Clipping and Hampel Filtering Algorithm to Mitigate Mutual Interference for Automotive Radars

Abstract

The frequency modulated continuous wave (FMCW) radars are widely adopted in the automotive industry to serve the emergency brake assistant (EBA) and automatic cruise control (ACC) functions. Nowadays, autonomous vehicles on-road density increasing and creating a problem of mutual interference. Due to this mutual interference, the structural properties like periodicity and amplitude of intermediate frequency (IF) signal varies and creates undeserved target detections. It is observed that, applying a one-dimension fast Fourier transform (FFT) on the intermediate frequency (IF) results in increased false alarms and missed detections. This paper process a clipping followed by a Hampel filtering on the IF signal to mitigate this mutual interference. Initially, the clipping framework chopoff the unwanted and abrupt amplitude information from the IF signal. The threshold of the clipper circuit is taken as the standard deviation of the acquired IF signal. Second, the Hampel filter was employed to identify the outliers in time-series data and replace them with more representative values. The Hampel is configured in a sliding window to calculate the median by providing the standard deviation of the acquired IF signal. This methodology implementation is easy and can be placed as an intermediate block between IF and FFT. The results demonstrate that the proposed methodology facilitating the good detection rate by decreasing the false alarm rate and missed detections.