Rich-ARQ: From 1-bit Acknowledgment to Rich Neural Coded Feedback

TL;DR

Rich-ARQ introduces neural-coded, high-dimensional feedback in wireless communication, enabling active collaboration between transmitter and receiver, leading to significant SNR improvements and latency reductions in practical over-the-air tests.

Contribution

This work pioneers neural-coded feedback in ARQ, developing a novel asynchronous feedback code and a full-stack SDR prototype for practical, high-performance wireless communication.

Findings

Achieves significant SNR gains over traditional hybrid ARQ.

Reduces latency compared to prior learning-based feedback codes.

Demonstrates practical implementation with over-the-air experiments.

Abstract

This paper reimagines the foundational feedback mechanism in wireless communication, transforming the prevailing 1-bit binary ACK/NACK with a high-dimensional, information-rich vector to transform passive acknowledgment into an active collaboration. We present Rich-ARQ, a paradigm that introduces neural-coded feedback for collaborative physical-layer channel coding between transmitter and receiver. To realize this vision in practice, we develop a novel asynchronous feedback code that eliminates stalling from feedback delays, adapts dynamically to channel fluctuations, and features a lightweight encoder suitable for on-device deployment. We materialize this concept into the first full-stack, standard-compliant software-defined radio prototype, which decouples AI inference from strict radio timing. Comprehensive over-the-air experiments demonstrate that Rich-ARQ achieves significant SNR gains over conventional 1-bit hybrid ARQ and remarkable latency reduction over prior learning-based feedback codes, moving the promise of intelligent feedback from theory to a practical, high-performance reality for next-generation networks.