Estimation of chirp signal parameters using state space modelization by incorporating spatial information

This paper considers the problem of estimating the parameters of chirp signals. Our approach is based on a non-linear state modelization of the signal where the state estimation is performed by the extended Kalman filtering (EKF). This approach exploits the spatial information provided by two sensors, which results in considering two different observations in the EKF equations. This leads to a double application of EKF. A first filter uses the direct signal observations provided by the first sensor. A second filter uses the second sensor measurements. The exact Cramer-Rao lower bounds (CRLB) are derived and simulation results are presented, illustrating the performance of the estimator and validating our CRLB analysis.