the management center according to a preset period , The value range of is 1 - 60 minutes.

[0013] Further preferably, the software-defined network traffic scheduling unit includes a controller module and a traffic scheduling policy module; The controller module is responsible for collecting network information, and its collection period can be dynamically adjusted according to the actual operation of the network. When the network is busy, the controller shortens the collection period, and when the network is idle, the controller extends the collection period.

[0014] The traffic scheduling policy module is used to dynamically adjust the traffic scheduling policy according to various factors, perform traffic scheduling based on device priority and network congestion status, and allocate different bandwidth resources for protocol traffic with different real-time requirements.

[0015] Further preferably, the blockchain protocol security management unit includes a blockchain network module and a smart contract module; The blockchain network module adopts a consortium blockchain architecture and has multiple verification nodes. The verification nodes verify and record transactions through a consensus algorithm. The smart contract module is responsible for executing smart contracts, and the execution results are recorded on the blockchain in real time.

[0016] Further preferably, the hot-pluggable unit of the protocol processing module includes a plug-in management interface module and a plug-in inspection module. The plug-in management interface module provides a unified interface for the loading and unloading operations of plug-ins. The plug-in inspection module checks the integrity and compatibility of the plug-ins during the plug-in loading process.

[0017] Further preferably, the intelligent protocol recognition unit includes a machine learning model module and a learning monitoring module. The machine learning model module periodically performs incremental learning on newly collected data to update the parameters and structure of the model. The learning monitoring module monitors and records the learning process of the machine learning model.

[0018] Further preferably, the distributed protocol management unit includes a proxy node monitoring module, a failover module, and a fault recording module. The proxy node monitoring module is used to monitor the running status of proxy nodes in real time. When a fault is detected in a proxy node, the failover module immediately sends a fault notification and a task transfer request to other proxy nodes. After receiving the request, other proxy nodes will select a suitable node to receive the transferred task according to their own load conditions and processing capabilities. The fault recording module records the time, cause, and handling results of the fault occurrence.

[0019] Due to the adoption of the above technical solutions in the embodiments of the present invention, the following advantages are achieved: 1. By adopting the hot-pluggable unit of the protocol processing module, the present invention encapsulates the IoT protocol into an independent plug-in, supports dynamic loading and unloading during runtime. Thus, when a new protocol or device is connected, there is no need to make large-scale modifications to the entire system. Instead, only the corresponding protocol plug-in needs to be added, reducing the system modification cost, shortening the development cycle, enabling the system to quickly adapt to the ever-changing IoT environment, and enhancing the flexibility and maintainability of the system.

[0020] Second, the present invention uses a hierarchical acquisition strategy through an intelligent protocol recognition unit, combines deep learning and traditional statistical feature extraction methods to quickly identify different protocols, and selects the optimal proxy node through a distributed protocol management unit considering geographical location, network latency, and node load to ensure communication quality. At the same time, data encryption and authentication are adopted between proxy nodes to ensure communication security and guarantee the stable operation of the Internet of Things system.

[0021] The above summary is for the purpose of the specification only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features of the present invention will become apparent by reference to the drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0023] Figure 1 It is a module architecture diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] In the following, only some exemplary embodiments are briefly described. As those skilled in the art can recognize, the described embodiments can be modified in various different ways without departing from the spirit or scope of the present invention. Therefore, the drawings and the description are considered to be exemplary in nature rather than restrictive.

[0025] The embodiments of the present invention will be described in detail below with reference to the drawings.

