2017 年 6 巻 2 号 p. 83-90
In this paper, a novel underdetermined blind source separation algorithm guided by particle swarm optimizer (PSO) is proposed for adjacent satellite interference in modern satellite communication systems. Different from traditional methods, we formulate the separation problem as clustering problem. Due to our algorithm is affected by the sparsity of source signals and the density of mixed vectors, our algorithm is motivated by the assumption is held that the distance between two arbitrary mixed signal vectors is less than the doubled sum of variances of distribution of the corresponding mixtures. In our method, we accomplish the underdetermined blind source separation by computing the Short Time Fourier Transform (STFT) to segment received mixtures and we use some estimates to separate the mixed source signals by PSO where the number of the mixed signals is unknown. In PSO, we define new parameters gather in formula (8) and cj in formula (11). We verify the proposed method on several simulations. The experimental results demonstrate the effectiveness of the proposed method.
