Research On Resource Allocation Algorithm For M2M Communication Uplink In LTE-A Network

With the rapid development of the Internet of things service,LTE-A network,as the most important bearing network of M2M services,has become the most widely used form of the Internet of things,and is a hot spot of development in the communication industry.Aiming at the characteristics of M2M communication terminal devices,such as frequent small data transmission and uplink superior.This dissertation takes the LTE-A uplink allocation problem as the main research object,and studies the energy efficiency and packet loss rate of the LTE-A uplink system M2M device.The main work is as follows:1.The current research status and development trend of M2M communication uplink allocation under LTE-A network at home and abroad are analyzed,and the technical characteristics of LTE-A network and M2M communication are studied.2.A LTE-A uplink resource allocation algorithm based on neural network is studied.Firstly,the model of M2M communication system under LTE-A network is established and the maximum energy efficiency objective function is defined.By reconstructing the original system model,the equivalence of the energy efficiency problem with the nonlinear constraint conditions is converted into the two dimensional knapsack problem with the highest value.The neural network model is introduced,and the knapsack problem is solved by integrating pointer network,attention mechanism and critic auxiliary network.Finally,simulation environment is constructed through pytorch software.Compared with the three classical algorithms of LTE-A uplink,the simulation results show that it can greatly improve the energy efficiency of the LTE-A system and reduce the packet loss rate,considering the fairness between the M2M devices and the quality of service.3.A saving resource allocation algorithm for M2M devices with low delay requirements in LTE-A uplink is studied.According to the characteristics of M2M communication and the limitation of SC-FDMA network resource allocation in LTE-A uplink,the system model under the LTE-A network is established and the maximum energy efficiency target function is defined;the original system model is reconstructed,the Charnes-Cooper transformation and the linearization scheme of Glover are used to transform the NP-Hard problem into a mixed integer linear programming problem which can be solved by the branch definition algorithm.Finally,the simulation environment is built by Matlab software.The simulation results show that,compared with three classical algorithms and genetic algorithms,this algorithm can greatly improve the system energy efficiency and reduce the probability of data packet loss while guaranteeing the priority and quality of service of M2M devices.