Performance evaluation of digital and RNS based filter for fast DSP processors

In the present scenario, the next generation DSP processors are critically desired for fast digital computations. Digital filters are the basic elementary units of DSP processor, which plays a very crucial role in digital signal processing. The performances of the DSP processors are majorly factored by their speed of arithmetic computation. Due to large carry propagation delay and sequential behavior, conventional digital FIR filters are sluggish in nature. The solution to this problem is conquered by Residue Number System (RNS) based digital filters. The RNS is carry free in nature and also provides parallelism, which makes it suitable for high speed computation. In this paper, the performance of RNS based digital filter has been analyzed for 15 tabs low-pass FIR filter on the basis of area, power, and delay. The functional verification is done on Xilinx ISE suite 14.7 and simulation is performed on Synopsis tool with 90nm technology file. This reveals the improvement in terms of area, power, and delay. The results have also been compared and contrasted with the available conventional digital filter at the same platform.