GPS nonstationary interference suppression based on sparse representation

For GPS (Global Position System) receivers, the nonstationary interferences may result from rapid relative motion between interference and GPS receiver, a change in the number of interferences and impulse interference, etc. The conventional interference suppression algorithms based on covariance matrix are ineffective with nonstationary interferences. By utilizing the sparse representation, a new kind of nonstationary interference suppression algorithm is proposed in this letter. In particular, a greedy sparse interference space reconstruction algorithm is utilized to directly estimate the received interferences instead of DOA (Direction of Arrival) estimation. The interferences are suppressed by subtracting the estimated interferences for every snapshot without covariance matrix estimation. There are two typical advantages: (1) the proposed algorithms do not require the number of interferences; (2) the proposed algorithms are robust for both DOA separation of interferences and grid mismatch of sparse representation. The simulated results demonstrate the effectiveness of the proposed algorithms.