Deep Autoencoder-Based Constellation Design in Multiple Access Channels

2025-05-01

AI Summary

Traditional constellations (e.g., QAM) in multi-access channels (MAC) suffer from severe inter-user interference, leading to high symbol error rates (SER) and suboptimal sum capacity. To address this, we propose an end-to-end joint constellation design framework based on deep autoencoders. This is the first work to introduce deep autoencoders into multi-user MAC settings, enabling interference-aware, dynamic symbol mapping optimization. Crucially, our approach abandons analytical assumptions about channel models or user count, supporting prior-free, data-driven constellation learning for arbitrary numbers of users. By jointly training channel-aware mapping functions with Monte Carlo gradient estimation, the learned constellations achieve optimal or near-optimal performance—demonstrated in both two- and multi-user scenarios. Results show substantial SER reduction and significant gains in constellation-constrained capacity, consistently outperforming existing analytical designs.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Search and Optimization: Learning to Search
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

In multiple access channels (MAC), multiple users share a transmission medium to communicate with a common receiver. Traditional constellations like quadrature amplitude modulation are optimized for point-to-point systems and lack mechanisms to mitigate inter-user interference, leading to suboptimal performance in MAC environments. To address this, we propose a novel framework for constellation design in MAC that employs deep autoencoder (DAE)-based communication systems. This approach intelligently creates flexible constellations aware of inter-user interference, reducing symbol error rate and enhancing the constellation-constrained sum capacity of the channel. Comparisons against analytically derived constellations demonstrate that DAE-designed constellations consistently perform best or equal to the best across various system parameters. Furthermore, we apply the DAE to scenarios where no analytical solutions have been developed, such as with more than two users, demonstrating the adaptability of the model.

Problem

• Design constellations for multiple access channels to reduce interference
• Improve symbol error rate and sum capacity using deep autoencoders
• Extend constellation design to scenarios lacking analytical solutions

Innovation

• Deep autoencoder designs flexible MAC constellations
• Reduces symbol error rate via intelligent interference awareness
• Adapts to multi-user scenarios without analytical solutions