Traffic Offloading in Heterogeneous Networks With Energy Harvesting Personal Cells-Network Throughput and Energy Efficiency

This work develops a tractable model to analyze the performance of downlink heterogeneous cellular networks (HCNs) with both power-grid-connected base stations (PG-BSs) and energy harvesting small cell access points (EH-SAPs). Each EH-SAP forms a personal cell that is active (and is available to serve others) only when its own priority user requests service and its battery contains sufficient energy to transmit. By modeling the battery dynamics of an EH-SAP as a discrete-time Markov chain and by considering a practical power consumption model, the rate coverage, network throughput, and energy efficiency are derived as functions of the PG-BS and EH-SAP densities, transmission powers, cell association biases, and energy harvesting capabilities. Cell association biases control the traffic load among different tiers and transmission powers affect EH-SAPs' probability of being active. Exact expressions of the performance metrics are first derived for the general multitier scenario and are then used to obtain approximate closed-form expressions for two-tier networks using the mean load approximation. The analytic results of the two-tier network provide valuable insights on the achievable performance and the choice of system parameters, and are further validated with numerical results.