Structured Superposition of Autoencoders for UEP Codes at Intermediate Blocklengths

2025-08-10

AI Summary

Training autoencoder (AE)-based unequal error protection (UEP) codes for medium-length codewords is computationally prohibitive due to high model complexity. Method: This paper proposes a structured superposition AE framework that decomposes the encoder/decoder into multiple parallelizable, modular AE sub-blocks. It integrates superposition coding with successive interference cancellation (SIC) decoding to explicitly model and flexibly control hierarchical reliability levels. Contribution/Results: Compared to conventional end-to-end AE architectures, the proposed framework significantly reduces training complexity and memory overhead while enabling scalable code-length design. Experimental results demonstrate consistent performance gains over the classical achievability bound of random superposition coding across diverse signal-to-noise ratios and code rates. The method exhibits superior robustness, generalization capability, and error-correction performance, particularly under practical channel conditions.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Search and Optimization: Learning to Search
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

Unequal error protection (UEP) coding that enables differentiated reliability levels within a transmitted message is essential for modern communication systems. Autoencoder (AE)-based code designs have shown promise in the context of learned equal error protection (EEP) coding schemes. However, their application to UEP remains largely unexplored, particularly at intermediate blocklengths, due to the increasing complexity of AE-based models. Inspired by the proven effectiveness of superposition coding and successive interference cancellation (SIC) decoding in conventional UEP schemes, we propose a structured AE-based architecture that extends AE-based UEP codes to substantially larger blocklengths while maintaining efficient training. By structuring encoding and decoding into smaller AE subblocks, our method provides a flexible framework for fine-tuning UEP reliability levels while adapting to diverse system parameters. Numerical results show that the proposed approach improves over established achievability bounds of randomized superposition coding-based UEP schemes with SIC decoding, making the proposed structured AE-based UEP codes a scalable and efficient solution for next-generation networks.

Problem

• Extends AE-based UEP codes to larger blocklengths efficiently
• Provides flexible framework for tuning UEP reliability levels
• Improves over randomized superposition coding-based UEP schemes

Innovation

• Structured AE architecture for UEP coding
• Superposition coding with SIC decoding
• Flexible AE subblocks for reliability tuning