🤖 AI Summary
Conventional multiple access techniques struggle to effectively exploit the characteristics of wireless propagation environments, thereby limiting performance gains in AI-native communication networks. To address this challenge, this work proposes Environment-Division Multiple Access (EDMA), which, for the first time, treats the propagation environment as an independent multiple access dimension. By integrating environment-aware sensing, reconfigurable intelligent surfaces, flexible antenna arrays, and artificial intelligence algorithms, EDMA actively reconfigures and leverages environmental properties to distinguish users. The study establishes a theoretical framework for EDMA and introduces both AI-assisted and AI-native architectural paradigms, advancing multiple access technologies toward environment intelligence. This approach significantly enhances system capacity and energy efficiency, while also clarifying EDMA’s application potential and future research directions within AI-native networks.
📝 Abstract
In this article, a new type of multiple access, termed Environment-Division Multiple Access (EDMA), is introduced and its interaction with AI-native communication networks is illustrated. In particular, the key properties of EDMA, such as utilizing the features of wireless propagation environments, integrating advanced flexible antennas, and proactively reconfiguring propagation environments, are described. The article also illustrates two types of applications of AI tools to multiple access, namely AI-assisted EDMA and AI-native EDMA. Finally, open problems and important directions for future research in AI-assisted EDMA are discussed.