Heterogeneous UAVs assisted mobile edge computing for energy consumption minimization of the edge side
Qiang Tang,Linjiang Li,Caiyan Jin,Lixin Liu,Jin Wang,Zhuofan Liao,Yuansheng Luo
DOI: https://doi.org/10.1016/j.comcom.2022.07.023
IF: 5.047
2022-10-01
Computer Communications
Abstract:The research on the Internet of Things (IoT) and edge computing, especially Unmanned Aerial Vehicles assisted Mobile Edge Computing (UAV-assisted MEC) attracts more and more interests of researchers. Nowadays, MEC systems with multiple UAVs have great research value, especially for emergency communication scenarios. In this paper, a heterogeneous UAVs assisted MEC system was proposed, and with the goal of minimization the edge side energy consumption, a parallel processing was involved to jointly optimize the communication scheduling, forwarding power, offloading data size, computing frequency and the trajectory of the UAV. The problem is formulated as a Mixed Integer Non-Linear Programming (MINLP), hard to solve. Therefore, we divided the MINLP into two sub-problems by applying the Block Coordinate Descent (BCD) method, and solved it using the Lagrangian Duality (LD) method and Successive Convex Optimization (SCO). The UAV’s trajectory was initialized as a circular, and then it was optimized by using the standardized convex solver design algorithm. In addition, a heuristic algorithm was put forward to obtain the optimization solution, while we set other benchmarks and the Monte Carlo (MC) algorithm to justify its validity and rationality. The simulation results showed that our strategy has better performance at minimizing energy consumption compared with other algorithms.
computer science, information systems,telecommunications,engineering, electrical & electronic