Resarch On Handover Technology Of LTE-M Communiction System

With the continuous development of rail transit,the speed of trains has gradually increased,and the operating environment has become more complex.The train-to-ground communication system requires safer and more efficient wireless communication technology.LTE-M wireless communication system is dedicated for rail transit,which has the advantages of high information transmission rate,strong anti-interference,and small transmission delay.For rail transit,trains are running at high speed and requiring frequent handoffs.Handoff technology is the basis for ensuring continuous train-to-ground communication,which directly affects the real-time and reliability of train-to-ground wireless communication.The existing handover algorithms have poor adaptability in the rail transit operating environment,and are prone to communication interruption and ping-pong handover.Therefore,this thesis optimizes the handover algorithms based on the characteristics of the rail transit operating environment.The main work of this thesis is as follows:(1)Study the LTE-M system architecture and handover procedures,and determine the required parameters and implementation methods for LTE-M handover.Analyze the specific impact of the rail transit operating environment on the handover,complete the specific wireless coverage planing and channel modeling,and build a theoretical basis for the optimization of the handover algorithm.(2)Analyze the principle of the A3 handover algorithm,study the influence of speed on handover,and find that the handover hysteresis threshold and delay trigger time need to decrease with the increase of speed.In this thesis,an adaptive switching algorithm based on speed is adopted,and the switching hysteresis threshold is dynamically changed with the speed by establishing a function model,and the delay trigger time is dynamically changing based on statistical characteristics and speed classification.This article proposes three functional models.This thesis proposes three function models.Simulations show that the handover algorithm based on the mean square function model has better handover trigger rate and handover success rate,so the function model is more suitable for the adjustment of handover decision parameters.(3)Aiming at the volatility of reference signal received power RSRP,this thesis adopts DGM(1,1)gray neural network method to perform fitting prediction to reduce the influence of channel mutation on handover measurement.Firstly,the principle of discrete gray prediction algorithm and BP neural network are studies separately,and it is found that the combination of the two can complement each other to obtain better prediction performance.Then the gray neural network based on DGM(1,1)is combined with the traditional switching algorithm,the related data preprocessing method is given,and the specific implementation steps of the switching algorithm based on DGM(1,1)gray neural network are designed.Finally,the simulation shows that the gray neural network handover algorithm based on DGM(1,1)can improve the volatility of RSRP and reduce the ping-pong handover rate.