[0026] As Figure 1 shown, the embodiment of the present invention provides an Internet of Things protocol management solution supporting hot pluggable expansion. The management solution includes a protocol processing module hot plug unit, an intelligent protocol recognition unit, a distributed protocol management unit, a software-defined network traffic scheduling unit, and a blockchain protocol security management unit; The protocol processing module hot plug unit, the intelligent protocol recognition unit, the distributed protocol management unit, the software-defined network traffic scheduling unit, and the blockchain protocol security management unit cooperate with each other. Among them, The hot-pluggable unit of the protocol processing module provides plug-in support for device communication, encapsulates the IoT protocol into independent plug-ins, supports dynamic loading and unloading during runtime to achieve hot-pluggable expansion. Encapsulating the IoT protocol into independent plug-ins and supporting dynamic loading and unloading during runtime enable the system to easily handle the access requirements of different IoT devices and protocols. When new IoT devices or protocols emerge, there is no need to make large-scale modifications to the entire system. Instead, only the corresponding plug-ins need to be developed and loaded into the system, improving the flexibility and scalability of the system. Moreover, the plug-in design reduces the coupling degree between protocol processing modules. The processing logics of different protocols are independent of each other, and the modification or update of one plug-in will not affect the normal operation of other plug-ins, thus enhancing the stability and reliability of the system; The hot-pluggable unit of the protocol processing module includes a data reception module, a data processing module, and a data transmission module; The data reception module is used to receive communication data from IoT devices. It includes a data buffer for temporary storage during peak data reception. The buffer size is dynamically adjusted according to the device processing capacity and network traffic. The basic buffer size is set in advance according to the device processing capacity , and monitors the network traffic in real time , when the network traffic exceeds the preset high-traffic threshold , the buffer size is increased according to the proportionality coefficient , that is, the buffer size , where , , and when the buffer is full, the first-in-first-out (FIFO) strategy is adopted for data replacement; The data reception module is also involved in data transmission mode, encryption method, and error retransmission. Among them, the data transmission mode is selected as synchronous transmission or asynchronous transmission according to device requirements. When asynchronous transmission is selected, the data transmission efficiency is expressed by the formula:

[0027] where is the total amount of transmitted data, which represents the total amount of data transmitted through the communication link within a certain period. For example, in a file transfer task, the size of the file is the total amount of transmitted data, and the unit can be bytes; is the total transmission time, that is, the time interval from the start of data transmission to the completion of transmission. For example, if it takes 10 seconds to transfer a 100MB file, then the total transmission time is 10 seconds; The transmission error rate refers to the ratio of the number of error - occurred data packets or bits to the total number of transmitted data packets or bits during data transmission. For example, when transmitting 1000 data packets and 10 of them are in error, then the transmission error rate

[0028] Among them,

[0029] This part calculates the data transmission rate, that is, the amount of data transmitted per unit time, which reflects the basic data - transmission ability. The higher the rate, the more data can be transmitted in the same time; This part is about the impact of transmission errors on data - transmission efficiency. Since transmission errors can lead to data re - transmission or partial data loss, the actual effective data - transmission amount is reduced. It represents the proportion of effective data transmission; for example, when At this time, , only 99% of the data is effectively transmitted.

[0030] The encryption method uses the AES symmetric - encryption algorithm, and the encryption strength is set according to security requirements, ensuring the security of data during transmission. In error re - transmission, the number of re - transmission times is set according to the network condition and device reliability, and the value range is 1 - 5 times, which can effectively improve the reliability of data transmission and reduce data loss or damage caused by transmission errors; There is a dedicated parsing and encapsulation logic library for different protocol types, which can accurately identify and process various IoT protocol data, enabling the system to uniformly manage and process IoT data in different formats, improving the accuracy and efficiency of data processing.

[0031] The data - sending module includes data compression. Whether to compress the data before sending is selected according to the network condition and device requirements, and the compression ratio is expressed by the formula:

[0032] Among them is the original data volume, is the data volume after compression.

[0033] The intelligent protocol - recognition unit selects appropriate plugins according to device communication data, uses machine - learning technology to analyze device communication data in real - time, identifies the protocol type and selects appropriate plugins for communication; The intelligent protocol - recognition unit includes a data - acquisition module, a feature - extraction module, and a machine - learning model training module; The data acquisition module adopts a hierarchical acquisition strategy to collect the communication data of the device; the collected data is divided into layers, and the acquisition frequency of each layer of data is set according to the importance and change frequency of the data, and the acquisition frequencies of all layers of data satisfy a preset total acquisition frequency; wherein, represents the number of layers into which the collected data is divided; represents the acquisition frequency of the th layer of data, which is set according to the importance and change frequency of the data in this layer. For example, for a data layer with frequent changes and crucial importance for protocol recognition, a higher acquisition frequency will be set; while for a relatively stable and less important data layer, the acquisition frequency is lower; is the total acquisition frequency, that is, in the entire data acquisition process, the total number of times all levels of data are collected per unit time; The feature extraction module uses a combined feature extraction algorithm, which includes a convolutional neural network algorithm based on deep learning and traditional statistical feature extraction methods; Among them, the traditional statistical feature extraction method calculates the mean value of the data for the collected data according to the formula:

[0034] calculates the variance of the data according to the formula:

[0035] wherein, is the number of data samples, that is, the number of data points participating in the calculation of the mean value; is the value of the th data sample, represents the calculated mean value, which represents the average value of this set of data and is obtained by summing all sample values and dividing by the number of samples; represents the calculated variance, which is obtained by summing the squares of the differences between each sample value and the mean value and then dividing by

