Research On Cross-correlation Characteristics Of Multi-link Wireless Channels In High Speed Railway Scenario

In recent years,the development of Chinese high-speed railway has achieved remarkable achievements.The rapid development of high-speed railway makes the demand for wireless communication business of high-speed railway increasing.The existing GSM-R(Global System for Mobile Communications-Railway)system is limited to the case of wireless communication of train control and can not provide data transmission services such as broadband access services for high-speed railway passengers.Therefore,it is necessary to evolve to the new generation of high-speed railway wireless communication system LTE-R(Long Term Evolution-Railway).The construction of LTE dedicated networks began in 2014 and the networks have covered most of the domestic high-speed rails.However,in the overlapping areas of the existing LTE dedicated networks,users will receive signals with same frequency and same cell identification which leads to the "echo effect".This effect can be regarded as multiple path interference and influence the demodulation performance of the receiver.Coordinated multiple points transmission/reception(CoMP)is a new technology which can solve the interference fundamentally by converting interference signals into useful signals through cooperative transmission.In order to solve the existing interference problems of LTE networks in high-speed railway scenarios and to optimize the networks,it is necessary to study the multi-link channels in the overlapping areas.Furthermore,the research on cross-correlation characteristics of multi-link channels can be used to evaluate the performance of using CoMP technology in high-speed railway scenarios.Based on the above practical research needs,this paper studied the multi-link channels in the overlapped areas of the LTE railway dedicated networks in HSR scenarios based on the channel measurement data,including the following innovative works:(1)By observing the characteristics of the received signals,a dynamic partition method based on time delay domain perspective is proposed to separate the impulse responses of multi-link channels.The multi-link channel impulse responses are successfully extracted by this method,which are prepared for the extraction of multi-link channel parameters.(2)Based on single-antenna channel measurement data,channel characteristic parameters of multi-link SISO(Single Input Single Output)channel are extracted,including shadow fading,K factor and delay spread and the statistical characteristics of channel parameters are then analyzed.Based on the extracted multi-link SISO channel parameters,the cross-correlation coefficients between parameters of multiple links are calculated,including the global cross-correlation coefficients and the local cross-correlation coefficients.The results show that the cross-correlation between multi-link SISO channels is low,and a good diversity gain could be obtained if CoMP technology is used in high-speed railway scenarios.(3)Based on the multi-antenna channel measurement data,multi-link MIMO(Multiple Input Multiple Output)channel spatial domain channel parameters are extracted,including angle of arrival,angle spread and power angle spectrum.As for the spatial cross-correlation of multi-link MIMO channels,the spatial cross-correlation of multi-link MIMO subchannels and the correlation matrix of collinearity(CMC)are studied.Finally,the capacity of multi-link MIMO channel are given,including single-link channel capacity and dual-link sum rate capacity under interference,and the relative sum rate capacity of multi-link MIMO channels are used to evaluate the space cross-correlation.The results show that the spatial cross-correlation between multi-link MIMO channels is larger than the cross-correlation between multi-link SISO channels.This study makes up for the blank of research on the cross-correlation characteristics of multi-link wireless channels in high-speed railway scenarios,which provides a new idea for the optimization of LTE dedicated network.Meanwhile,the research results based on measured data can provide some references for related theoretical and simulation researches.