Power-Delay Tradeoff in Mobile-Edge Computation Offloading with Heterogeneous Applications

Power-delay tradeoff scheme in smart device (SD) tasks execution by the joint optimization of dynamic voltage and frequency scaling (DVFS) problem has been studied over the last few years. Since, it is a great achievement for reducing power consumption but there is still need for the reduction of power consumption to its maximum extent without affecting the computational experience. In this paper, we proposed a new system model with the use of Mobile-Edge Computing (MEC) server to improve the standards of processing by serving the networking tasks of smart cellular device. An energy harvesting problem is generated by isolating networking and non-networking applications. The tradeoff is analyze by queue stability constraints and an online algorithm considering MEC and multiple wireless interfaces, e.g., Wi-Fi, LTE. Simulation results are taken based on a numerical example reveals that our smart device major part of power consumption is consisting of networking applications because of their continuous execution. The introduction of MEC server reduces the power consumption approximately 90 % of its actual power (current state) by executing the continuous networking applications on behalf of the Smart device and trading small delays of 10 ms between each executed task.