Characteristic frequency extraction of HVAP Based on wavelet analysis

Abstract

Bad shunt of track circuit is a common problem in railway signal system all over the world, the High Voltage Asymmetric Pulse (HVAP) track circuit is an effective solution to it. How to extract the characteristic frequency of HVAP is an important task. The Fast Fourier Transform (FFT) can analyze only stationary signals at single resolution. Wavelet analysis is an algorithm of multiresolution approximations, it can decompose a signal into high-frequency and low-frequency compositions. If the HVAP signal contains transient noises, its characteristic frequency can not be extracted using FFT. This paper proposes an algorithm of extracting the characteristic frequency of HVAP signal which contains transient noises based on wavelet analysis. And computer simulations show that this algorithm has not only anti-interference ability, but also high precision.