Machine Learning Message Decoding Strategy for Security Engine Memory Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current security engines face inefficiencies and memory resource constraints due to the need to execute multiple decoding strategies for various server types, making it difficult to effectively parse and decode messages across different servers, which can lead to attack bypasses.
Innovation Solution
A method utilizing a machine learning model trained on normal and malicious messages to determine the optimal decoding strategy for a message, allowing the security engine to execute this strategy only once, reducing memory occupation and improving parsing efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the security engine executes multiple decoding strategies for different server types, then the compatibility with various servers is improved, but the memory resource consumption and processing time increase significantly
Solution Approach 1:
The patent applies preliminary action by pre-training a machine learning model with decoding strategy information from multiple servers before actual message processing. The model learns to predict the appropriate decoding strategy based on message characteristics, eliminating the need to execute multiple decoding strategies during runtime. This pre-computation approach resolves the contradiction by preparing the solution in advance, allowing fast single-strategy execution while maintaining multi-server compatibility.
2Adaptability or versatility
If all parsing and decoding methods are sent to the security engine, then the coverage of different server types is improved, but the memory occupation and system performance consumption increase
Solution Approach 1:
The patent extracts the essential decoding strategy information from multiple server types and embeds it within a machine learning model during the training phase. Instead of storing and executing all possible decoding methods in the security engine, only the trained model parameters are retained. This extraction process separates the strategic decision-making (stored in the model) from the execution, significantly reducing memory occupation while preserving coverage of different server types through the model's predictive capability.
3Reliability
If multiple decoding strategies are executed for message analysis, then the accuracy of malicious message detection is improved, but the processing time and resource consumption increase
Solution Approach 1:
The patent applies preliminary action by pre-training the machine learning model with decoding strategy information from multiple servers before actual message processing. The model learns to predict the appropriate decoding strategy based on message characteristics, eliminating the need to execute multiple decoding strategies during runtime. This pre-computation approach resolves the contradiction by preparing the solution in advance, allowing fast single-strategy execution while maintaining multi-server compatibility.
Solution Approach 2:
The patent replaces the mechanical execution of multiple decoding strategies with a machine learning-based predictive system. Instead of systematically trying multiple decoding methods (mechanical approach), the trained model intelligently predicts the most appropriate decoding strategy based on learned patterns from training data. This substitution transforms a resource-intensive mechanical process into an efficient intelligent decision-making process, maintaining detection accuracy while reducing processing time.
Data Source
AI summary
The present disclosure discloses a method for determining a message decoding strategy and apparatus, related to the field of message decoding. After the message is acquired, the message is inputted to a preset machine learning model to determine the type of the message and the decoding strategy of the message. In response to the message being a normal message, the message is decoded based on the obtained decoding strategy and the message is released; in response to the message being a malicious message, the message is rejected, and the preset machine learning model is obtained by training multiple normal messages and malicious messages in advance. Firstly, the type of message and the decoding strategy needed are determined by machine learning model, which effectively reduces the memory occupation of the system and improves the parsing and decoding efficiency of the message.