Modulation scaling for real-time energy aware packet scheduling

Portable wireless communication systems operate on a limited battery supply, and energy efficiency is therefore crucial. Voltage scaling techniques have been proposed to lower the energy consumption of embedded processors and real-time operating systems have incorporated these schemes in their task scheduling engine. However, the actual data transmission itself constitutes a major portion of the total energy consumption in these wireless communication systems. In this paper, we extend the scaling notion to the realm of wireless communications and propose a novel technique called modulation scaling to decrease the energy consumed during data transmission. Modulation scaling trades off energy consumption against transmission delay and as such, introduces the concept of energy awareness in communications. We investigate how modulation scaling can be exploited to design a dynamic power management engine at the level of the radio. This engine coordinates the packet transmission schedule while optimizing energy efficiency. We demonstrate such a power management module for real-time traffic and show that it reduces the energy consumption of data transmissions by up to 50 % through smart traffic scheduling.