Fast-Converging Conjugate-Gradient Adaptive Algorithm For Complex Channel Estimation

An efficient adaptive digital filtering algorithm called complex block conjugate least mean square (CBC-LMS) is presented for complex wireless channel estimation. The proposed technique incorporates the conjugate gradient principle into the complex block LMS algorithm. In addition, the novel algorithm generates a unique optimal stepsize along each conjugate direction at each iteration. Computer simulations show that the proposed CBC-LMS yields better performance than existing complex block LMS algorithms, while maintaining computational efficiency. © 2010 The Institution of Engineering and Technology