The application of core functions to residue number system signal processing

The authors briefly outline core-based residue number system (RNS) processing. An RNS adaptive array processor using Choleski decomposition and forward elimination/back substitution is described as an example. The RNS adaptive processor was simulated, and preliminary results that indicate the feasibility of the approach are presented. The most significant of these results is the ability of the eight 5-b moduli system to operate well with both 12- and 32-element arrays. Because the core function is used to perform nonmodular operations, a large number of array elements can be accommodated with a modest dynamic range. Potential applications for special-purpose core-based RNS processors include adaptive array processing, Kalman filtering, fast Fourier transforms, and image processing for communications, surveillance, and intelligence systems.<
>