Detection and estimation of chirp signals in non-Gaussian noise

Detection of chirps in (non-Gaussian) additive and multiplicative noise is explored via a novel cyclostationary approach. A least-squares method for parameter estimation in colored noise is proposed and asymptotic bounds are derived on the estimation error. Cramer-Rao lower bounds are developed for chirp parameter estimation if the multiplicative and additive noise is Gaussian. Finally, simpler parameter estimation algorithms that only require linear searches are proposed as alternatives to the computationally complex optimal schemes