Iterative Soft-Input Soft-Output Decoding with Ordered Reliability Bits GRAND

Abstract

Guessing Random Additive Noise Decoding (GRAND) is a universal decoding algorithm that can be used to perform maximum likelihood decoding. It attempts to find the errors introduced by the channel by generating a sequence of possible error vectors in order of likelihood of occurrence and applying them to the received vector. Ordered reliability bits GRAND (ORBGRAND) integrates soft information received from the channel to refine the error vector sequence. In this work, ORBGRAND is modified to produce a soft output, to enable its use as an iterative soft-input soft-output (SISO) decoder. Three techniques specific to iterative GRAND-based decoding are then proposed to improve the error-correction performance and decrease computational complexity and latency. Using the OFEC code as a case study, the proposed techniques are evaluated, yielding substantial performance gain and astounding complexity reduction of 48% to 85% with respect to the baseline SISO ORBGRAND.