Design of FIR digital filters in complex domain by complex-valued neural networks

Complex-valued neural networks, whose inputs, weights, and outputs are all complex numbers, can be used in signal processing fields such as communication, speech recognition, and image processing. In this paper, we propose an algorithm based on complex-valued neural networks for the design of complex-coefficient FIR digital filter with arbitrary amplitude-frequency response. Examples of optimized design of the FIR low-pass filter and band-pass filter based on the complex-valued neural networks are also presented. Simulation results show that the complex-valued neural network is more natural and effective for the design of FIR filter, has higher training speed and has stronger generalization ability, than real-valued neural networks.