Smartphone Digital Key Preventing Relay Attacks via Movement Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Keyless entry systems in vehicles are vulnerable to relay attacks, where a signal repeater bypasses encryption and security protocols, allowing unauthorized access by mimicking the electronic key's location, and existing solutions like placing keys in Faraday cages are not satisfactory.
Innovation Solution
A computer system and portable electronic device combination that uses machine learning to train a model on user movement signals, generating direction vectors and characteristic vectors to authenticate access to vehicles, incorporating sensors like accelerometers and gyroscopes, and applying supervised classification algorithms to prevent relay attacks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If keyless entry systems are implemented, then ease of operation is improved, but security against relay attacks deteriorates
Solution Approach 1:
The system performs preliminary actions by training a machine learning model in advance with movement data collected during acquisition sessions. The model learns to recognize legitimate user movement patterns before actual authentication occurs, enabling it to detect and prevent relay attacks in real-time without compromising the convenience of contactless access
Solution Approach 2:
The patent introduces an intermediary machine learning model that acts as a mediator between the physical key and the vehicle's authentication system. This model analyzes movement signals and determines whether the key's movement is legitimate or part of a relay attack, thereby enhancing security while maintaining the ease of contactless operation
2Reliability
If Faraday cages are used to prevent relay attacks, then security is improved, but device complexity and ease of operation worsen
Solution Approach 1:
The patent replaces the mechanical/physical Faraday cage solution with a computational approach using machine learning. Instead of using electromagnetic shielding to block signals, the system uses an ML model to analyze movement patterns and detect relay attacks, thereby maintaining security without adding physical complexity or restricting user operations
3Measurement precision
If movement signals are collected and analyzed, then security precision is improved, but loss of time increases
Solution Approach 1:
The system performs the time-consuming training process in advance during acquisition sessions, so that when actual authentication is needed, the pre-trained model can quickly and accurately analyze movement signals without causing noticeable delays to the user. This preliminary preparation resolves the contradiction between achieving high measurement precision and minimizing time loss
Data Source
AI summary
Computer systems for training a machine learning model for preventing relay attacks. Including portable electronic devices capable of controlling the opening or closing, contactlessly, of an access to a road vehicle on the basis of a pretrained machine learning model. The general principle is based on the use of smartphones as a digital key for accessing a road vehicle. Machine learning is used to train a learning model capable of predicting the movement of a smartphone of this type as it approaches or moves away from the road vehicle. Subsequently, access to the road vehicle is authorized only when same receives the information on the movement of the smartphone.