[0036] In the training process of the machine learning model training module, cross-validation and regularization are adopted. The cross-validation adopts k-fold cross-validation method, divides the data set into subsets on average, selects subsets for training each time, and the remaining 1 subset is used for validation, and repeats times this process; The regularization adopts L2 regularization, and a regularization term is introduced into the loss function ; Among them, is the regularization parameter, which controls the strength of regularization. The larger it is, the stronger the regularization effect, and the greater the restriction on the model weights. The smaller it is, the weaker the regularization effect; is the model weight. In a machine learning model, the weight determines the relationship between the input features and the output results; represents the sum of the squares of all model weights. By minimizing this regularization term, the model weights can be made as small as possible, avoiding overfitting the training data due to an overly complex model.

[0037] The intelligent protocol recognition unit includes a machine learning model module and a learning monitoring module; The machine learning model module performs incremental learning on newly collected data regularly to update the model's parameters and structure. The learning monitoring module will monitor and record the learning process of the machine learning model.

[0038] The distributed protocol management unit is responsible for the selection of device access proxy nodes, constructing a distributed proxy node network, and connecting the device to the appropriate node according to the geographical location and network topology during device access; The distributed protocol management unit includes a routing algorithm module, a proxy node communication module, and a proxy node monitoring module; The routing algorithm module is used to select a suitable proxy node for the device. For each proxy node , obtain the geographical location distance from the device to this node , network latency , and node load , and calculate the comprehensive evaluation index according to the formula :

[0039] Among them, , , are the weight coefficients and satisfy + + = 1; , , respectively represent the maximum values of the geographical location distance, network latency, and node load among all proxy nodes; represents the geographical location distance from the device to the th proxy node. The geographical location distance affects the efficiency and stability of data transmission. The farther the distance, the greater the signal attenuation may be, and the higher the transmission cost may also be; Represents the maximum value of the geographical location distances among all proxy nodes, which is used to perform normalization processing so that the geographical location distance metrics of different proxy nodes are comparable; Represents the contribution of the geographical location distance to the comprehensive evaluation index ; is the weight coefficient of the geographical location distance, which determines the importance of the geographical location distance in the comprehensive evaluation; Normalize the geographical location distance to the interval [0, 1] to facilitate comprehensive comparison with other factors; Represents the network latency from the device to the th proxy node. Network latency directly affects the real-time performance of data transmission. For Internet of Things applications with high real-time requirements, proxy nodes with low latency are more suitable; Represents the maximum value of the network latencies among all proxy nodes, which is used to perform normalization processing so that the network latency metrics of different proxy nodes have a unified measurement standard; Represents the contribution of the network latency to the comprehensive evaluation index ; is the weight coefficient of the network latency, which reflects the relative importance of the network latency in the comprehensive evaluation, Normalize the network latency for comprehensive calculation with other factors; Represents the load of the th proxy node. Node load reflects the current busyness of the proxy node. Nodes with excessive load may cause data processing delays and packet loss problems, affecting the management efficiency of the Internet of Things protocol; Represents the maximum value of the node loads among all proxy nodes, which is used to perform normalization processing so that the load metrics of different proxy nodes are comparable; Represents the contribution of the node load to the comprehensive evaluation index ; is the weight coefficient of the node load, which determines the weight of the node load in the comprehensive evaluation, Normalize the node load to the interval [0, 1] to facilitate comprehensive evaluation with other factors.

[0040] According to the calculation results, select the proxy node with the minimum value as the target node for device connection; The node communication security module is used to ensure the security of information interaction and collaborative work between proxy nodes. It encrypts communication data using the AES symmetric encryption algorithm, with the key length set to 128 bits or 256 bits. The encryption process is as follows: the sender uses the key to encrypt the plaintext data to generate ciphertext, and the receiver uses the same key to decrypt the ciphertext to obtain the plaintext. At the same time, digital certificate authentication is used for identity verification; The node status monitoring module is used for the management center to comprehensively monitor and manage proxy nodes. The proxy nodes report their own running status and the protocol traffic information processed to the management center according to a preset cycle The value range of is 1 - 60 minutes.

[0041] The distributed protocol management unit includes a proxy node monitoring module, a failover module, and a fault record module; The proxy node monitoring module is used to monitor the running status of proxy nodes in real time. When a fault is detected in a proxy node, the failover module immediately sends a fault notification and a task transfer request to other proxy nodes; After receiving the request, other proxy nodes will select a suitable node to receive the transferred task according to their own load conditions and processing capabilities; The fault record module records the time, cause, and processing result of the fault occurrence.

