Fine-grained Task Offloading Policy In Dense Small Cell Network

In recent years,the amount of data processed by mobile wireless network is dramatically increasing,and applications such as virtual reality,telemedicine,are becoming popular.However,this has brought many challenges to mobile wireless networks and users.For example,network traffic is too heavy,applications cannot run smoothly,and user experience is bad.According to these problems above,small cell network and mobile edge computing are proposed.Small cell network can provide users with higher data transmission rate and better energy efficiency because it shortens the distance between users and small cell base station.In mobile edge computing,mobile edge hosts are deployed closer to users,which reduces transmission delay,process delay and burden on core network.Small cell network and mobile edge computing are two key technologies of 5G(5th Generation).Combination of small cell network and mobile edge computing has many benefits,such as making network more powerful and providing users with more computation power.However,there are several problems and challenges.For example,task offloading problem and joint allocation of computation and storage resources.For issues mentioned above,the main work of this thesis is as follows:Firstly,in multiple small cell scenario,propose a task allocation scheme with mobile edge computing.Consider a complex computation task model which contains many dependent subtasks.The task can be modeled as directed acyclic graph.Secondly,establish a subtask-offloading problem based on delay optimization is,which jointly considers the transmission and offloading delay from users to small cells.Thirdly,convert the proposed problem into a scheduling problem,which is NP-hard(non-detenninistic polynomial).A heuristic algorithm based on the list scheduling is proposed to solve this problem.Finally,do simulations on the proposed algorithm and verify the effectiveness of the algorithm.Secondly,in a multi-user scenario,propose a joint small cell clustering and resource allocation scheme with mobile edge computing technology.Firstly,established the system model in which different user's needs are considered.Secondly,formulate the delay optimization problem by jointly consider transmission power and computation task allocation.The formulated problem is a mixed integer nonlinear programming problem.Thirdly,propose a low-complexity heuristic algorithm for small cell clustering and a genetic algorithm based resource allocation scheme.Finally,various simulation results prove the proposed algorithm can minimize the delay of network.