A modified constrained stability-least mean square algorithm and its application in chaos communication

The constrained stability-least mean square (CSLMS) provides faster convergence speed and higher flexibility than the normalized least mean square (NLMS) by minimizing the Lagrangian function under a constraint stability update condition. In this paper, a modified CSLMS is proposed by adaptively adjusting step size according to a nonlinear function based on error signal. The modified CS-LMS is then applied to perform chaos communication. Simulation results show that the proposed novel CS-LMS can effectively recover information signals, and outperform the other LMS algorithms.