Multi-mode Blind Equalization Algorithm with Dynamic Momentum Factors and Conjugate Gradient
Yanping Zhang · Journal of Henan University of Science & Technology · 2013
In order to overcome the slow convergence rates of the traditional multi-mode algorithm(MMA),a multi-mode blind equalization algorithm with the dynamic momentum factors and conjugate gradient was proposed.In this proposed algorithm,the conjugate gradient and the momentum term were introduced into the multi-mode algorithm to get the iterative formula of the momentum term and conjugate gradient based on multimode blind equalization algorithm.Through analyzing the dynamic momentum factor and introducing the deformation form of exponential function to construct a non-linear function between the momentum factors and the error,the function uses the decreasing of error to decrease momentum factor adaptively to achieve the purpose of speed up the convergence rate and reduce the steady-state error.Theoretical analysis and simulation results show that the proposed algorithm compared with the MMA and conjugate gradient muti-mode algorithm can improve the convergence rate,and reduce the steady-state mean square error.