DC-offset effect cancelation method using mean-padding FFT for automotive UWB radar sensor

Abstract

To improve road safety and realize intelligent transportation, Ultra-Wideband (UWB) radars sensor in the 24 GHz domain are currently under development for many automotive applications. Automotive UWB radar sensor must be small, require low power and inexpensive. By employing a direct conversion receiver, automotive UWB radar sensor is able to meet size and cost reduction requirements. We developed Automotive UWB radar sensor for automotive applications. The developed receiver of the automotive radar sensor is direct conversion architecture. Direct conversion architecture poses a dc-offset problem. In automotive UWB radar, Doppler frequency is used to extract velocity. The Doppler frequency of a vehicle can be detected using zero-padding Fast Fourier Transform (FFT). However, a zero-padding FFT error is occurs due to DC-offset problem in automotive UWB radar sensor using a direct conversion receiver. Therefore, dc-offset problem corrupts velocity ambiguity. In this paper we proposed a mean-padding method to reduce zero-padding FFT error due to DC-offset in automotive UWB radar using direct conversion receiver, and verify our proposed method with computer simulation and experiment using developed automotive UWB radar sensor. We present the simulation results and experiment result to compare velocity measurement probability of the zero-padding FFT and the mean-padding FFT. The proposed algorithm simulated using Matlab and experimented using designed the automotive UWB radar sensor in a real road environment. The proposed method improved velocity measurement probability.