The parameter estimation of frequency-hopping signals via l1-SPICE

Frequency-hopping (FH) signals are widely applied in the commercial and military fields because of good robustness to interference. However, even with the best parameter estimation of FH signals, there will be a heavy computational burden. The cycle diagram method which is widely used in the estimation of FH signals is very rough and unreliable, while they limit the estimation performance. In this paper, we present an adaptive estimation method for the sparse spectrum based on the l1-norm sparse iterative covariance-based estimation (SPICE), namely l1-SPICE. l1-SPICE is used in the parameter estimation of FH signals, the process of which is then compared with the traditional short-time Fourier transform (STFT) and the conventional SPICE. Simulation results show that the proposed method is better than STFT and SPICE as regards the aspects of frequency resolution, sparse feature and accuracy, and this which makes l1-SPICE a good choice in respect of hopping time estimation and chip period estimation in FH signals.