Joint Computation Offloading and Resource Allocation in UAV Swarms with Multi-access Edge Computing

Abstract

With the emergence of multi-access edge computing (MEC) paradigm, mobile users who have computation-intensive and low-latency tasks can exploit computing resources of network edge MEC servers or other computing nodes. For a multi-user and multi-helper coexisting system (e.g., a cluster of UAVs), when computation offloading occurs, an efficient computation offloading scheme which can reduce users’ energy consumption or latency is required. In this paper, we study the computation offloading scheme in a multi-user UAV system in which UAVs with missions can offload part of the missions to helpers nearby to meet the delay constraint. And we minimize the total energy consumption of all devices by optimizing users’ strategy of offloading objects, transmission channel, and offloading rate. To solve the difficult problem of optimization variable coupling, we proposed a two-stage resource allocation scheme that exploits the convex optimization method and stochastic learning automata (SLA) algorithm respectively. Finally, simulation results indicate that our proposed schemes can effectively decrease energy consumption compared to several benchmark schemes.