Reliability-based Error Detection for Feedback Communication with Low Latency

TL;DR

This paper introduces a reliability-based decoding scheme for feedback communication that achieves higher rates than existing bounds by using the ROVA algorithm to decide when to stop transmission based on error probability, requiring minimal feedback.

Contribution

It proposes a novel decoding scheme utilizing ROVA for variable-length feedback coding that surpasses previous theoretical limits without additional CRCs.

Findings

Achieves higher rates than Polyanskiy et al.'s lower bound on BSC and AWGN channels.

Uses minimal feedback: a single bit indicating low error probability.

Eliminates the need for rate-reducing CRCs in decoding process.

Abstract

This paper presents a reliability-based decoding scheme for variable-length coding with feedback and demonstrates via simulation that it can achieve higher rates than Polyanskiy et al.'s random coding lower bound for variable-length feedback (VLF) coding on both the BSC and AWGN channel. The proposed scheme uses the reliability output Viterbi algorithm (ROVA) to compute the word error probability after each decoding attempt, which is compared against a target error threshold and used as a stopping criterion to terminate transmission. The only feedback required is a single bit for each decoding attempt, informing the transmitter whether the ROVA-computed word-error probability is sufficiently low. Furthermore, the ROVA determines whether transmission/decoding may be terminated without the need for a rate-reducing CRC.