V2X Misbehavior Detection via CPM Cross-Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
V2X messages can be subjected to attacks by malicious parties, where attackers transmit false data describing the driving environment, leading to misbehavior and potential safety hazards for Advanced Driver Assistance Systems (ADAS) and autonomous driving systems.
Innovation Solution
A detection system that receives V2X messages and Collective Perception Messages (CPMs) from remote devices, determines the relevance and accuracy of the data by checking timestamps, location, and score thresholds, and identifies misbehavior by verifying the presence of objects, thereby ignoring erroneous data to prevent adverse effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If V2X messages are used to transmit environmental data for ADAS and autonomous driving systems, then the performance and functionality of these systems are improved, but the system becomes vulnerable to cyber-attacks and spoofing by malicious parties
Solution Approach 1:
The patent introduces Collective Perception Messages (CPMs) as intermediary data sources from multiple remote devices to verify the authenticity of V2X messages. Instead of directly trusting a single V2X transmitter, the system uses CPMs from multiple independent sources as mediators to cross-validate environmental data, thereby maintaining system performance while enhancing security against spoofing attacks
Solution Approach 2:
The system implements a feedback mechanism where received V2X messages are continuously verified against CPM data from multiple remote devices. The verification process provides feedback on message authenticity by comparing sensor data from different sources, allowing the system to detect and reject spoofed messages while maintaining reliable operation of ADAS and autonomous driving functions
2Measurement precision
If multiple CPMs from multiple remote devices are collected and verified to detect misbehavior, then the accuracy of detecting spoofed V2X messages is improved, but the system complexity and computational requirements increase
Solution Approach 1:
The patent segments the verification process into distinct modular steps: receiving V2X messages, collecting CPMs from multiple remote devices, filtering relevant CPMs based on spatiotemporal criteria, comparing sensor data, and detecting misbehavior. This segmentation allows the complex detection system to process information in manageable stages, improving detection accuracy while controlling system complexity through structured data flow
Solution Approach 2:
The system applies partial verification by selectively processing only those CPMs that meet specific relevance criteria (spatial proximity, temporal proximity, sensor type matching) rather than verifying all received messages equally. This partial action approach maintains high detection accuracy for suspicious messages while reducing unnecessary computational overhead from processing irrelevant data
3Measurement precision
If strict verification criteria including time range, location, and score thresholds are applied to filter relevant CPMs, then the precision of data filtering is improved, but the processing time and computational load increase
Solution Approach 1:
The patent applies preliminary filtering actions by pre-establishing relevance criteria for CPM selection, including spatial proximity thresholds, temporal windows around the target time, and sensor type matching rules. By pre-defining these filtering parameters, the system可以快速 identify and process only relevant CPMs without performing exhaustive analysis on all received data, thus maintaining high filtering precision while minimizing processing time
Data Source
AI summary
The disclosure includes embodiments for an ego vehicle to detect misbehavior. According to some embodiments, a method includes receiving a V2X message from an attacker. The V2X message includes V2X data describing a location of an object at a target time. The method includes receiving a set of CPMs from a set of remote devices. The set of CPMs include remote sensor data describing a free space region within the roadway environment. The method includes determining a relevant subset of the CPMs include remote sensor data that is relevant to detecting misbehavior. The method includes determining, based at least in part on the remote sensor data of the relevant subset, that the object is not located at the location at the target time. The method includes detecting the misbehavior by the attacker based on the determination that the object is not located at the location at the target time.