Task Offloading in UAV-Enabled MEC Against Smart Jammers Relying on Evolutionary Game Theory

Unmanned aerial vehicles (UAVs)-enabled mobile edge computing (MEC) has emerged as a key enabler to tackle the increasing computation requirement of mobile devices (MDs). However, existing research assumes ideal conditions and fails to characterize the interference and interaction between MDs and jammers. In this paper, we jointly optimize latency and energy efficiency, considering the interference from smart malicious nodes. To elaborate, we utilize the evolutionary game model (EGM) to analyze the confrontational behaviors between MDs and jammers in the MEC system. Furthermore, we propose a matching access solution and derive the existence conditions of evolutionary stable strategies relying on the Friedman method, which forms the basis of an algorithm to analyze the boundedly rational strategies of both players. Finally, extensive experimental results demonstrate that our EGM and algorithm can optimize the edge computation performance.