ERI:Interference-Aware Constellation Design

无线网络需要接入的用户数量呈指数级增长，这使得为每个用户分配专用频谱变得不切实际。为了解决这个问题，非正交传输被提出作为一个潜在的解决方案。通过在多个用户之间共享频谱，可以使更多的设备接入频谱，或者可以将更多的带宽分配给每个收发器，从而实现更高的数据传输速率。然而，非正交传输存在用户之间的干扰，需要有效的传输和解码方法才能成功部署。本研究的主要目标是开发这种方法并推进该领域，这对下一代无线网络和物联网（IoT）等应用具有重要价值。此外，该项目将提供教育机会，包括让本科生参与通信深度学习的研究和设计研究生水平的课程材料。这项研究的结果将与学术界和工业界分享。该项目旨在通过提出一种新颖的端到端基于学习的数字调制设计来解决用户间的干扰。数字调制技术用于实现信息传输的高比特率和低误码率之间的平衡。本研究的一个重点是在设计阶段同时考虑噪声和干扰的非正交多址（NOMA）的新叠加星座。与传统的正交调幅等星座不同，所提出的变换产生的非均匀星座可以适应不同的干扰水平，并在它们存在的情况下有效工作。该方法为多用户星座设计提供了新的见解和工具，具有提高频谱效率、降低误码率和减少下一代无线网络通信延迟的潜力。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Wireless networks have seen an exponential increase in the number of users requiring access, making it impractical to allocate a dedicated spectrum to each user. To address this issue, non-orthogonal transmission has been proposed as a potential solution. By sharing spectrum among multiple users, more devices can access the spectrum or more bandwidth can be committed to each transceiver, achieving higher data transmission rates. However, non-orthogonal transmission suffers from interference among users, and effective transmission and decoding methods are needed for its successful deployment. The primary goal of this research is to develop such methods and advance the field, which has significant value for next-generation wireless networks and applications like the Internet of Things (IoT). In addition, the project will provide education opportunities, including involving undergraduate students in research on deep learning for communications and designing graduate-level course material. The findings of this research will be shared with both academic and industrial communities. This project aims to address inter-user interference by proposing a novel end-to-end learning-based digital modulation design. Digital modulation techniques are used to achieve a balance between high bit rates and low bit-error rates for information transmission. A particular focus of this research is on designing new superimposed constellations for non-orthogonal multiple access (NOMA) that take into account both noise and interference during the design phase. Unlike traditional constellations such as quadrature amplitude modulation, the proposed transform results in non-uniform constellations that can adapt to different interference levels and work effectively in their presence. This approach provides new insights and tools for multi-user constellation design, which has the potential to improve spectral efficiency, reduce bit-error rate, and decrease latency in communication for next-generation wireless networks.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.