946 publications from this institution
Current spectrum-sharing frameworks struggle with adaptability, often being either static or insufficiently dynamic. They primarily emphasize temporal sharing while overlooking spatial and spectral dimensions. We propose an adaptive, AI-driven spectrum-sharing framework within the O-RAN architecture, integrating discriminative and generative AI (GenAI) to forecast spectrum needs across multiple timescales and spatial granularities. A marketplace model, managed by an authorized spectrum broker, enables operators to trade spectrum dynamically, balancing static assignments with real-time trading. GenAI enhances traffic prediction, spectrum estimation, and allocation, optimizing utilization while reducing costs. This modular, flexible approach fosters operator collaboration, maximizing efficiency and revenue. A key research challenge is refining allocation granularity and spatio-temporal dynamics beyond existing models.
In this paper, we develop novel two-tier interference management strategies that enable macrocell users (MUEs) to improve their performance, with the help of open-access femtocells. To this end, we propose a rate-splitting technique using which the MUEs optimize their uplink transmissions by dividing their signals into two types: a coarse message that is intended for direct transmission to the macrocell base station and a fine message that is decoded by a neighboring femtocell and subsequently relayed over a heterogeneous (wireless/wired) backhaul. For deploying the proposed technique, we formulate a non-cooperative game between the MUEs in which each MUE can decide on its relaying femtocell while maximizing a utility function that captures both the achieved throughput and the expected backhaul delay. Simulation results show that the proposed approach yields up to 125% rate improvement and up to 2 times delay reduction with wired backhaul and, 150% rate improvement and up to 10 times delay reduction with wireless backhaul, relative to classical interference management approaches, with no cross-tier cooperation.
