Energy consumption of multi-hop wireless networks under throughput constraints and range scaling

Abstract

This paper studies the joint optimization of throughput and energy consumption in multi-hop wireless networks and associated scaling trends. We employ transmission range as the main control knob and forfeit the common, but often unrealistic, premise that communication energy is proportional to the transmission range raised to a path loss exponent. In real systems, idle power, overhearing and power amplifier play a major role and energy efficiency may potentially favor long transmission ranges. On the other hand, throughput maximization favors short transmission ranges. We derive approximate expressions for average per-node energy consumption as a function of the transmission range for various representative system behaviors. Based on these, we build a joint energy-throughput analytical model and validate it through simulations. Then, we address two key problems: Given an upper bound on the average per-node power consumption (or energy budget) find the maximum per-node throughput (or number of transferred bits) and the range that achieves it. And, given a lower bound on average per-node throughput (or minimum bit volume to be transferred), find the transmission range that results in the minimum per-node power consumption (or minimum energy dissipation).