Project Grant 2431594

The National Science Foundation (NSF) awarded a $199,999 Project Grant under the Computer and Information Science and Engineering (CISE) program (CFDA 47.070) to San Francisco State University (SFSU) to conduct research and development on resilient next-generation (NextG) network design for federated learning over mobile devices. The award aims to address challenges such as bursty workloads, spectrum constraints, and privacy concerns when supporting federated learning applications on resource-constrained mobile devices. The key products and services to be delivered include:

1. Developing a serverless computing architecture at the network edge to efficiently provide machine learning as a service for large-scale federated learning over mobile devices.
2. Designing an energy-efficient differential privacy mechanism for mobile devices to protect training data privacy against inference attacks in federated learning.
3. Proposing a multi-bit over-the-air computation-based spectrum accessing approach to improve the resilience, scalability, and efficiency of federated learning model updates under limited spectrum availability.

This collaborative research project will enrich the knowledge of wireless systems and machine learning, while providing multidisciplinary training opportunities for underrepresented students. The award period is from January 1, 2025 to December 31, 2026.