Automotive Radar Interference Cancellation Using Dual Receivers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automotive radar systems face interference issues due to mutual interference between radars, leading to information errors and false target recognition, especially in next-generation high-resolution radars for autonomous driving, where signal-to-noise ratio (SNR) decreases, affecting angular resolution and target detection rates.
Innovation Solution
A device and method for processing automotive radar interference signals using a signal receiver with a main receiver and a sub-receiver, filters, and a processor to detect and cancel interference signals by subtracting the interference signal from the target signal, employing frequency offsets and phase correction to isolate and remove interference, thereby improving target detection rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current signal processing techniques with threshold values are used to detect interference, then interference detection is possible, but interference below the threshold value cannot be detected and complicated signal processing is required
Solution Approach 1:
The patent replaces conventional signal processing techniques with machine learning-based interference detection. The trained model automatically identifies interference signals below threshold values without requiring complex signal processing algorithms, thereby improving detection precision while reducing processing complexity.
Solution Approach 2:
The patent transforms the interference detection approach by changing from fixed threshold-based parameter comparison to dynamic machine learning model prediction. The model learns optimal detection parameters during training and adapts to various interference scenarios, enabling detection of sub-threshold interference without manual parameter tuning.
2Measurement precision
If current interference detection methods are used, then interference presence can be determined, but loss and distortion of target signal occurs leading to decrease in target detection rate
Solution Approach 1:
The patent replaces threshold-based interference determination with machine learning-based classification. The trained model distinguishes between interference and target signals more accurately, preserving target signal integrity while detecting interference, thus maintaining high target detection rates.
Solution Approach 2:
The patent uses trained machine learning models that have learned from labeled interference and non-interference data. These models create a virtual representation of interference patterns, enabling accurate interference identification without directly manipulating or potentially distorting the original target signal.
3Reliability
If FMCW signals with fast chirps are used, then resistance to interference is improved, but interference from radars with different chirp slopes appears as broadband noise causing SNR decrease
Solution Approach 1:
The patent replaces conventional FMCW signal processing with machine learning-based interference detection and cancellation. The system identifies interference components in the received signal and subtracts them, preserving the target signal and maintaining high SNR even when interference with different chirp slopes is present.
Solution Approach 2:
The patent extracts and removes interference components from the received FMCW signal using machine learning-based detection. By separating the interference from the target signal and eliminating it through subtraction, the system maintains the integrity and SNR of the remaining target signal.
Data Source
AI summary
The present disclosure relates to an apparatus and method for processing interference signal of an automotive radar, and more particularly, to an apparatus and method for processing interference signal of an automotive radar to detect and cancel interference signal of the automotive radar.