[0042] The software-defined network traffic scheduling unit dynamically adjusts the protocol traffic path and strategy. With the help of SDN technology, the controller collects information, formulates strategies, and distributes them to switches; The software-defined network traffic scheduling unit includes a controller module and a traffic scheduling strategy module; The controller module is responsible for collecting network information, and its collection cycle can be dynamically adjusted according to the actual operation of the network. When the network is in a busy state, the controller shortens the collection cycle, and when the network is in an idle state, the controller extends the collection cycle.

[0043] The traffic scheduling strategy module is used to dynamically adjust the traffic scheduling strategy according to various factors, perform traffic scheduling based on device priority and network congestion status, and allocate different bandwidth resources for protocol traffic with different real-time requirements.

[0044] The blockchain protocol security management unit ensures the secure storage and access control of protocol information. Using blockchain characteristics, it constructs a blockchain network to store protocol information and realizes protocol access control, authorization, and settlement through smart contracts; The blockchain protocol security management unit includes a blockchain network module and a smart contract module; The blockchain network module adopts a consortium blockchain architecture and has multiple verification nodes. The verification nodes verify and record transactions through a consensus algorithm; The smart contract module is responsible for executing smart contracts, and the execution results are recorded on the blockchain in real time.

[0045] In this embodiment, the hot pluggable unit of the protocol processing module includes a plug-in management interface module and a plug-in inspection module; The plug-in management interface module provides a unified interface for the loading and unloading operations of plug-ins; The plug-in inspection module checks the integrity and compatibility of the plug-ins during the plug-in loading process.

[0046] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various changes or substitutions, and these should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. An IoT protocol management solution supporting hot-plug expansion, characterized in that: The management scheme includes a protocol processing module hot-swap unit, an intelligent protocol identification unit, a distributed protocol management unit, a software-defined network traffic scheduling unit, and a blockchain protocol security management unit; The protocol processing module hot-swap unit, intelligent protocol identification unit, distributed protocol management unit, software-defined network traffic scheduling unit and blockchain protocol security management unit cooperate with each other, wherein: The hot-swap unit of the protocol processing module provides plug-in support for device communication, encapsulates the IoT protocol as an independent plug-in, and supports dynamic loading and unloading at runtime to achieve hot-swap expansion; The intelligent protocol identification unit selects a suitable plug-in according to the device communication data, uses machine learning technology to analyze the device communication data in real time, identifies the protocol type and selects a suitable plug-in for communication; The distributed protocol management unit is responsible for selecting the device access proxy node and building a distributed proxy node network. When the device accesses, it is connected to the appropriate node according to the geographical location and network topology; The software-defined network traffic scheduling unit dynamically adjusts the protocol traffic path and strategy, collects information through the controller, formulates strategies and sends them to the switch with the help of SDN technology; The blockchain protocol security management unit ensures the secure storage and access control of protocol information, utilizes the characteristics of blockchain, builds a blockchain network to store protocol information, and implements protocol access control, authorization and settlement through smart contracts.

2. According to claim 1, a hot-swap expansion-supporting IoT protocol management solution is characterized in that: The protocol processing module hot-swap unit comprises a data receiving module, a data processing module and a data sending module; The data receiving module is used to receive communication data from the IoT device, and includes a data cache for temporary storage during peak data reception. The cache size is dynamically adjusted according to the device processing capacity and network traffic, and the basic cache size is pre-set according to the device processing capacity. , real-time monitoring of network traffic , when network traffic Exceeding the preset high traffic threshold When the proportionality coefficient Increase the cache size, i.e. cache size ,in , ,When the cache is full, a first-in-first-out FIFO strategy is used to replace data; The data receiving module also involves data transmission mode, encryption method and error retransmission, wherein the data transmission mode selects synchronous transmission or asynchronous transmission according to the device requirements. When asynchronous transmission is selected, the data transmission efficiency is The formula means: in is the total amount of data transmitted, is the total transmission time, is the transmission error rate; the encryption method adopts the AES symmetric encryption algorithm, and the encryption strength is set according to the security requirements. In the error retransmission, the number of retransmissions The value range is 1-5 times according to the network status and device reliability. The data processing module has a special parsing and encapsulation logic library for different protocol types, which is used to identify and process various IoT protocol data; The data transmission module includes data compression, and selects whether to compress the data before transmission according to the network status and device requirements. The formula means: in is the original data volume, The amount of data after compression.

