Collaborative research: vines: track 1: AI-empowered semantic communications with energy efficiency and privacy preserving for IOT-enabled emergency response

The National Science Foundation Division of Information and Intelligent Systems awarded the University of Louisiana at Lafayette $475,000 on October 1, 2026, for collaborative research on AI-empowered semantic communications with energy efficiency and privacy preservation for IoT-enabled emergency response systems under the Computer and Information Science and Engineering program (CFDA 47.070).

The award funds development of a semantic communication framework that reduces communication overhead and improves energy efficiency in data transmissions from IoT sensors during emergency response and disaster management operations. Research activities include building an AI-native semantic communication framework with efficient representation embedding and data regeneration; designing methodologies to handle unseen emergency events and enhance energy efficiency for model training and updating; developing advanced neural network pruning methods for lightweight and computation-efficient semantic communications; and investigating privacy protection mechanisms for sensitive emergency data. The project also supports educational opportunities through summer tutorials, workshops, curriculum development, and research activities for students and young talent.

Work is performed in Lafayette, Louisiana. The period of performance runs from October 1, 2026, through September 30, 2029. No subawards are planned.

Abstract

Efficient and robust data transmissions are critical to make internet-of-things (IOT)-enabled emergency response systems work effectively because massive data are collected by IOT sensors and then transmitted to servers for real-time processing. This project designs a novel semantic communication framework for smart emergency responses. Through AI-empowered information extraction and semantic-based data regeneration, the project will reduce communication overhead and improve energy efficiency in data transmissions, enabling reliable and efficient mission-critical applications for emergency responses and disaster management. The project also provides new educational opportunities to students and young talents through summer tutorials, workshops, curriculum development, and research activities. This project aims to develop innovative AI-driven semantic communication techniques for IOT-enabled emergency response applications. Specifically, the proposed research activities are structured into four major tasks: 1) building an AI-native semantic communication framework with efficient representation embedding and data regeneration; 2) designing effective methodologies and strategies to deal with unseen emergency events and enhance energy efficiency for model training and updating; 3) developing advanced neural network pruning methods to enable lightweight and computation-efficient semantic communications; and 4) investigating effective mechanisms for privacy protection of sensitive emergency data during semantic communications. 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.- subawards are not planned for this award.