Study on detection of small-sized range-spread target by using chaotic Duffing oscillator

Abstract

A chaotic detection scheme for a small-sized rangespread target in white Gaussian noise is developed. The basic idea is that a chaotic oscillator system is less influenced by noise, but it is very sensitive to periodic or pseudo-periodic signals. Firstly we concatenate the echoes with the same interested range cells from three successive pulses transmitted by high resolution radar end to end into a time series. The time series has certain pseudo-periodicity. Then the pseudoperiodic signal generated artificially is introduced into a Duffing oscillator system which is in the critical chaotic phase state. By adjusting the intensity of the noisy signal according to the estimated noise power, the false alarm ratio can be kept within an acceptable range. Then we determine the presence or absence of a target by judging the negative or positive maximal Lyapunove exponent of the Duffing equation. The Monte Carlo simulations for a small-sized range-spread target show that the proposed method significantly outperforms the traditional integrator detector and can improve about 2-5 dB in signal-to-noise ratio performance versus the integrator.