Optimizing Cellular Network Binary Rate and Fairness: Examination of a Two-Tier Hybrid Duplex Architecture Employing Full-Duplex and Half-Duplex Modes

Full-duplex HCNs have emerged as a compelling solution to meet the escalating traffic demands of future fifth-generation networks. Theoretical advantages of full-duplex cellular networks include the potential for doubling throughput compared to half-duplex cellular systems, as base stations can both receive and transmit within the same frequency band, thereby enhancing various performance metrics. However, in practice, achieving this twofold increase in throughput and maintaining other performance metrics can be challenging due to interference from full-duplex base stations, particularly those of large-scale macro BSs. In this study, we explore a two-tier hybrid duplex HCN architecture, where small-scale BSs operate in full-duplex mode while large-scale BSs and user equipment utilize traditional half-duplex communication. This hybrid approach aims to mitigate interference issues and manage costs effectively. We investigate the impact of various network parameters, coupling scenarios, duplexing modes, and association criteria on the performance of the two-tier HCN. Our simulation results reveal the significant influence of user and small-scale BS densities, as well as the maximum user limit, on binary rate and the Jains fairness index. Remarkably, both binary rate and Jains fairness index exhibit optimal performance at specific values of cell range expansion. In the context of downlink and uplink binary rates, FD and hybrid duplex modes outperform other modes, particularly when the association criterion is based on maximizing average received power. However, when considering the Jains fairness index under the nearest association criteria, half-duplex mode consistently outperforms other modes. This research sheds light on the intricate interplay of duplexing modes, association criteria, and network parameters in the context of FD HCNs, providing valuable insights for optimizing network performance in the era of 5G and beyond.