Information processing method and system for unified decoding of different private communication protocols

Abstract

The application discloses an information processing method and system for unified decoding of different private communication protocols, and the method comprises the following steps: S1, receiving original communication data frames from different devices; S2, inputting the original communication data frames into a pre-trained deep learning model, identifying the private communication protocol type to which the data frames belong, and outputting a protocol type identifier; S3, performing transmission layer protocol compatible processing on the original communication data frames according to the protocol type identifier, and generating data frames in a unified transmission format; S4, performing private protocol decoding on the data frames in the unified transmission format based on the protocol type identifier, and extracting original data content; S5, converting the original data content into standard data in a unified format through preset standardization rules; and S6, buffering the standard data to a data pool and distributing the standard data according to a scheduling request of an upper-layer business system. The application has the advantages of efficient and stable information transmission.

Description

Technical Field

[0001] This invention mainly relates to the field of communication technology, specifically to an information processing method and system for unified decoding of different proprietary communication protocols. Background Technology

[0002] In modern power dispatching systems, industrial automation control systems, and other fields related to real-time data transmission and processing, the diversity and complexity of communication protocols are increasingly becoming a major challenge in system integration. Each device often uses its own proprietary communication protocol for data transmission. These protocols differ significantly in data frame structure, encoding methods, transmission protocols, and error checking mechanisms, leading to compatibility issues for the dispatching end when processing data from different devices. This situation severely impacts the uniformity and efficiency of the dispatching system, increasing system complexity and maintenance costs. Specifically, this manifests in: (1) Diversity of communication protocols: There are significant differences in the underlying transmission protocol, data encoding format, packet structure and transmission frequency of different devices. This means that the scheduling end needs to develop specific parsing logic for each protocol when parsing data, which increases the complexity of the system.

[0003] (2) Complexity and maintenance difficulty of the scheduling terminal: As the scale of system integration expands, the types of protocols and the number of devices that the scheduling terminal needs to handle increase rapidly. The traditional protocol processing method usually involves writing a dedicated parsing module for each protocol, which can easily introduce compatibility issues during upgrades and expansions.

[0004] (3) Insufficient data uniformity: Due to the different data formats and transmission methods of each device, even after decoding is completed, differences in data formats still exist. Different formats mean that each data type needs to be processed separately, making it impossible to achieve a standardized data flow, which affects the real-time performance and processing efficiency of the system.

[0005] A common approach to solving multi-specification problems using existing technologies is to develop a dedicated parsing module for each specification. While this method can achieve parsing of different specifications to a certain extent, the number and complexity of parsing modules increase with the increase in devices and specification types. This leads to poor system maintainability, and requires significant manpower to update or introduce new specifications. Some systems use data conversion tools to convert data of different formats to a unified format, but these tools typically require manual configuration for each new specification, making it difficult to adapt to frequently changing specification versions. Due to the complexity of data formats and content, this manual approach is prone to errors and struggles to meet the demands of real-time data processing. Summary of the Invention

[0006] To address the technical problems existing in the prior art, this invention provides an information processing method and system for efficient and stable information transmission and unified decoding of different proprietary communication protocols.

[0007] To solve the above-mentioned technical problems, the technical solution proposed by this invention is as follows: A method for information processing that unifies decoding of different proprietary communication protocols includes the following steps: S1: Receive raw communication data frames from different devices; S2: Input the original communication data frame into a pre-trained deep learning model, identify the private communication protocol type to which the data frame belongs, and output the protocol type identifier; S3: Based on the protocol type identifier, perform transport layer protocol compatibility processing on the original communication data frame to eliminate format differences caused by different transport layer protocols and generate a data frame with a unified transmission format. S4: Based on the protocol type identifier, dynamically call the decoding unit corresponding to the protocol to perform private protocol decoding on the data frame of the unified transmission format and extract the original data content; S5: Input the raw data content into the data standardization module, and convert it into standard data that conforms to a unified format through preset standardization rules; S6: Cache the standard data in the data pool and distribute it according to the scheduling request of the upper-layer business system.

[0008] Preferably, in step S2, the deep learning model uses a recurrent neural network (RNN) to extract and classify the features of the data frame; the RNN receives the data frame sequence as input and processes it into a sequence data vector, then maps the features in the input data frame to a high-dimensional space to form a continuous feature vector sequence; the time series features are extracted through the recurrent layer, and finally the classification layer outputs the reduced type probability distribution.

