Sidelobes Suppression For Polyphase Code Radar Signals Based On Wavelet Neural Network

The pulse compression is an advanced technique widely applied in the modern Radar. It has effectively solved the contradiction between the radar range resolution and the average power. The phase coded singals are commonly used in pulse compression. And with the development of the digital and DSP technology, people began to seek the signal with greater pulse compression radio, therefore the polyphase coded signal received wide attention. However, their range sidelobes are too high and inadequate for most applications. So, how to minimize the peak range sidelobe of a phase coded waveform is a key to practical applications of phase coded pulse compression.This article mainly introduced and has analyzed a method of sidelobes suppression for radar polyphase code signal using wavelet neural network. First, introduce some characteristic of polyphase coded signal such as ambiguity function and echo, and introduced some polyphase coded such as Taylor code, frank code, P3, P4 code and so on. Second, intrduced two common used method of sidelobes suppression for polyphase coded singal, windowing weighted sidelobe suppression technology and least squares mismatched filtering algorithm. Combined with the principle of the two methods, we carried on the corresponding simulation. Third, we study the theory of the wavelet and the learning methed of the wavelet neural network, focusing on BP algorithm besed on the wavelet neural network and put forward several improved BP algorithm. Last, using the improved BP network for suppressing sidelobes of the polyphase coded signal, and test its resistance to noise, Doppler tolerance and ability of resolution. The result show that the wavelet network sidelobes suppression method has better characteristic of sidelobe suppression and Doppler tolerance than windowing weighted sidelobe suppression technology and iterative least squares mismatched filtering algorithm.