A neural-like network approach to finite ring computations

Computation over finite rings using networks modeled after the general neural network approach is discussed. In this case, the neurons are arithmetic elements that have modulo operator characteristics, rather than the usual nonlinear, saturating characteristics of learning and associative memory neural network applications. Following an analysis of finite-ring arithmetic, a computing model based on an iterative, bit-level modulo reduction scheme is built, from which a basic operator is extracted. A corresponding subnet is designed to implement this operator, and its effectiveness is illustrated in two examples of computing finite-ring operations for residual number system computations.<
>