[0009] Preferably, the recurrent layers in the recurrent neural network employ Long Short-Term Memory (LSTM) units to capture long sequence dependency features in the data frames and avoid gradient vanishing.

[0010] Preferably, an attention layer is further introduced into the recurrent neural network; the attention layer uses the hidden state sequence of each time step of the recurrent neural network. As input, through a trainable query vector Calculate the attention weights at each time step and hide the state The global context vector is generated by weighted summation. The global context vector is used for subsequent reduction type classification.

[0011] Preferably, the decoding unit in step S4 adopts a modular design, with each decoding unit corresponding to a private communication protocol and interacting with the system through a unified interface; the system dynamically loads the corresponding decoding unit according to the protocol type identifier to achieve plug-and-play functionality.

[0012] Preferably, the decoding unit is deployed as a dynamic link library or plugin.

[0013] Preferably, the data standardization module in step S5 includes a rule engine and a standardization template; the rule engine executes predefined field mapping, data type conversion and unit conversion rules to map the original data content to the standardization template and generate a data structure in a unified format.

[0014] Preferably, the standardized template is defined in XML or JSON format, including field names, data types, value ranges and default values, and supports dynamically loading different templates according to different business scenarios.

[0015] This invention also discloses an information processing system for unified decoding of different proprietary communication protocols, comprising: The protocol identification module is used to automatically identify the protocol type of data frames based on a deep learning model; The transport layer compatibility module is used to uniformly process data frames from different transport layer protocols; The protocol decoding module includes multiple modular decoding units, which are used to call the decoding logic of the corresponding protocol based on the recognition result; The data standardization module is used to convert decoded data into a unified format. The data caching and scheduling module is used to cache and schedule standardized data. The system interface module is used to provide standardized data to the upper-layer business system.

[0016] The present invention further discloses a computer-readable storage medium having a computer program stored thereon, the computer program executing the steps of the method described above when run by a processor.

[0017] Compared with the prior art, the advantages of the present invention are as follows: This invention effectively addresses the shortcomings of current technologies in handling multi-reduction parsing, data format unification, and system maintenance, providing a highly efficient, stable, and easily scalable solution for scheduling systems. Specific technical effects are as follows: 1. Unified Protocol Identification and Processing: This invention proposes a unified protocol identification and processing technical solution, which can automatically identify and parse different private communication protocols, reducing the cost of developing and maintaining different protocol parsing modules; at the same time, through modular design, it can quickly adapt to the introduction and upgrading of new protocols.

[0018] 2. Transport Layer Compatibility and Data Standardization: This invention incorporates a transport layer compatibility module and a data standardization module, enabling the standardization of data formats with different protocols, ensuring the uniformity of the transport layer and data content. This innovation solves the problem of inconsistent data formats in existing technologies, greatly simplifies the processing flow at the scheduling end, and provides standardized data resources for subsequent business processing.

[0019] 3. Scalability and Real-time Performance: The system architecture of this invention is highly scalable, allowing users to support new specifications through simple configuration and module loading. The system also possesses excellent real-time performance, capable of quickly completing decoding and standardization processing after data reception, ensuring that the scheduling end can respond to and process various business needs in a timely manner. Attached Figure Description

[0020] Figure 1 This is a block diagram of an information processing system according to an embodiment of the present invention.

[0021] Figure 2 This is a diagram of the protocol identification network model based on RNN in an embodiment of the present invention. Detailed Implementation

[0022] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0023] The information processing method for unified decoding of different proprietary communication protocols provided in this embodiment of the invention includes the following steps: S1: Receive raw communication data frames from different devices; S2: Input the original communication data frame into a pre-trained deep learning model, identify the private communication protocol type to which the data frame belongs, and output the protocol type identifier; S3: Based on the protocol type identifier, perform transport layer protocol compatibility processing on the original communication data frame to eliminate format differences caused by different transport layer protocols and generate a data frame with a unified transmission format. S4: Based on the protocol type identifier, dynamically call the decoding unit corresponding to the protocol to perform private protocol decoding on the data frame of the unified transmission format and extract the original data content; S5: Input the raw data content into the data standardization module, and convert it into standard data that conforms to a unified format according to the preset standardization rules; S6: Cache standard data in the data pool and distribute it according to the scheduling requests of the upper-layer business system.

