The Optimization Of Radial Basis Function Neural Network And The Application In Embedded System

Radial Basis Function a kind of Neural Network is a high-efficiency Neural Network, with its simple network structure, fast learning ability, good approximation performance, higher convergence speed advantage and widely used in pattern recognition, nonlinear function approximation and intelligent control, signal processing, etc. Although Radial Basis Function Neural Network's structure is simple, but its structure parameter determined has no unified method. Swarm intelligence algorithm is a new type of evolutionary algorithms; it is according to the biology scholars to natural ecological environment of long-term evolution, which traditional methods can not solve the complex structure optimization problems.In allusion to being difficult to determine the parameters of radial basis functions neural network (RBFNN), a new method on the parameters optimization of radial basis function neural network based on Shuffled Frog Leaping Algorithm (SFLA) is proposed. The parameters of the RBFNN compose a multidimensional vector. It is regarded as a frog in this algorithm to evolve. Then, according to the fitness function, the feasible sampling space is searched for the global optima, further more, we have improved the SFLA.The simulation test on nonlinear function approximation shows that the new method compares to GA and PSO has a less mean square error and has better approximation ability. RBFNN algorithm also combines with ARM9 embedded system. This optimization algorithm combines with ARM9 embedded system identification color, making the RBF neural network algorithm application.