Reliability-Aware Neural Decoding with Adaptive Multi-Source Information Fusion

TL;DR

This paper introduces a neural decoder that automatically adjusts to the reliability of different information sources in communication systems.

Contribution

A neural decoder with adaptive fusion of multi-source information using a learnable gating module and continuous injection strategy.

Findings

The decoder demonstrates Bayesian-like behavior by adjusting reliance on statistical models based on uncertainty.

A continuous injection strategy maintains auxiliary information quality in deep architectures.

The approach improves performance and robustness when auxiliary information degrades.

Abstract

Modern communication systems increasingly leverage multiple information streams—including channel observations, statistical models, and contextual knowledge—to enhance decoding reliability. However, the varying and often unpredictable quality of these sources poses a critical challenge: rigid combination rules fail when source reliability fluctuates, while manual tuning cannot adapt to dynamic operating conditions. This paper presents a neural decoder architecture that automatically learns to assess and fuse heterogeneous information sources based on their instantaneous reliability. Central to our design is a learnable gating module that dynamically weights information streams, demonstrating emergent Bayesian-like behavior—increasing reliance on statistical models under high uncertainty while transitioning to observation-dominated processing as signal confidence improves. To combat the progressive dilution of auxiliary information in deep architectures, we propose a continuous injection strategy that refreshes auxiliary features at each processing layer through dedicated encoding pathways. The underlying message-passing network adopts a heterogeneous bipartite structure with direction-dependent edge parameterization, respecting the asymmetric computational roles inherent in iterative decoding algorithms. Comprehensive experiments validate that the proposed approach not only improves nominal performance but critically maintains robustness when auxiliary information quality degrades or becomes mismatched with actual conditions.

Figures