Research On Uplink Channel Parameter Estimation Based On SRS Signal In High-Speed Railway Scenario

High-speed railway is one of the important transportation infrastructure of a country.In recent years,the high-speed railway in China has been developing rapidly,and its technology is now at the forefront of the world.Nowadays,high-speed railway has become one of the most preferred trasnporations for many people.However,the increasing traffic demand has put forward higher requirements for the wireless communication system of high-speed rail.At present,GSM-R is mainly used in railway wireless communication.The system cannot provide train users with wireless Internet access capability,and its narrowband characteristic makes it difficult to transmit more abundant train monitoring information.It needs to evolve to the next generation communication network system.Due to the rapid development of 5G systems in public mobile communication system,the next generation wireless communication systems based on 5G-R have attracted widespread attention.By comparison with traditional mobile communication scenarios,high-speed rail broadband wireless access has problems such as huge Doppler frequency offset and frequent cell switching due to the high mobility of trains.It is also of great significance to estimate the direction of the uplink signal for beamforming and detection of the downlink channel.This dissertation considers a two-hop access system based on Multi-TRP in the high-speed railway scenario,in which the firsthop link is the communication between the roadside TRP and the on-board relay station MRS,and the second hop link is the communication between the invehicle access point and the user terminal.It mainly focuses on the first hop link,and studies the estimation method of the uplink channel parameters from the perspective of SRS signal.This dissertation mainly includes the following content:1、Uplink channel synchronization including timing synchronization method and frequency offset estimation method.For timing synchronization,this dissertation studies three estimation algorithms based on training sequences: SC algorithm,Minn algorithm and A.B algorithm.In terms of simulation performance,due to the influence of the flat area,the estimation accuracy of the SC algorithm is maintained at a fixed value with the increase of the signal-to-noise ratio,while the Minn algorithm and the A.B algorithm approach zero.Among them,the A.B algorithm achieves better performance,but it has the highest algorithm complexity.From the perspective of frequency offset estimation,this dissertation studies the Moose algorithm and the Classen algorithm and conducts the simulation comparison.In terms of performance,Classen has better estimation accuracy,but the implementation of Classen method is more complicated2、Estimation of the uplink signal angle.Firstly,the estimation method of signal angle of arrival based on subspace decomposition is introduced,and the performance is compared.Afterwards,a joint estimation method based on the fusion of multiple TRP estimations is proposed for the Multi-TRP system in the high-speed railway scenario,and the simulation is given.The result shows that the algorithm achieves higher estimation accuracy than a single TRP.It can make up for the loss of estimation accuracy caused by the insufficient number of single TRP antennas due to the limited space.Moreover,when the train enters the roadside TRP array,the root mean square error of estimation fluctuates less with the change of train position,which shows that the algorithm has more stable estimation performance.