Energy-Aware Cooperative Computation in Mobile Devices

Abstract: New data intensive applications, which are continuously emerging in daily routines of mobile devices, significantly increase the demand for data, and pose a challenge for current wireless networks due to scarce resources. Although bandwidth is traditionally considered as the primary scarce resource in wireless networks, the developments in communication theory shifts the focus from bandwidth to other scarce resources including processing power and energy. Especially, in device-to-device networks, where data rates are increasing rapidly, processing power and energy are becoming the primary bottlenecks of the network. Thus, it is crucial to develop new networking mechanisms by taking into account the processing power and energy as bottlenecks. In this paper, we develop an energy-aware cooperative computation framework for mobile devices. In this setup, a group of cooperative mobile devices, within proximity of each other, (i) use their cellular or Wi-Fi (802.11) links as their primary networking interfaces, and (ii) exploit their device-to-device connections (e.g., Wi-Fi Direct) to overcome processing power and energy bottlenecks. We evaluate our energy-aware cooperative computation framework on a testbed consisting of smartphones and tablets, and we show that it brings significant performance benefits.