含噪BSS模型的BJD-Hough算法*

当噪声存在时，信号和混合阵的盲估计变得相当困难。针对信号源噪声污染情形，假设信号和噪声的时频谱不同，提出了一种时频去噪盲源分离方法。该方法以Born-Jordan分布计算混合信号的时频矩阵并将信号的时频分布看作图像，利用广义Hough变换将信号检测转换为在参数空间寻找局部极大值的问题，再运用自项点理论选择合适的时频阵进行对角化，进而估计源信号和混合阵。该方法扩展了盲源分离的限制条件，能有效分离各种非平稳源信号、非独立源信号，且通过把噪声能量扩展到整个时频面而只选择信号能量占主导的时频点，对噪声具有一定的抑制能力。

BJD-Hough algorithm for noisy BSS model

The blind estimation of signals and mixed matrix become very difficult when noise exists. This paper introduces a time-frequency denoising blind source separation method via assuming different time-frequency localization properties between signals and noise. The method firstly calculated the time-frequency matrix of mixed-signal by Born-Jordan distribution, and used the generalized Hough transform to convert the signals detection to find the local peak values in the parameter space though considering the time-frequency distribution as an image. Then, this method diagonalized a combined set of time-frequency distributions chosen by auto-term theory and therefore estimated source signals and mixing matrix. This method extends the blind source separation constraints, can effectively separate the various non-stationary sources, non-independent source and the effect of spreading the noise power while localizing the source energy in the time-frequency domain amounts to increasing the robustness of the proposed approach with respect to noise.