Task And Channel Prediction Based Single User Computation Offloading

With the development of information and communication technology,mobile communication plays an increasingly important role in daily production and life.Because the growth of computing power of mobile devices is lower than the demand of consumers,the improvement of battery energy density is limited,the computing power and endurance of mobile devices cannot meet the growing demand of consumers.In order to solve this problem,previous studies have transferred computing tasks to the base station side,but there are some problems that cannot accurately predict the future time of computing task attributes and channel status,resulting in the limitation of the scope of application of related task offloading algorithms.In order to solve this challenge,this thesis used the statistical characteristics of single-user computing tasks and channel states to predict the future computing task attributes and channel states,aiming at minimizing energy consumption or delay,and chooses appropriate offload paths for computing tasks to improve the endurance,computing power and response speed of mobile devices.The main contributions are as follows:In order to solve the problem of the uncertainties of computing tasks and channel states in the future of mobile internet,this thesis studied the statistical characteristics of computing tasks and channel states by analyzing the historical data of computing tasks and channel states,predicted the future computing tasks and channel states,and obtained the attribute of computing tasks and channel states at the next time.The simulation results showed that computation task and channel state prediction algorithm can significantly improve the reliability of computational task attributes and channel state prediction.In order to solve the problem of dependence of offloading path selection on computational tasks and channel states in computational offloading algorithm,this thesis proposed an offloading path selection algorithm to improve energy consumption and delay performance.On the basis of calculating task attributes and channel state prediction,energy consumption and delay performance after choosing offloading paths are simulated.The simulation results showed that the energy ef'ficiency of the new algorithm is improved by more than 90%compared with the traditional single-user partial computational offloading algorithm without computational task attributes and channel prediction.In order to solve the problem of unstable transmission rate caused by frequent handover of mobile devices in the Fifth Generation(5G)mobile communication system,this thesis proposed a Long Term Evolution(LTE)link-assisted handover method for mobile devices,and studies the continuity of service transmission rate when mobile devices switch base stations.Finally,the feasibility and reliability of the proposed scheme in engineering are verified by testing.