A Time-Frequency Information Based Method for BSS Output FH Signal Recognition

Frequency hopping (FH) communication signal is usually statistically independent from common jamming signals. Blind source separation (BSS) or independent component analysis (ICA) can be introduced to separate the useful FH communication signal from the jamming signals. Through the separation, the jamming signals are suppressed and the quality of the object signal is improved. However, BSS is suffered from the inherent permutation ambiguity, which makes it difficult to select object signal from the multiple separated signals. According to the frequency spectrum characteristics of FH signal, a time-frequency (TF) information based method for FH signal recognition is proposed. Firstly, the TF matrices of the separated signal and the expected FH signal are constructed respectively. Secondly, the correlation value between each separated TF matrix and the expected TF matrix is calculated. Finally, the separated signal which has the biggest correlation value is identified as the object FH signal. The proposed method can eliminate the permutation ambiguity of BSS, so as to make BSS more applicable. Simulation experiments are carried out to test the performance of the proposed method. The simulation results show that the proposed method can recognize the FH signal among the multiple separated signals effectively.