Multi-parameter decision algorithm for mobile computation offloading

Today, mobile handsets are more and more capable to run complex applications. Mobile computation offloading offers the potential of extending mobile devices capabilities and battery lifetime. Traditional mobile computation offloading decision algorithms are mainly based on the offloading energy trade-off between locally consumed energy and offloading energy. Femtocloud is a novel paradigm introduced by the European project TROPIC [1] that merges cloud computing services and the benefits of femtocell networks. In this paper we present a novel offloading algorithm that takes decision about offloading mobile computation to a femtocloud. The proposed offloading algorithm incorporates a multitude of parameters in the offloading decision process while reducing the mobile handset energy consumption and keeping a good user quality of experience. Simulations results show that our proposed algorithm is able to extend the mobile battery life and to assure the computation of all the applications while respecting latency and memory constraints.