Low Snr Of Linear Fm Signal Detection And Parameter Estimation Study

As a mature LPI radar signal , linear-frequency modulated (LFM) signal is widely used in various radar systems. How to detect the LFM signal and estimate the signal parameters in low signal-to-noise rate is very important in modern Electronic Warfare. In this dissertation, the performances of some detection methods based on typical time-frequency analysis theories are researched and compared, and two new methods about LFM parameters estimation are proposed.Starting with the discussions about LFM signal, which is one of the typical non-stationary signals ,the general situation of development about the researches and applications of time-frequency analysis theories and technologies is studied. The performances of various time-frequency analysis methods about the LFM signal detection plus noise are analyzed via experiments. Because of the linearity of LFM signal in time-frequency plane and the feature of detecting straight line of Radon transform in image processing, the methods of combining Radon transform with some typical time-frequency analysis methods, such as short-time Fourier transform, wavelet transform, Wigner-Ville distribution and ambiguity function , can well detect LFM signal embedded in noise. The performances of above detection methods are analyzed and compared, and the advantages and deficiencies are indicated based on experiments.After introducing the theory of maximum likelihood estimation of LFM signal, two new algorithms of fast detection and parameters estimation based on dechirp are presented . The algorithms which have low computational complexity are considerably simpler to implement than maximum likelihood estimatition , and the effectiveness of them can be proved by computer simulation.At last, the conclusions are drawn , and some further research about LFM signal detection and parameter estimation is pointed out.