文档介绍：Mathematical puter Modelling 47 (2008) 982–996
ate/mcm
FPGA implementation of a wavelet work with particle
swarm optimization learning
Cheng-Jian Lina,∗, Hung-Ming Tsaib
a Department of Electrical Engineering, National University of Kaohsiung, Kaohsiung, Taiwan 811, ROC
b Institute working munication Engineering, Chaoyang University of Technology, 168 Gifong E. Rd., Wufong, Taichung County,
413, Taiwan, ROC
Received 14 October 2006; received in revised form 9 February 2007; accepted 22 May 2007
Abstract
This paper introduces implementation of a wavelet work (WNN) with learning ability on field programmable gate
array (FPGA). A learning algorithm using gradient descent method is not easy to implement in an electronic circuit and has local
minimum. A more suitable method is the particle swarm optimization (PSO) that is a population-based optimization algorithm.
The PSO is similar to the GA, but it has no evolution operators such as crossover and mutation. In the approximation of a nonlinear
activation function, we use a Taylor series and a look-up table (LUT) to achieve a more accurate approximation. The results of the
two experiments demonstrate the essful hardware implementation of the wavelet works with the PSO algorithm using
FPGA. From the results of the experiment, it can be seen that the performance of the PSO is better than that of the simultaneous
perturbation algorithm at sufficient particle sizes.
c 2007 Elsevier Ltd. All rights reserved.
Keywords: Wavelet works (WNN); Field programmable gate array (FPGA); Particle swarm optimization (PSO); Prediction; Identification
1. Introduction
In recent years, artificial works (ANN) have been widely used in many fields, such as chemistry [1],
diagnosis [2], control [3], and image processing [4]. The monly used artificial work is the multi-
layer perceptron (MLP) proposed by Gibson et al. [5].The MLP is a full connection structure that uses sigmoid
functions as hidden node functions. MLP equ