Research And Application Of Artihcial Fish School Algorithm In Blind Detection

Artificial fish school algorithm(AFSA) builds a fish school behavior by simulating the fish inthe water.It’s a specific application of the swarm intelligence and can be used for solvingoptimization problems.It doesn’t need to know the problem’s special information and only tocompare the result of the problem. Artificial fish school algorithm is a new swarm intelligenceoptimization algorithm of fast convergence,global search and adaptive abilities. It has been widelyused in optimization,signal processing,neural networks and intelligent control. For the influence ofrecurring fading channel,nonlinear and time-varying characteristics and multipathtransmission,blind detection technique doesn’t require training sequences and just uses the receivedsequences’ prior knowledge,so it will be able to balance channel characteristics adaptivelyand.complete the estimate of the best signal.It’s a new technology of blind signal processing.It hasbecome a hot research topic in the field of communication.This technique can be effective incompensating the non-ideal characateristics of the channel,overcoming the inter-symbolinterference(ISI),reducing the error rate and improving the quality of communication.The first chapter of the paper gives an overview of the research background,the developmentof swarm intelligence algorithm and the work this paper did.The second chapter describes theprinciple,application and directions of AFSA.It analyzes the various behavior of the fish school aswell as the influence of parameters to the performance of the algorithm.The third chapter describesthe principle of blind equalization and its implementation method.Some classic blind equalizationalgorithm are mentioned.The forth chapter is the focus of the paper.In this chapter it proposes blinddetection algorithm based on AFSA,at the same time,it gives the related simulation to theperformance of the proposed algorithm.The results of simulation show that the swarm intelligenceblind detection algorithm proposed in this paper has several advantages,like lower error rate,fastconvergence speed,stable performance.It can solve the problem of blind signal processingeffectively.