Low complexity and narrow transition band filter banks for software defined radio applications

Abstract

We presented an improved efficient architecture of low complexity and narrow transition band filter banks for software defined radio applications. A drawback in filter banks is that the narrow transition band filter is designed with high computation complexity. In order to get around this problem, this paper shows how the filter banks architecture can be designed with low complexity by using frequency response masking technology. We derives efficient architecture of exponentially modulated filter banks based the theory of polyphase, and uses frequency response masking technology to design narrow transition band filter. With employed this method we can extract uniform bandwidths and very narrow bandwidth channels compared to polyphase filter banks. A simulation is provided to illustrate the method of the proposed filter banks architecture. It is shown that the resulting filter banks entail substantially less computational complexity compared to the polyphase filter banks.