[0024] like Figure 1 As shown, this embodiment of the invention also provides an information processing system for unified decoding of different proprietary communication protocols, which mainly includes the following modules: The protocol identification module is used to analyze the header information of the data frame after receiving the original communication data frames from different devices, and to identify the source of the data and the communication protocol used. The transport layer compatibility module is used to perform unified processing on different transmission methods (different transport layer protocols, packet structures, etc.) based on the identification results, and to standardize the data. The protocol decoding module is used to call the corresponding private protocol decoding logic based on the output of the transport layer compatibility module, parse the data content, and convert the original encoded data into a unified standard data format. The data standardization module is used to further standardize the decoded data to ensure that all data conforms to the unified data format requirements of the scheduling end. The data caching and scheduling module is used to cache standardized data and dynamically allocate and process data according to scheduling requirements to ensure the real-time performance and efficiency of the system. The system interface module provides a unified interface to pass the processed standardized data resources to the upper-layer business system, supporting the expansion of various business applications.

[0025] This invention effectively addresses the shortcomings of current technologies in handling multi-reduction parsing, data format unification, and system maintenance, providing a highly efficient, stable, and easily scalable solution for scheduling systems. Specific technical effects are as follows: 1. Unified Protocol Identification and Processing: This invention proposes a unified protocol identification and processing technical solution, which can automatically identify and parse different private communication protocols, reducing the cost of developing and maintaining different protocol parsing modules; at the same time, through modular design, it can quickly adapt to the introduction and upgrading of new protocols.

[0026] 2. Transport Layer Compatibility and Data Standardization: This invention incorporates a transport layer compatibility module and a data standardization module, enabling the standardization of data formats with different protocols, ensuring the uniformity of the transport layer and data content. This innovation solves the problem of inconsistent data formats in existing technologies, greatly simplifies the processing flow at the scheduling end, and provides standardized data resources for subsequent business processing.

[0027] 3. Scalability and Real-time Performance: The system architecture of this invention is highly scalable, allowing users to support new specifications through simple configuration and module loading. The system also possesses excellent real-time performance, capable of quickly completing decoding and standardization processing after data reception, ensuring that the scheduling end can respond to and process various business needs in a timely manner.

[0028] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0029] The information processing system for unified decoding of different private communication protocols of the present invention includes a protocol identification module, a transport layer compatibility module, a protocol decoding module, a data standardization module, a data caching and scheduling module, and a system interface module. The protocol identification module is developed based on artificial intelligence technology. The transport layer compatibility module and the protocol decoding module support different protocols and transmission methods through modular design. The data standardization module achieves standardized conversion through a rule engine or template method.

[0030] Specifically, the protocol identification module is based on the identification of data protocols using artificial intelligence technology; the protocol identification module uses deep learning technology to build a model that can dynamically learn and identify different protocol features.

[0031] First, communication data samples from various devices are collected. These samples include data frames with different protocols. The dataset needs to include various data frame structures, header information, data encoding methods, etc., to cover as many protocol types as possible. Then, each sample data frame is labeled with the corresponding protocol type to build a training dataset. like Figure 2 As shown, the deep learning model uses a recurrent neural network (RNN) to extract and classify features from data frames. An RNN is a neural network specifically designed for processing sequential data. Communication protocol data frames are typically sequential data, and their features may be correlated at different positions within the frame. The RNN can utilize its recurrent structure to capture time-series features in the data frame, thereby effectively identifying different protocol types. The recurrent neural network model accepts a sequence of data frames as input, including fields such as a frame header, data portion, and checksum. These fields are processed into sequential data vectors at the input layer, and then the features in the input data frame are mapped to a high-dimensional space, forming a continuous sequence of feature vectors. The embedding layer transforms discrete symbolic features into vector forms that can be input into the neural network. Since data frame sequences can be long, this invention uses a Long Short-Term Memory (LSTM) unit in the recurrent neural network to capture dependencies in long sequences. By introducing a gating mechanism, features at the time step can be retained or forgotten, solving the gradient vanishing problem in traditional RNNs when processing long sequences.

