Source Signals' Number Estimation Based on Fuzzy Clustering in Blind Separation of BPSK Signals

Digital signals are used in modern communication, and it is very important to focus on it. Although there have existed a lot of algorithms of blind separation, algorithms for blind separation of digital signals or blind separation of finite characters set are still few. What's more, algorithms for estimating the number of source signals in the case don't appear. In this paper, it gives a novel algorithm for BPSK signals' blind separation. First, it can estimate the number of source signals according to the characteristics of sensor signals in no noise and noise circumstance respectively. Finally, the mixture matrix is estimated accurately by using the relations of sensor signals, which is a primary column transformation of the original mixture matrix. They can be corrected by introducing headers in the bit-streams and differently encoding them. The algorithms are shown simple and efficient in last simulations.