3. The IoT protocol management solution supporting hot-plug expansion according to claim 1, characterized in that: The intelligent protocol identification unit includes a data acquisition module, a feature extraction module and a machine learning model training module; The data acquisition module uses a hierarchical acquisition strategy to collect the communication data of the device; the collected data is divided into Layer, each layer of data The acquisition frequency It is set according to the importance and change frequency of the data, and the collection frequency of all layers of data meets is the preset total acquisition frequency; in is the total acquisition frequency; The feature extraction module uses a combined feature extraction algorithm, which includes a convolutional neural network algorithm based on deep learning and a traditional statistical feature extraction method; Among them, the traditional statistical feature extraction method is based on the collected data. Calculate the mean of the data according to the formula: Calculate the variance of the data according to the formula: in, is the number of data samples, For the data samples; The machine learning model training module uses cross-validation and regularization during the training process. The cross-validation uses The fold cross validation method divides the data set into Each time you select a subset The remaining 1 subset is used for validation and repeated times the process; The regularization adopts L2 regularization, and the regularization term is introduced into the loss function. ; in, is the regularization parameter, is the model weight.

4. The IoT protocol management solution supporting hot-plug expansion according to claim 1, characterized in that: The distributed protocol management unit includes a routing algorithm module, an agent node communication module and an agent node monitoring module; The routing algorithm module is used to select a suitable proxy node for the device. , get the geographical distance from the device to the node , network delay and node load , calculate the comprehensive evaluation index according to the formula : in, , , is the weight coefficient and satisfies + + =1; , , Represents the maximum value of geographical distance, network delay and node load among all proxy nodes. According to the calculation results, select The proxy node with the smallest value is used as the target node for the device to connect to; The node communication security module is used to ensure the security of information interaction and collaborative work between proxy nodes. The AES symmetric encryption algorithm is used to encrypt the communication data. The key length is set to 128 bits or 256 bits. The encryption process is: the sender uses the key to encrypt the plaintext data to generate ciphertext, and the receiver uses the same key to decrypt the ciphertext to obtain the plaintext. At the same time, the digital certificate authentication method is used for identity authentication; The node status monitoring module is used by the management center to comprehensively monitor and manage the proxy nodes. The proxy nodes monitor and manage the proxy nodes according to the preset period. Report its own operating status and processed protocol traffic information to the management center, The value range is 1-60 minutes.

5. The IoT protocol management solution supporting hot-plug expansion according to claim 1, characterized in that: The software-defined network traffic scheduling unit includes a controller module and a traffic scheduling strategy module; The controller module is responsible for collecting network information, and its collection cycle can be dynamically adjusted according to the actual operation of the network. When the network is busy, the controller shortens the collection cycle, and when the network is idle, the controller extends the collection cycle; The traffic scheduling strategy module is used to dynamically adjust the traffic scheduling strategy according to multiple factors, perform traffic scheduling according to device priority and network congestion conditions, and allocate different bandwidth resources to protocol traffic with different real-time requirements.

6. The IoT protocol management solution supporting hot-plug expansion according to claim 1, characterized in that: The blockchain protocol security management unit includes a blockchain network module and a smart contract module; The blockchain network module adopts a consortium chain architecture and is equipped with multiple verification nodes, which verify and record transactions through a consensus algorithm; The smart contract module is responsible for executing the smart contract, and its execution results are recorded in real time on the blockchain.

7. The IoT protocol management solution supporting hot-plug expansion according to claim 1, characterized in that: The protocol processing module hot-swap unit comprises a plug-in management interface module and a plug-in inspection module; The plug-in management interface module provides a unified interface for plug-in loading and unloading operations; The plug-in checking module checks the integrity and compatibility of the plug-in during the plug-in loading process.

8. The IoT protocol management solution supporting hot-plug expansion according to claim 1, characterized in that: The intelligent protocol identification unit includes a machine learning model module and a learning monitoring module; The machine learning model module regularly performs incremental learning on newly collected data to update the parameters and structure of the model; The learning monitoring module monitors and records the learning process of the machine learning model.

9. The IoT protocol management solution supporting hot-plug expansion according to claim 1, characterized in that: The distributed protocol management unit includes an agent node monitoring module, a fault switching module and a fault recording module; The proxy node monitoring module is used to monitor the operating status of the proxy node in real time. When a proxy node failure is detected, the fault switching module will immediately send a fault notification and a task transfer request to other proxy nodes; After receiving the request, other proxy nodes will select a suitable node to receive the transferred task based on their own load and processing capabilities; The fault recording module will record the time, cause and processing result of the fault.

Images