Research On Task Offloading And Resource Allocation Scheme In Fog Networks

With the development of the Internet of Things(IoT),the number of heterogeneous IoT devices(IDs)in the network is growing exponentially,which generate a large number of latency-sensitive and computation-intensive tasks.However,due to the limited computation capability of IDs,tasks need to be transferred to other servers with large computation capability to ensure the task processing latency.In the traditional cloud computing network,tasks need to be transmitted to a central cloud server that is far away from IDs,resulting in huge network data traffic and congestions.Currently,fog computing networks provide task computing services for IDs by deploying a large number of fog nodes(FNs)with certain computing and storage capabilities,which shorten the task transmission and computing latency,and improve network transmission performance.Moreover,FNs are randomly deployed in the unsupervised area,the security of IDs data is one of the important issues in fog computing networks.In this thesis,we focus on the resource allocation and user privacy schemes in fog computing networks.Firstly,this thesis investigates the task offloading and resource allocation problem in fog computing networks.To ensure the loading balance of the FNs,the amount of computation resources involved in task offloading is dynamically adjusted,and the available computational capacity will be forwarded to IDs.Based on this,each ID establishes a candidate offloading FN set for the task offloading;Secondly,the problem is modeled as a network energy efficiency optimization resource allocation problem,which could be solved by the Energy Efficiency Resource Allocation Algorithm Based on the Candidate Fog Nodes(EE-CN);Then,the original optimization problem is divided into two sub-optimization problems: the local optimal offloading decision and the transmission power allocation.For the local optimal offloading decision problem,a Channel-based Optimal Transmission Resource Allocation Algorithm(COTRA)is proposed to obtain the local optimal offloading decision and transmission resource allocation.For the optimal transmission power allocation,a Lyapunov-based optimal power allocation algorithm is proposed to maximize the network energy efficiency while ensure the stability of the task data queue.Finally,the effectiveness of the algorithm is verified by simulation results.Furthermore,to improve the IDs data security and privacy,a blockchain-enabled task offloading scheme in fog computing networks is studied.Firstly,a two-stage Stackelberg game model is designed for purchasing the computing resource.The amount of available computing resources for each ID is obtained by maximizing the revenue of both parties through the price game between resource supply and demand.Secondly,the network utility is defined as the total energy consumption of the fog computing network and the total latency of the blockchain network.To minimize the network utility,a task offloading and resource allocation algorithm in Blockchain-enabled fog computing network(TRBFCN)is proposed;then,the original optimization problem is divided into the energy consumption-based task offloading problem and the transmission resource allocation problem.To solve the computational resource allocation,a network utility-based computational resource allocation algorithm is proposed to obtain local optimal computational resource allocation by alternating iterative fog computing and blockchain network resource allocation algorithms;finally,the effectiveness of the algorithm is verified by simulation.