Collaborative research: vines: track 1: NSF-RCF: FORMULA3: framework for optimal resource management using generalizable AI in FR3 dynamic programmable wireless environments

The National Science Foundation Division of Information and Intelligent Systems awarded Texas A&M Engineering Experiment Station $250,000 on October 1, 2026, under the Computer and Information Science and Engineering program (CFDA 47.070) for collaborative research on proactive wireless network control in future broadband systems.

The project develops framework and artificial intelligence models for optimal resource management in frequency range 3 (FR3) upper mid-band wireless environments. Work focuses on four technical thrusts: developing realistic channel models and datasets using experimental platforms with programmable reflective surfaces; designing AI models that proactively configure beams and reflective surfaces to prevent connection problems; improving model generalization through environmental context information from integrated sensing and communication with continual learning; and providing rigorous evaluation through combined simulation and measurement in dynamic indoor and outdoor-to-indoor scenarios. The research addresses delivery of fast, stable, and reliable wireless connections inside buildings and from outdoors to indoors using smart programmable surfaces placed on walls to redirect signals around obstacles.

Performance occurs in College Station, Texas through September 30, 2029. The project is collaborative, bringing together investigators from Tennessee Tech University, Texas A&M, and other participating institutions. Applications enabled by the research include augmented reality, virtual reality, remote healthcare, and advanced manufacturing.

Abstract

This project studies how future wireless networks can deliver fast, stable, and reliable connections inside buildings and from outdoors to indoors. IT focuses on a new range of radio frequencies that can carry much more data than current systems while still covering useful distances. The project explores how smart programmable surfaces placed on walls can redirect wireless signals around obstacles. By combining these smart surfaces with artificial intelligence-based control, the project aims to prevent connection problems before users notice them. At its core, the project advances the science of proactive wireless network control through four tightly integrated technical thrusts. The first thrust develops realistic channel models and datasets using experimental platforms with programmable reflective surfaces operating in the frequency range 3 (FR3) upper mid-band. The second thrust designs novel artificial intelligence models that proactively configure beams and reflective surfaces. The third thrust improves model generalization by incorporating environmental context information from integrated sensing and communication, along with continual learning. The fourth thrust provides rigorous evaluation through combined simulation and measurement across dynamic indoor and outdoor-to-indoor scenarios. The project has broad impacts on society, education, and technology development. Reliable and low-latency wireless connectivity supports applications such as augmented reality, virtual reality, remote healthcare, and advanced manufacturing. The project trains graduate students in wireless systems, sensing, and artificial intelligence through hands-on research experiences across participating institutions. This collaborative project brings together investigators from tennessee tech university, texas a & m university, and new york university in the united states, and aalto university and tampere university in finland, to advance next-generation wireless connectivity through an international partnership. Through this international collaboration between the united states and finland, this project advances the foundation of artificial intelligence native wireless networks that can sense, predict, and adapt to change. 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.