A Stable DOA Tracking Method Using FDPM and Kalman Filter

To solve the problems that eigenvalue decomposition of the covariance matrix is computationally intensive and the target moving can cause a spatial spectrum spread in the GNSS (global navigation satellite system) interference detection and localization system using antenna array, in this paper we propose a low-complexity GNSS signal DOA tracking algorithm using a Kalman filter implementation. This method firstly utilizes an improved FDPM (fast data projection method) for noise subspace tracking, and sets the DOA estimation predicted by Kalman filter as the initial value of Newtown iteration, then using the iteration result as a new DOA observation the sate vector of the Kalman filter is updated and calibrated, therefore a stable and effective DOA tracking loop is achieved. Simulation results show that using the method the performance of DOA tracking can be significantly improved while the computational burden is reduced simultaneously.