Vehicle Network Message Compression Using ML Pattern Replacement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated and autonomous vehicle systems face challenges in efficiently reducing the size of messages transmitted over communication channels, leading to high bandwidth and storage requirements, especially when large volumes of data need to be transmitted or stored from multiple vehicles to remote networked resources.
Innovation Solution
Implementing a computer-implemented method and system that uses trained machine learning models to identify data patterns in messages and replace them with predefined lossless representations of reduced size, which are then transmitted or stored, thereby reducing the overall message size without losing data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If large volumes of message data are transmitted from vehicles to remote networked resources, then comprehensive data analysis and processing can be performed, but bandwidth requirements and communication resources are significantly increased
Solution Approach 1:
The patent extracts only the essential and variable parts of message data for transmission to remote resources. By identifying and separating critical information elements from redundant or static content, the system transmits minimal necessary data while preserving analytical value, thus reducing bandwidth consumption without compromising data completeness for analysis purposes
Solution Approach 2:
The message data is segmented into different components based on importance and variability. Critical time-sensitive and variable parameters are extracted and transmitted separately from static or less important data, allowing selective transmission that optimizes bandwidth usage while maintaining the integrity of essential information for remote analysis
2Loss of information
If message data is stored locally in vehicles for later analysis or upload, then data availability is improved, but storage resources in vehicles are significantly consumed
Solution Approach 1:
The system extracts and stores only the essential variable parameters locally in vehicles rather than retaining complete message datasets. By identifying and preserving only the critical data elements needed for local analysis and potential upload, the system maintains data availability for immediate processing while dramatically reducing local storage requirements
Solution Approach 2:
Data is segmented into locally-stored essential parameters and remotely-stored comprehensive datasets. The vehicle retains only the minimal necessary data for local operations and emergency analysis, while comprehensive archives are maintained remotely, optimizing the balance between local data availability and storage resource consumption
3Device complexity
If rule-based pattern detection is used for message optimization, then implementation simplicity is maintained, but adaptability to new message content patterns is limited
Solution Approach 1:
The system employs machine learning models that automatically learn and adapt to new message patterns without requiring manual rule updates. The models self-improve by processing incoming data and identifying emerging patterns autonomously, providing both adaptability to new content and maintaining relative implementation simplicity through automated learning rather than manual rule management
Solution Approach 2:
The pattern detection mechanism transitions from static rule-based systems to dynamic machine learning models that continuously adapt to changing message patterns. The system evolves its detection capabilities over time by learning from new data, allowing it to maintain simplicity of operation while gaining versatility in handling diverse and evolving message content
Data Source
AI summary
A computer implemented method of reducing a size of a message intercepted on a communication channel of a vehicle, comprising using one or more processors of a vehicular device. The processor(s) is adapted for receiving one or more of a plurality of messages intercepted by one or more devices adapted to monitor messages transmitted via one or more segments of one or more communication channels of a vehicle, applying one or more trained machine learning models to identify one or more of a plurality of data patterns in one or more of the messages, adjusting one or more of the messages by replacing each of the identified data pattern(s) with a respective predefined lossless representation having a reduced size compared to the identified data pattern and transmitting the adjusted message(s) to a remote system via one or more upload communication channels.