Signal MP Sparse Decomposition Based On Artificial Fish-Swarm Algorithm

The signal sparse decomposition is emerging as a new method for analyzing and processing signals, which has many excellent characteristics. And signal sparse decomposition based on Matching Pursuit (MP) is used commonly in signal sparse decomposition. It has been applied to many areas such as data compression, signal feature extraction, time-frequency analysis and etc.. But it is a NP difficult problem. The large computational cost is the bottleneck of sparse decomposition.In recent years, swarm intelligence algorithm has attracted lots of researches of numerous science areas, which has higher adaptability, robustness, parallel and global quality etc., and are used widely in some areas such as function optimization, pattern recognition and image processing etc. This article studies the application of a new swarm intelligence algorithm-Artificial Fish-Swarm Algorithm (AFSA), in solving signal sparse decomposition and the related application. This evolution algorithm can overcome local extremum and get some global extremums, initial values can be chosen random and carrying out this algorithm doesn’t need to know gradient value of the objective function, so it has robustness for the searching area.The paper starts with the signal sparse decomposition and Matching Pursuit algorithm. Afterwards Artificial Fish-Swarm Algorithm is introduced. It presents systemic expatiation and study about the principle, structure, astringency and implements methods of the algorithm. And an improved artificial fish-Swarm algorithm is proposed. In the prey action of artificial fish, the pull means is added which can collect more artificial fish near the global extremum. The Improved algorithm not only retains the precision of original algorithm, but also advances the efficiency and convergent speed. Then the improved Artificial Fish-Swarm Algorithm is applied to the signal sparse decomposition based on MP, and the simulation results and analysis are given. It shows that the improved Artificial Fish-Swarm Algorithm can fast search for approximately optimal atom at each step of MP and the quantum of computing is reduced a lot. In addition, it is inevitable to be interfered by a large amount of noise signal in the process of signal gathering and transmission. So it’s essential to denoise and extract original signal. At the end of the article, dealing noised signal with signal MP sparse decomposition based on Artificial Fish-Swarm Algorithm has made certain success in improving the ratio of signal to noise and denoising.