Abstract
In future 5G networks, in-band interference is perceived as one of the most critical performance bottlenecks. While current solutions classically treat interference as an additional source of noise, recent advances in information theory show that interference is not necessarily an opponent, but might be cancelled or supressed. In this paper, we propose a three-fold optimization method, which couples reduced complexity interference classification and matching techniques, to enhance the system performance. The matching objective consists in defining coalitions of users assigned to each Access Points (AP); match interferers that will be transmitting on the same spectral resources into groups of interferers; and define the transmission rates and interference regimes for each user inside each group. Our analytical study and simulations results show that our proposed solution allows for system spectral efficiency enhancements, compared to classical reference scenarios.