Pairing for resource sharing in cellular device-to-device underlays

D2D communications has attracted substantial research attention recently due to its potential for improving spectrum and energy efficiency within existing cellular infrastructures. This work investigates methods for matching a DUE link (or links) to CUE to form spectrum sharing partners. The objective is to maximize various sum performance metrics (reward) of all D2D links and CUEs after successful pairing under individual power constraints. In our approach, a set of DUE candidates for each CUE uplink for a requisite SINR level is first identified, which allows us to form a bi-partite graph linking CUEs with their candidate DUE links for spectrum sharing. Classic algorithms, such as the Hopcroft-Karp, Gale-Shapley, and Kuhn-Munkres algorithms, are investigated for maximizing different performance metrics (reward) of all spectrum sharing CUEs and D2D links. Their performance is compared tp a low-complexity fairness-based algorithm that can deliver solid sum rate performance without high complexity.