Wireless Channel Measurement And Modeling For Vehicle-to-Infrastructure Communications

Vehicular ad-hoc networks(VANETs)has become one of the research hotspots in the field of future wireless communications.Wireless channel measurement and modeling constitute the basis of network design,and provide an important reference standard for whether wireless signals can be received.Especially,the correct reception of signals is directly related to personal life safety when vehicular communication services for the traffic safety related applications.The wireless links for vehicular communication are mainly divided into two categories: Vehicle-to-Vehicle(V2V)communication and Vehicle-to-Infrastructure(V2I)communication.The propagation characteristics of the vehicle-to-vehicle channel have been extensively studied.This thesis mainly studies the propagation characteristics and modeling analysis of the V2 I channel at 2.4 GHz,including the path loss,large-scale fading and small-scale fading.The applicable scenario we considered is that the roadside trees constitute the main obstacles,this vehicular communication scenario widely exists on some urban,suburban,and rural traffic roads,which is an essential category for deploying roadside units.The main work of this thesis is as follows:1.The path loss model of vehicle-to-infrastructure channel is studied.During the movement of the vehicle,the wireless signal may be blocked by the roadside trees,depending on the deployment height of the roadside infrastructure and the distance between the vehicle and the infrastructure.Based on the spatial information such as geographical location and geometric features,we classify the wireless link into two types: Line of Sight(LOS)link and Non-Line of Sight by foliage(NLOSf)link,in which the NLOSf link is blocked by the roadside tree canopies.For these two different link types,the proposed path loss model has high prediction accuracy,especially for the NLOSf link,the prediction accuracy is nearly 15 d B higher than the classical two-ray path loss model.The path loss models help to optimize the deployment strategy of roadside units,reduce unnecessary equipment investment and improve the benefit of VANETs system.2.The large-scale fading characteristics of the vehicle-to-infrastructure channel are studied.On the basis of the previous work,we further divide the vehicle-to-infrastructure wireless channel into three-segment links: the line-of-sight path transmits beneath the tree canopy,LOS-B,the non-line-of-sight path blocked by the canopy,NLOSf,and the line of sight path transmits above the tree canopy,LOS-A,they cover all altitude ranges deployed by the base station.For these three different link types,we analyze the large-scale fading characteristics of the vehicle-to-infrastructure channel,including the path loss exponent and the shadow fading analysis.The measurements verify the rationality of the proposed model.Accurately describing the large-scale fading characteristics of the vehicle-to-infrastructure channel is crucial to successfully predict the wireless network coverage and link budget.The study on the correlation and distribution of shadow fading can help to design the shadow fading margin and perfect the propagation model library.3.The small-scale fading characteristics of the vehicle-to-infrastructure channel are studied.Firstly,the time correlation caused by vehicle movement and the frequency selectivity caused by multipath effect are analyzed;Secondly,we analyze the statistical characteristics of the V2 I channel,mainly the statistic variations of fading depth,level crossing rate,average fading duration and envelop autocoefficient;Thirdly,we use the Akaike Information Criteria(AIC)method and Kolmogorov-Smirnov(K-S)method to come to a conclusion that the Ricean distribution is the most suitable distribution function for describing the V2 I channels;Finally,we estimate the Ricean K-factor with the classical moment-based method and establish a distance-based time-varying channel distribution parameter prediction model.The small-scale fading effect of the V2 I channel characteristics reflects the microscopic variation of the received power,and helps to design the physical layer coding and frame structure of VANETs system.It is of great significance for the future implementation of V2 I communication-based security applications.In summary,this thesis conducts wireless channel measurement and modeling research on the vehicle-to-infrastructure communication,which provides a tool for predicting electromagnetic wave propagation for the design of the vehicle-to-infrastructure wireless communication system.This thesis makes up the inadequate theory of channel modeling in this scenario,and it is the basis for constructing and perfecting the VANETs channel model library.