Velocity measurement of the automobile based on the wavelet de-noising and STFT

Abstract

This paper mainly deals with the velocity measurement of automotive millimeter wave radar. The echo from the noisy traffic environment often makes it difficult for the traditional frequency spectrum analysis and the complex time-frequency processing methods to accurately and briefly represent the automotive velocity. A new method based on the wavelet de-noising and the short-time Fourier transform is proposed here to extract desired signal from interference and give a new time-frequency spectrum of the Doppler frequency, and calculate the moving velocity effectively. Higher signal to clutter ratio and higher resolution in the time-frequency distribution are obtained via the computer simulation and the actual gathered signal processing, and the velocity is also measured effectively. The method may widely apply to the velocity measurement of the moving goals of projectiles, ships and so on.