Computer Arithmetic Preserving Hamming Distance of Operands in Operation Result

TL;DR

This paper introduces a novel fault tolerant computing method that preserves Hamming distances between inputs and outputs, enabling efficient correction of transient faults without extensive resource use.

Contribution

It proposes a new approach that maintains Hamming distances during computation, reducing resource requirements compared to traditional replication and voting methods.

Findings

Preserves Hamming distance between inputs and outputs during processing

Enables correction of up to transient faults per cycle

Analyzed cost and performance compared to existing methods

Abstract

The traditional approach to fault tolerant computing involves replicating computation units and applying a majority vote operation on individual result bits. This approach, however, has several limitations; the most severe is the resource requirement. This paper presents a new method for fault tolerant computing where for a given error rate, the hamming distance between correct inputs and faulty inputs as well as the hamming distance between a correct result and a faulty result is preserved throughout processing thereby enabling correction of up to transient faults per computation cycle. The new method is compared and contrasted with current protection methods and its cost / performance is analyzed.