该文首先基于Oja定义的非线性PCA准则J1(W),利用矩阵广义逆递推得到一种盲信号分离算法,然后对Karhunen给出的非线性PCA加权误差平方和准则J2(W),采用梯度下降算法和线性寻优而得到另一种自适应盲信号分离算法。对这两个分离算法进行了计算机仿真,仿真结果表明它们的有效性。
Firstly,based on the nonlinear PCA criterion J1 (W)defined by Oja,We propose a blind signal separation algorithm by using recursion of generalized inverse matrix.Then,based on the nonlinear PCA criterion J2(W) defined by Karhunen,We present the other blind signal separation algorithm by using gradient descent method and linear search optimization.Computer simulation results show that the proposed algorithms are effective.