Multi-Factor Pruning for Recursive Projection-Aggregation Decoding of RM Codes

TL;DR

This paper introduces a multi-factor pruning technique that significantly reduces the computational complexity of recursive projection aggregation decoding for Reed-Muller codes while maintaining near-ML performance.

Contribution

The paper proposes a novel multi-factor pruning method that cuts down RPA decoding complexity by up to 92% without sacrificing error correction accuracy.

Findings

Complexity reduced by up to 92% for RM(8,3)

Maintains comparable error-correcting performance

Effective for resource-critical applications

Abstract

The recently introduced recursive projection aggregation (RPA) decoding method for Reed-Muller (RM) codes can achieve near-maximum likelihood (ML) decoding performance. However, its high computational complexity makes its implementation challenging for time- and resource-critical applications. In this work, we present a complexity reduction technique called multi-factor pruning that reduces the computational complexity of RPA significantly. Our simulation results show that the proposed pruning approach with appropriately selected factors can reduce the complexity of RPA by up to $92\%$ for $\text{RM}(8,3)$ while keeping the comparable error-correcting performance.