Abstract:

The parameter estimation of chirp signals contaminated by additive Gaussian white noise is investigated, and a phase domain estimation algorithm is presented. Firstly, the initial frequency and the frequency rate are estimated coarsely, then the received signals are transformed to baseband signals, which contain the residual error of the initial frequency and the frequency rate. Consequently a phase unwrapped algorithm is performed to the baseband signals. The estimates of the residual error are extracted by least square fitting to the unwrapping phase. Simulation shows that when the SNR is higher than the threshold SNR, the accuracy of the algorithm is close to Cramer-Rao lower bound (CRLB) and is more accurate than that of discrete polynomial transform (DPT) based on discrete Fourier transform (DFT). Its threshold SNR is lower than that of conventional phasedomain methods.