Increasing Opportunistic Gain in Small Cells Through Base Station-Driven Traffic Spreading

Abstract

Dense deployment of small cells is an important, emerging trend to enable future cellular networks to cope with growing traffic demand. However, this reduces the number of users per cell and thus opportunistic scheduling gain. We propose a base station-driven energy-aware approach to exploit user-user communication to increase the opportunistic gain. We use tools from stochastic Lyapunov optimization to determine the optimal scheduling policy subject to a constraint on energy consumption for user-user communication. Our simulation results show that with a large energy budget, packet transfer delay is reduced by up to 70%. The bulk of the performance improvement can be achieved with only a small increase in energy consumption, where 60% of the improvement is achieved at only 20% of the additional energy consumption. Further, we evaluate our algorithm using realistic video traffic traces and show that frame loss ratio is reduced by 90% and PSNR is improved by 4dB.