A batch back-propagation neural networks (BBPNN) approach was presented to design general two-dimensional (2-D)
quasi-equripple zero-phase finite-impulse response (FIR) digital filters. By minimizing the frequency-domain weighted
error function, the BBPNN design method was obtained. The solution was presented as a parallel algorithm to approximate
the desired frequency response specification. Thus, the method makes a fast calculation of the filter's coefficients
possible. It is shown that the method leads to an optimal solution for the filter coefficients. The implementation of the
approach was described together with some design guidelines, and some optimal design examples were given to demonstrate
the effectiveness of the proposed approach.
An equiripple FIR linear-phase digital filters design approach is proposed based on a novel neural network optimization technique. Its goal is to minimize the weight square-error function in the frequency domain. The design solution is presented as a parallel algorithm to approximate the desired frequency response specification, and the weight coefficients are updated according to error function. Thus, the proposed approximation method can avoid the overshoot phenomenon which may happen near the pass-band and stop-band edge of the designed filter, and may make a fast calculation of the filter's coefficients possible. Several optimal design examples are given and the performance comparison between the proposed design approach with some conventional methods, and the results show that the proposed neural network method can easily achieve higher design accuracy.
Two-dimensional (2-D) digital filters are widely useful in image processing and other 2-D digital signal processing fields,but designing 2-D filters is much more difficult than designing one-dimensional (1-D) ones.In this paper, a new design approach for designing linear-phase 2-D digital filters is described,which is based on a new neural networks algorithm (NNA).By using the symmetry of the given 2-D magnitude specification,a compact express for the magnitude response of a linear-phase 2-D finite impulse response (FIR) filter is derived.Consequently,the optimal problem of designing linear-phase 2-D FIR digital filters is turned to approximate the desired 2-D magnitude response by using the compact express.To solve the problem,a new NNA is presented based on minimizing the mean-squared error,and the convergence theorem is presented and proved to ensure the designed 2-D filter stable.Three design examples are also given to illustrate the effectiveness of the NNA-based design approach.