Optimization of base station density and user transmission power in multi-tier heterogeneous cellular systems

Abstract In this paper, a loss minimization issue is proposed, which includes both cost of user power consumption and base station (BS) deployment. A multi-tier heterogeneous network (HetNet) is considered and modeled with random geometry of Poisson point process (PPP). BSs and users are distributed randomly following PPP with the specific densities. Cutoff threshold ( ρ k ) as the average received power at the kth tier BS is proposed to model uplink channels. The kth tier user transmission power is formulated through a channel inversion power control. Tier Selection Probability indicating the possibility of a user accessing a certain tier is derived. Bias ( B k ) is proposed to change the kth tier Tier Selection Probability. Thus, an operator is able to change bias based on nearby environment and communication status. Also, Truncated Outage Probability and signal-to-interference ratio (SINR) Outage Probability are derived, which represents a connection failure status. Truncated outage results from insufficient power supply at users and SINR outage results from poor communication environment. Genetic algorithm (GA) is introduced to determine the trade off between outage constraints and the loss minimization issue. Simulation shows that, when equipped with equal bias, multi-tier network has smaller Truncated Outage Probability but larger SINR Outage Probability than single-tier network. On the contrary, with different bias, Truncated Outage Probability in multi-tier is larger than the single-tier case and SINR Outage Probability in multi-tier is less than that in single-tier. A considerable loss reduction can be witnessed in both cases.