J4 ›› 2015, Vol. 42 ›› Issue (3): 129-134.doi: 10.3969/j.issn.1001-2400.2015.03.022

• Original Articles • Previous Articles     Next Articles

FH signal extraction using the improved symmetric  co-occurrence matrix

GU Pei;WANG Bin   

  1. (Institute of Information Systems Engineering, PLA Information Engineering Univ., Zhengzhou  450001, China)
  • Received:2013-12-27 Online:2015-06-20 Published:2015-07-27
  • Contact: GU Pei E-mail:grainme@hotmail.com

Abstract:

Aiming at the problem of removing the background noise and fixed-frequency interference in the blind detection of FH signals from the HF channel, this paper proposes an FH signals extraction algorithm based on the noise-sparse symmetric co-occurrence matrix. Firstly, we define the noise-sparse symmetric co-occurrence matrix in the direction of frequency and time to improve the calculation of the co-occurrence matrix. Secondly, we estimate the noise threshold based on the frequency-sparse symmetric co-occurrence matrix and then extract the FH signals from the time-sparse co-occurrence matrix with the threshold. Simulation results show that the algorithm can realize the blind extraction of the FH signal on the condition of a low SNR. Estimation of the threshold in background noise is more accurate and stable. The performance of FH signal extraction is better and algorithm is simple with less computation and is easy to apply in engineering.

Key words: detection of FH signal, time-frequency analysis, image processing, symmetric co-occurrence matrix