Delay-energy-aware joint multi-cell association, service caching, and task offloading in hybrid-task heterogeneous edge computing networks

Abstract

In highly dense networks with huge computational requirements, mobile edge computing has been proposed to alleviate network traffic congestion and reduce system latency by offloading the intensive computational tasks to the network edges for execution. As a result, achieving low energy consumption and reduced system latency has become increasingly important under this paradigm. In this paper, we propose a delay-energy-aware algorithm for minimizing the overall system latency, energy consumption and balancing the load among base stations, particularly in the case of hybrid-task scenarios. A novel crafted weighted-sum objective function for the total system latency and energy consumption is designed to formulate a non-convex joint optimization problem. The Gibbs sampling algorithm is used to solve the formulated optimization problem through updating the caching and offloading decision variables. The proposed framework investigates the optimal multi-cell association, power allocation, service data caching, and computational task offloading for multi-tier communication and edge computing networks. The effect of limited quota on multi-tier heterogeneous networks is investigated under Rayleigh fading channels. Simulation results demonstrate the superiority of the proposed algorithms over the state-of-the-art works in terms of reducing the system latency and energy consumption.