[0032] The protocol recognition module of this invention introduces an attention mechanism to enhance the model's ability to focus on features at specific time steps when processing long data frames containing complex structures. The attention mechanism improves recognition accuracy by assigning different weights to features at each time step, allowing the network to focus on the most important parts for reduction recognition. Specifically, based on extracting data frame sequence features using a recurrent neural network (RNN) or its variant, a long short-term memory network (LSTM), an attention layer is constructed at the top layer of the network: this layer uses the hidden state sequence of each time step of the LSTM... As input, a trainable query vector is used. Calculate a similarity score with each hidden state (e.g., using additive attention). Or point accumulation attention, where This is the sequence length of the data frame, i.e., the total number of time steps; For the first The hidden state vector at each time step; For row vector parameters; , (This is the weight matrix), which is then normalized using Softmax to obtain the attention weight distribution. These weights reflect the importance of each time step to the reduction recognition task. The weights are then summed with the corresponding hidden states to generate a global context vector. This vector integrates key information from the entire frame, especially enhancing local features such as the frame header, control fields, and checksums that play a decisive role in distinguishing protocols.

[0033] Through the aforementioned mechanism, the model can dynamically ignore irrelevant or redundant information, focusing on the most discriminative time step, thus maintaining high-precision reduction recognition even with varying data frame lengths and complex structures. Compared to traditional RNNs that rely solely on the last time step or average pooling, the attention mechanism not only alleviates the information decay problem in long sequences but also provides the model with a degree of interpretability—visualizing the weights allows for intuitive identification of the intra-frame regions the model focuses on. Furthermore, this invention seamlessly integrates the attention layer with the recurrent structure of LSTM, supporting end-to-end training. This enables the attention weights to adaptively adjust according to the recognition task, significantly improving the robustness and generalization ability of protocol recognition and laying a reliable foundation for the accurate invocation of subsequent decoding modules.

[0034] Finally, the RNN output is transformed into a fixed-dimensional vector through a fully connected layer for subsequent classification tasks. The output of the fully connected layer is then transformed into a reduced probability distribution using a softmax function, allowing the model to determine the most likely reduced type of the input data frame.

[0035] Compared to traditional feature matching methods, RNNs can capture the dependencies of long-distance features in data frames. Furthermore, RNNs can identify the reduction type in real time during data frame reception, enabling the system to respond quickly and invoke the corresponding decoding logic. In addition, through continuous online learning or incremental learning, RNNs can gradually adapt to new reduction types, achieving adaptive recognition.

[0036] Specifically, the protocol decoding module adopts a modular design, supporting protocol decoding for various devices. This design ensures the system's flexibility and scalability, making the introduction and integration of new protocols simpler and more efficient.

[0037] The protocol decoding module is designed as a set of independent decoding units, each responsible for decoding a specific device protocol. Each unit is an independent module with clearly defined input / output interfaces. Using a dynamic library or plug-in mechanism, the decoding units can be dynamically loaded and unloaded during system runtime, facilitating maintenance and updates. For each protocol, a corresponding decoding logic module is developed, including steps such as data frame parsing, field extraction, and data format conversion. Each decoding unit has a unified interface specification to ensure compatibility between different units. The unified interface includes the input raw data stream and the output standardized data structure.

[0038] After the protocol identification module determines the protocol type, the system dynamically calls the corresponding decoding unit to parse the data. The decoding result is returned to the system through a unified interface and stored in a standardized data structure for subsequent module processing. Integration of new protocols is achieved through a new decoding unit registered in the system. Modular design allows the system to quickly adapt to the introduction of new protocols without requiring large-scale system modifications. In the transparent transmission architecture, the decoding module can be embedded in the data channel to process transmitted data in real time and feed the decoding results back to the upper-layer application.

[0039] Specifically, the data standardization module achieves data standardization transformation through a rule engine and standardized templates. The task of the data standardization module is to transform the decoded data into a unified format, providing consistent data resources for subsequent business processing.

[0040] The rules engine is the core of the data standardization module, responsible for executing predefined standardization rules and converting decoded data into a standard format. The rules engine allows users to define transformation rules, including data field mapping, data type conversion, and unit conversion. Rules are stored in configuration files for easy modification and expansion. Standardized data templates are designed to meet different business needs. Each template defines the standardized data structure, including field names, data types, and default values. Templates can be defined based on XML or JSON format and loaded and parsed by the template engine.

[0041] The decoded data first undergoes preliminary processing by a rules engine, transforming and mapping various data fields according to rules. The data processed by the rules engine is then further matched to standardized templates to ensure all data conforms to a unified format. Standardized data undergoes a validation mechanism to check its integrity and consistency, ensuring the data can be correctly used by subsequent business modules. In the transparent transmission architecture, the data standardization module can be directly embedded into the data processing pipeline, performing real-time standardization transformations on the transmitted data. The system allows different standardized templates to be applied to different data streams to adapt to the needs of various business scenarios.

