Research On Task Allocation And Information Awareness In Fog Radio Access Networks

With the arrival of full connection era,it is predicted that new services and application scenarios will continuously emerge and hence diverse terminals will exist.Besides mobile intelligent terminals,vehicle terminals and Internet of Things(IoTs)terminals will also be in 5G radio networks.Service types and network performance asked for by diverse terminals are different.Fog radio access networks(F-RANs)are able to support different kinds of services.The scalable fog computing composed of fog access points(F-APs)and fog user equipments(F-UEs)can provide real-time collaboration radio signal processing(CRSP)and flexible cooperative radio resource management(CRRM),which makes F-RANs adaptive to the different service types asked for by diverse terminals.Based on data mining technology,F-RANs can utilize terminals' historic data to infer terminals' type,and it can enable F-RANs respond proactively and intelligently to improve subscribers' quality of service(QoS)and the efficiency of network resources usage.In this paper,the information awareness technology and task allocation approach in F-RANs are studied,and a hierarchical terminal awareness approach and a task allocation approach that minimize the total training time of terminal awareness are proposed.The main contributions of this thesis are listed as follows:1.In this thesis,due to the acquisition of raw data and the execution of computation tasks in F-RANs,centralized approaches for terminal awareness can put heavy burdens on fronthaul and cloud.To alleviate these burdens and infer the terminals' type precisely,a hierarchical terminal awareness approach integrating data mining technology is proposed.The core idea is that the raw data processing and the training of classifiers are shifted to fog access points(F-APs).Meanwhile,using Support Vector Machine to establish the prediction model to infer the terminals' type,and the prediction accuracy is close to 90%.2.In order to minimize the total training time of information awareness,the training task allocation can be formulated as an integer problem.The integer problem which can be solved by a branch and bound based method and a low complexity heuristic method.Simulation result shows that the heuristic method can achieve optimal performance,and the impacts of the heterogeneity in F-APs computing capabilities on the training time and the task allocation are demonstrated and analyzed.The training task allocation approach of the hierarchical terminal awareness can make F-RANs infer terminals' type more quickly,and the information perception technology can make F-RANs respond more proactively and intelligently and so improve the subscribers' quality of service(QoS)and the efficiency of network resources usage.