Blind Equalization Algorithm Based On Adaptive Double-stranded DNA Genetic Algorithm

Due to the distortion caused by complicated communication channel and inter-symbol interference caused by limited bandwidth affect the quality of communication seriously in wireless communication, so those efforts need to be compensated properly on the receiving end of communication system. Because blind equalization technology does not require sending training sequence and save bandwidth, it can overcome the effectively inter-symbol interference and improve communication speed. Blind equalization technology has become a research hotspot in the field of signal processing.Aiming at the defects of traditional blind equalization algorithm such as slow convergence speed, large steady-state error, this paper optimizes the performance of blind equalization algorithm by using DNA genetic algorithm and orthogonal wavelet transform. The research content includes mainly following several aspects:(1) Constant modulus blind equalization algorithm based on DNA genetic algorithm is proposed. Traditional blind equalization algorithm use the steepest descent method to update equalizer weight vector, the steepest descent method, however, require cost function must be continuous and differentiable, and easy to fall into local extremes, leading to slow convergence speed, large steady-state error. Therefore, for those short comes, using global search ability of DNA genetic algorithm to optimize the equalizer weight vector can avoid local convergence and improve the performance of blind equalization algorithm. The simulation results show the effectiveness of this algorithm.(2) Blind equalization algorithm based on tabu search & adaptive double-stranded DNA genetic algorithm. Because the probability of crossover operation can not be adjusted following population change in general DNA genetic algorithm, so the search efficiency of traditional DNA genetic algorithm is not high. In addition, while DNA genetic algorithm search for the optimal solution to the problem, sometimes repeated search may appear, which that also affect the searching efficiency of the algorithm. So, for these shortcomings, blind equalization algorithm based on tabu search & adaptive double-stranded DNA genetic algorithm is formed by using tabu search algorithm and method that the probability of crossover operation change with population evolution.(3) Multi-modulus blind equalization algorithm based on tabu search & adaptive double-stranded DNA genetic algorithm. For the reason that traditional multi-modulus blind equalization algorithm can not equalize high order modulation signal effectively, tabu search & adaptive double-stranded DNA genetic algorithm is applied to the multi-modulus blind equalization algorithm to form multi-modulus blind equalization algorithm based on tabu search & adaptive double-stranded DNA genetic algorithm by using orthogonal wavelet transform. Because tabu search & adaptive double-stranded DNA genetic algorithm has faster searching efficiency and better local searching feature, therefore, the DNA genetic algorithm is applied to multi-modulus blind equalization algorithm can significantly improve the performance of MMA.(4) Orthogonal wavelet transform fractionally spaced multi-modulus blind equalization algorithm based on multiple population & tabu search DNA genetic algorithm. In order to improve search efficiency of DNA genetic algorithm and equalization performance of orthogonal wavelet transform fractionally spaced multi-modulus blind equalization algorithm(WTFSE-MMA), this chapter improve DNA genetic by using tabu search algorithm and multiple population evolution strategy. Therefore, orthogonal wavelet transform fractionally spaced multi-modulus blind equalization algorithm based on multiple population & tabu search DNA genetic algorithm is proposed by applying improved DNA genetic algorithm to WTFSE-MMA. DNA population is divided into multiple subpopulations in this algorithm, and each subpopulation focuses on different search purposes, that are the main population focus on local search and auxiliary population focus on global search. Besides, tabu crossover operations and different mutation operations are used in different subpopulations, which improves search ability of DNA genetic algorithm and performance of orthogonal wavelet transform fractionally spaced multi-modulus blind equalization algorithm.