Research On The Improvement Of Squirrel Search Algorithm And Its Application In Blind Source Separation

Optimization problems are widely present in scientific research and practical life.With the development of information technology,optimization problems have become increasingly complex and finding effective solutions has become more and more important.The swarm intelligence algorithm is an effective solution to optimization problems,and with the development of artificial intelligence technology,new types of swarm intelligence algorithms are constantly proposed and applied.The squirrel search algorithm(SSA)is a well-performing swarm intelligence algorithm proposed in 2018.By simulating social behaviors such as the dynamic foraging and gliding of a group of squirrels,a model is established to solve the complicated global optimal problems.However,the search capability of the SSA is limited,and it often requires a great many of iterations to achieve satisfactory accuracy.There is still much space for improvement in its global exploration ability and local exploitation ability.At present,the research and application of the SSA still in its infancy,much work remains to be expanded.Blind source separation is an important research direction of information processing technology.Independent component analysis(ICA)is an important method to deal with the blind source separation problem.Most traditional ICA algorithms have defects such as excessive reliance on gradient information,easy to fall into local optima,and the quality of the solution cannot be guaranteed.The optimization process of swarm intelligence algorithm does not rely on gradient information,which can make up for the shortcomings of traditional ICA algorithms.This article focuses on the research of the SSA,improves its defects,and applies it to solve the problem of blind source separation.The main content and innovations of the article are as follows:(1)This paper proposes a squirrel search algorithm with an enhanced search capability(ESSA).First,the global exploration ability of the algorithm is enhanced by employing an opposition-based learning strategy for general individuals that meet the seasonal monitoring conditions.Second,the local exploitation ability of the algorithm is enhanced by a chaotic search around the better individuals of each iteration.To verify the efficiency of the proposed algorithm,a set of 32 test functions,including unimodal,multimodal,CEC2013 test functions,and 3 classical engineering optimization problems,are selected to test the proposed ESSA.Experimental results show the greatly improved convergence accuracy and speed of the ESSA compared with those of the SSA and that its performance is better than that of other compared algorithms.More accurate solutions can be obtained by the ESSA with fewer iterations,and it can efficiently solve practical engineering optimization problems.(2)The ESSA is applied to solve the blind source separation problem.An independent component analysis algorithm based on ESSA(ESSA-ICA)is proposed to overcome the defects of traditional ICA algorithms,such as excessive dependence on gradient information and easily fall into local optimum.The ESSA-ICA takes the sum of the absolute value of the kurtosis as the objective function,and combines the advantages of ESSA to optimize the objective function,so as to improve the performance of the ICA algorithm and realize blind separation of mixed signals.Simulated separation experiments were performed on the mixed image signal and the mixed speech signal.The results show that the ESSA-ICA algorithm can effectively separate each source signal,which is an effective blind source separation algorithm.Its separation performance has a greater improvement than that of the traditional ICA algorithm,and better than that of other intelligent algorithms.ESSA-ICA can deal with the problem of blind source separation well.The research in this paper has enriched the theoretical basis of the SSA,enhanced the search capability of the algorithm,and broadened the application range of the algorithm.It has certain theoretical value and practical significance for the development of the SSA.