In recent times, there has been a significant focus on the development of automotive high-resolution 77 GHz CS (Chirp Sequence) radar, a technology essential for autonomous driving. However, with the increasing popularity of vehicle-mounted CS radars, the issue of intensive inter-radar wideband interference has emerged as a significant concern, leading to undesirable missed targe detection. To solve this problem, various algorithm and learning based approaches have been proposed for wideband interference suppression. In this study, we begin by conducting extensive simulations to assess the SINR (Signal to Interference plus Noise Ratio) and execution time of these approaches in highly demanding scenarios involving up to 7 interfering radars. Subsequently, to validate these approaches could generalize to real data, we perform comprehensive experiments on inter-radar interference using multiple 77 GHz MIMO (Multiple-Input and Multiple-output) CS radars. The collected real-world interference data is then utilized to validate the generalization capacity of these approaches in terms of SINR, missed detection rate, and false detection rate.

In recent years, high-resolution 77 GHz band automotive radar, which is indispensable for autonomous driving, has been extensively investigated. In the future, as vehicle-mounted CS (chirp sequence) radars become more and more popular, intensive inter-radar wideband interference will become a serious problem, which results in undesired miss detection of targets. To address this problem, learning-based wideband interference mitigation method has been proposed, and its feasibility has been validated by simulations. In this paper, firstly we evaluated the trade-off between interference mitigation performance and model training time of the learning-based interference mitigation method in a simulation environment. Secondly, we conducted extensive inter-radar interference experiments by using multiple 77 GHz MIMO (Multiple-Input and Multiple-output) CS radars and collected real-world interference data. Finally, we compared the performance of learning-based interference mitigation method with existing algorithm-based methods by real experimental data in terms of SINR (signal to interference plus noise ratio) and MAPE (mean absolute percentage error).

A non-linear extension of generalized hyperplane approximation (GHA) method is introduced in this letter. Although GHA achieved a high-confidence result in motion parameter estimation by utilizing the supervised learning scheme in histogram of oriented gradient (HOG) feature space, it still has unstable convergence range because it approximates the non-linear function of regression from the feature space to the motion parameter space as a linear plane. To extend GHA into a non-linear regression for larger convergence range, we derive theoretical equations and verify this extension's effectiveness and efficiency over GHA by experimental results.

Considering the real existent fact of the ideal edge and the learning style of image analogy without reference parameters, a blind image recovery algorithm using a self-adaptive learning method is proposed in this paper. We show that a specific local image patch with degradation characteristic can be utilized for restoring the whole image. In the training process, a clear counterpart of the local image patch is constructed based on the ideal edge assumption so that identification of the Point Spread Function is no longer needed. Experiments demonstrate the effectiveness of the proposed method on remote sensing images.