[0042] The present invention further discloses a computer-readable storage medium having a computer program stored thereon, the computer program executing the steps of the method described above when run by a processor.

[0043] The medium of the present invention corresponds to the method described above and also has the advantages described above.

[0044] The present invention can implement all or part of the processes in the methods of the above embodiments, or it can be implemented by hardware related to computer program instructions. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of the above method embodiments. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable storage medium includes: any entity or device capable of carrying computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. The memory is used to store computer programs and / or modules. The processor implements various functions by running or executing the computer programs and / or modules stored in the memory, and by calling data stored in the memory. The memory may include high-speed random access memory, as well as non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital (SD) cards, flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0045] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. An information processing method for unified decoding of different proprietary communication protocols, characterized in that, Including the following steps: S1: Receive raw communication data frames from different devices; S2: Input the original communication data frame into a pre-trained deep learning model, identify the private communication protocol type to which the data frame belongs, and output the protocol type identifier; S3: Based on the protocol type identifier, perform transport layer protocol compatibility processing on the original communication data frame to eliminate format differences caused by different transport layer protocols and generate a data frame with a unified transmission format. S4: Based on the protocol type identifier, dynamically call the decoding unit corresponding to the protocol to perform private protocol decoding on the data frame of the unified transmission format and extract the original data content; S5: Convert the original data content into standard data that conforms to a unified format through preset standardization rules; S6: Cache the standard data in the data pool and distribute it according to the scheduling request of the upper-layer business system.

2. The information processing method for unified decoding of different proprietary communication protocols according to claim 1, characterized in that, In step S2, the deep learning model uses a recurrent neural network (RNN) to extract and classify the features of the data frames. The RNN receives the data frame sequence as input and processes it into a sequence data vector. Then, it maps the features in the input data frame into a high-dimensional space to form a continuous feature vector sequence. The time series features are extracted through the recurrent layer, and finally, the classification layer outputs the reduced type probability distribution.

3. The information processing method for unified decoding of different proprietary communication protocols according to claim 2, characterized in that, The recurrent layers in the recurrent neural network employ Long Short-Term Memory (LSTM) units to capture long sequence dependency features in data frames and avoid gradient vanishing.

4. The information processing method for unified decoding of different proprietary communication protocols according to claim 2 or 3, characterized in that, An attention layer is also introduced into the recurrent neural network; the attention layer uses the hidden state sequence of each time step of the recurrent neural network. As input, through a trainable query vector Calculate the attention weights at each time step and hide the state The global context vector is generated by weighted summation. The global context vector is used for subsequent reduction type classification.

5. The information processing method for unified decoding of different proprietary communication protocols according to claim 1, 2, or 3, characterized in that, The decoding unit in step S4 adopts a modular design. Each decoding unit corresponds to a private communication protocol and interacts with the system through a unified interface. The system dynamically loads the corresponding decoding unit according to the protocol type identifier to achieve plug-and-play functionality.

6. The information processing method for unified decoding of different proprietary communication protocols according to claim 5, characterized in that, The decoding unit is deployed as a dynamic link library or plugin.

7. The information processing method for unified decoding of different proprietary communication protocols according to claim 1, 2, or 3, characterized in that, In step S5, the data standardization module has preset standardization rules. The data standardization module includes a rule engine and a standardization template. The rule engine executes predefined field mapping, data type conversion, and unit conversion rules to map the original data content to the standardization template and generate a data structure in a unified format.

8. The information processing method for unified decoding of different proprietary communication protocols according to claim 7, characterized in that, The standardized templates are defined in XML or JSON format and include field names, data types, value ranges, and default values. They support dynamically loading different templates according to different business scenarios.

9. An information processing system for unified decoding of different proprietary communication protocols, characterized in that, include: The protocol identification module is used to automatically identify the protocol type of data frames based on a deep learning model; The transport layer compatibility module is used to uniformly process data frames from different transport layer protocols; The protocol decoding module includes multiple modular decoding units, which are used to call the decoding logic of the corresponding protocol based on the recognition result; The data standardization module is used to convert decoded data into a unified format. The data caching and scheduling module is used to cache and schedule standardized data. The system interface module is used to provide standardized data to the upper-layer business system.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program, when run by a processor, performs the steps of the method as described in any one of claims 1-8.

Images