| With the advent of the "Internet of Everything" era,the amount of information generated by Io T devices has exploded.Cloud computing,with its powerful computing capacity and on-demand features,has long been favored for handling massive amounts of data.However,the high latency brought by cloud computing and the inability to provide instant and mobility support are undesirable for networks in mobile scenarios such as intelligent traffic.So Fog computing has been introduced into computationally intensive applications at the edge of networks.Compared to cloud computing,fog computing is more distributed,emphasizing quantity and location rather than computing capacity.Fog computing not only reduces computing delay,but also brings severe challenges to the complex resource allocation problem of available computing and communication resources with strict quality of Service(QOS)requirements.In this paper,we investigate the computation offloading and resource allocation problem of cloud and fog co-computing in heterogeneous networks by joint optimization of unloading decision,transmission power and computing resources.However,in heterogeneous environment,the computation complexity of the optimal strategy will be very large.Therefore,an improved genetic algorithm(IGA)is proposed in this paper,which can maintain a low computation complexity while obtaining the optimal solution.In the IGA algorithm,we proposed to use penalty function to express the constraints condition of optimization problem,and use non-uniform mutation operator to accelerate the convergence speed.Besides,an improved method of parameter self-adaptation and a perturbation method of mutation probability based on population fitness standard deviation are proposed to optimize genetic algorithm.The results show that the proposed genetic algorithm can obtain lower average cost of the system while keeping smaller computational complexity compared with the existing algorithms. |
PDF Full Text Request