Peer-to-peer computation offloading has been a promising approach that enables resource-limited Internet of Things (IoT) devices to offload their computation-intensive tasks to idle peer devices in proximity. Different from dedicated servers, the spare computation resources offered by peer devices are ran-dom and intermittent, which affects the offloading per-formance. The mutual interference caused by multi-ple simultaneous offloading requestors that share the same wireless channel further complicates the offload-ing decisions. In this work, we investigate the oppor-tunistic peer-to-peer task offloading problem by jointly considering the stochastic task arrivals, dynamic inter-user interference, and opportunistic availability of peer devices. Each requestor makes decisions on both lo-cal computation frequency and offloading transmis-sion power to minimize its own expected long-term cost on tasks completion, which takes into considera-tion its energy consumption, task delay, and task loss due to buffer overflow. The dynamic decision process among multiple requestors is formulated as a stochas-tic game. By constructing the post-decision states, a decentralized online offloading algorithm is proposed, where each requestor as an independent learning agent learns to approach the optimal strategies with its lo-cal observations. Simulation results under different system parameter configurations demonstrate the pro-posed online algorithm achieves a better performance compared with some existing algorithms, especially in the scenarios with large task arrival probability or small helper availability probability.

Siqi Mu;Yanfei Shen

School of Sports Engineering(China Big Data Center for Sports),Beijing Sport University,Beijing 100084,China

中国通信（英文版）

[1]Siqi Mu;Yanfei Shen-.Stochastic Learning for Opportunistic Peer-to-Peer Computation Offloading in IoT Edge Computing)[J].中国通信（英文版）,2022(07):239-256

requestors,tunistic,requestor

Stochastic,Learning,Opportunistic,Peer,Computation,Offloading,IoT,Edge,Computing,peer,computation,offloading,been,promising,approach,that,enables,limited,Internet,Things,devices,their,intensive,tasks,idle,proximity,Different,from,dedicated,servers,spare,resources,offered,by,dom,intermittent,which,affects,mutual,interference,caused,simultaneous,share,same,wireless,channel,further,complicates,decisions,this,work,investigate,problem,jointly,considering,stochastic,arrivals,dynamic,user,opportunistic,availability,Each,makes,both,cal,frequency,transmis,power,minimize,its,own,expected,long,cost,completion,takes,into,considera,energy,consumption,delay,loss,due,buffer,overflow,process,among,multiple,formulated,game,By,constructing,post,states,decentralized,online,proposed,where,each,independent,learning,agent,learns,optimal,strategies,observations,Simulation,results,under,different,system,parameter,configurations,demonstrate,achieves,better,performance,compared,some,existing,algorithms,especially,scenarios,large,probability,small,helper

0.553384