Improved Target Detection Through DNN-Based Multi-Channel Interference Mitigation in Automotive Radar
Shengyi Chen, Marvin Klemp, Jalal Taghia, Uwe Kühnau, Nils Pohl, Rainer Martin
Abstract
Deep learning methods have triggered significant progress in automotive radar-based object detection and classification. However, with an increasing number of radar sensors on the road, mutual interference is unavoidable since these sensors share the same frequency spectrum. Mutual interference affects the robustness of radar processing schemes and thereby the object detection accuracy. Unlike many recent works which focus on interference mitigation for a single receive channel, this paper proposes a multi-channel mitigation approach and seeks to analyze the effect of mutual interference in a multiple-input multiple-output (MIMO) radar. To this end, we first formulate a general signal model for multi-channel interference scenarios. Then, a novel signal separation neural network is proposed for multi-channel interference mitigation which eliminates the efforts of interference detection. We assess the impact of interference in terms of the reconstruction error, angle estimation error, and the target detection accuracy in both real-world and simulated interference scenarios. It is demonstrated that the proposed neural network can provide superior signal recovery, massively reduces the false-positive rate, and significantly improves the accuracy of object detection even in the presence of severe interference.