Parameter Estimation Of LFM-BPSK Hybrid Modulated Signals In Impulsive Noise Environment

Hybrid modulated signals combining binary phase shift keying(BPSK) with linear frequency(LFM) modulation have better performance than BPSK signals and LFM signals with lower probability of interception, superior Doppler properties than BPSK signals, and have been applied in various radar, micro detector and communication systems. Gaussian distribution satisfies the central limit theorem, and it is convenient applied in signal processing and analyzing. Therefore, the background noises are generally assumed to be Gaussian distributed in traditional signal processing. However, there exist intense impulsive characteristics in such realistic noise and clutter as lightning, sea clutter and low-frequency atmospheric noise, and alpha stable distribution is a model that is effective for these complicated noises. In electronic countermeasures(ECM), an important mission for electron reconnaissance and electronic jamming is to capture and analyze the parameter information of radar signals of the enemy. Therefore, a systematic study on estimation methods of such parameters of LFM-BPSK signals as initial frequency, chirp rate and symbol rate in alpha stable noise is conducted. The author’s major contributions are outlined as follows:1. In order to solve the problem that the existing parameter estimation algorithms for hybrid modulated signals will significantly degrade in alpha stable noise environment, an L-DFT filter based LSHT method which is combined with short time Fourier transform(STFT) and Hough transform is proposed; an LDCFT(L-DFT based DCFT) method for initial frequency and chirp rate based on L-DFT and discrete chirp Fourier transform(DCFT) is proposed. Simulation results show that the LDCFT method offers superior parameter estimation performance in impulsive noise environment.2. A research on symbol rate estimation in alpha stable noise environment is conducted. The modified L-DFT filter is introduced to solve the problem that the probability of appearance of pulses in cyclic autocorrelation is increased, and it can be extended as a filter for general bilinear transform; a novel Cauchy based maximum-likelihood estimator(CMLE) for symbol rate of PSK signals is proposed. The parameters of the timing offset and the symbol rate can be estimated simultaneously through this method.Windowed procedure is utilized in CMLE and the noise polluted PSK signal is divided into a timing offset window and multiple windows with certain width which are non-overlapping and synchronized in time domain, and only one symbol code is contained in each window; In alpha stable noise environment, the symbol in the window is utilized and a likelihood function based on Cauchy distribution is built, then the maximum-likelihood estimation of window width for the timing offset and the symbol rate can be achieved simultaneously. Simulation results show that the two proposed methods can suppress the alpha stable noise efficiently and offer superior parameter estimation performance.