Energy-minimized Partial Computation Offloading for Delay-sensitive Applications in Heterogeneous Edge Networks

Mobile Devices (MDs) support delay-sensitive and computation-intensive applications. Yet they only have limited resources, thereby failing to totally run all applications. A mobile edge computing (MEC) paradigm has been proposed to provide additional resources for MDs. Servers in MEC are often deployed in both macro base stations (MBSs) and small base stations (SBSs). Thus, it is highly challenging to associate resource-limited MDs to them with high performance, and realize partial computation offloading among them for minimizing total energy consumption of an MEC system. To tackle these challenges, this work proposes a novel computation offloading approach for delay-sensitive applications in hybrid networks including MDs, SBSs and an MBS. It formulates total energy consumption minimization as a constrained mixed integer nonlinear program. It designs an improved meta-heuristic algorithm called Particle swarm optimization based on Genetic Learning, which integrates strong local search capacity of a particle swarm optimizer, and genetic operations of a genetic algorithm. It jointly optimizes task offloading among MDs, SBSs and MBS, users connection to SBSs, MDs CPU speeds and transmission power, and channels bandwidth allocation. Simulations with real-world data demonstrate that it significantly outperforms other methods in energy consumption of an entire system consisting of MDs, SBSs and MBS.