An Alerting System For Foreign Matter Invasions And Natural Disaster In Railway Based On Integrated Network Architecture

The increasing speed of high-speed rail needs higher demand on the operation safety,preventing foreign matter invasions and natural disasters such as landslides or mud-rock flow.Meanwhile,to satisfy the rapidly growing capacity requirements of future railway wireless communications systems,the International Railways Alliance(UIC)has proposed the evolution strategy of GSM-R to LTE-based high-speed railway wireless communications systems.On one hand,as the millimeter wave(mmWave)has a very wide bandwidth and more "blank" non-licensed frequency band,mmWave wireless communications will continue to be one of the hot topics in wireless communication field.On the other hand,the mmWave also plays an important role in radar detection.In this context,this thesis proposes an LTE-based network architecture combining mmWave radar detection with wireless communication for railway.The network architecture aggregates mmWave detection and mmWave communication on the basis of the original LTE,which provides a cost-effective method to construct a radar monitoring system based on the existing communication network infrastructure by using the wide bandwidth unlicensed mmWave band.At the same time,the network architecture can provide high-capacity wireless communications services for trains.As for the radar monitoring system,in order to improve the detection accuracy and reduce the occurrence of false alarm,this thesis studies the plot processing techniques of foreign matter invasions and natural disaster detection.On the basis of the integrated network architecture,this thesis designs an alerting system satisfying the actual demand of foreign matter invasions and natural disaster warning,to ensure the timeliness and accuracy of the alerting information,and to provide a solution for the safe driving of the high-speed train.For the research on integrated network architecture combining mmWave radar detection with wireless communication,this thesis firstly studies the BBU(Building Baseband Unit)pool + RRU(Radio Remote Unit)cloud wireless access scheme,network coverage scheme and frequency band selection,and then investigates the radar detection system.Finally,the simulation analysis of the network architecture performance is carried out.According to the BBU pool + RRU cloud wireless access scheme,all RRUs are connected to the same BBU pool form a logical unit.Regarding to the network coverage scheme,intertwined bilateral layout is deployed for RRU plan.The LFMCW(Linear Frequency Modulation Continuous Wave)radar is mainly exploited that is suitable for the near-field detection.The echo signal processing scheme adopts a two-dimensional signal processing algorithm and a two-dimensional CFAR(Constant False Alarm Rate)algorithm to detect the target,which can effectively suppress the influence on detection performance caused by complex background clutter.The simulation results of the network architecture show that the capacity of communication could be greatly improved due to the introduction of the mmWave band RRUs,while ensuring the effective monitoring of the radar monitoring system.For the plot processing techniques of foreign matter invasions and natural disaster detection,the plot preprocessing plot coherence,target filtering algorithm and plot parameter estimation algorithms are studied respectively,and the related simulation analysis is performed.The plot preprocessing algorithm can filter out most of the interference plots and directly raise the alert when the number of target plots is unusually large.In this thesis,a plot coherence algorithm based on DBS CAN(Density-Based Spatial Clustering of Applications with Noise)algorithm is proposed,which can adapt to the discontinuous and irregular target plots in complex scene.The target filtering algorithm can reduce false alarms caused by clutter or non-threatening objects(such as birds,small stones,etc.).In this thesis,the two-dimensional plot parameter estimation algorithm based on similarity is well improved,which is showed by simulation that the improved algorithm is more accurate and has stronger anti-interference ability.For the alerting system,firstly overall framework of the alerting system and alerting strategies are designed and analyzed,and then the alerting signal processing algorithm of the alerting system is introduced.Then,the physical layer of the integrated network is studied,including frame structure and signaling flow.The related parameters of the millimeter-wave frequency band in OFDM(Orthogonal Frequency Division Multiplexing)are studied according to the propagation characteristics of millimeter wave,and the relevant feasibility scheme is given.Finally,the notification algorithm of alerting system is studied to ensure the timeliness and accuracy of system alerting,including the generation of the alerting information,the train positioning process and the notification signaling flow of the alerting information.