In-Memory Computing of Akers Logic Array

Abstract

This work studies memories from a different perspective, while the goal is to explore the concept of in-memory computing. Our point of departure is an old study of logic arrays by Akers in 1972. We demonstrate how these arrays can simultaneously store information and perform logic operations. We first extend the structure of these arrays for non-binary alphabets. We then show how a special structure of these arrays can both store elements and output a sorted version of them. We also study other examples of the in-memory computing concept. In this setup, it is shown how information can be stored and computed with, and the array can tolerate or detect errors in the stored data.