Research On Modeling Of E-band Low-altitude Air-ground Channel

With the rapid development of aviation industry in China,the military and civil fields have put forward higher and higher requirements on the performance of aviation communication.At present,small bandwidth and crowded frequency spectrum restrict the capacity expansion of aviation communication,and millimeter wave communication provides a solution.Compared with millimeter wave frequency band,E-band(71-76 GHz,81-86 GHz)has smaller longdistance transmission loss and larger available continuous bandwidth.Introducing it into aviation communication scenario can not only effectively relieve the problem of spectrum strain but also meet the demand of high bandwidth and high transmission rate of low altitude channel.Accurate and efficient channel model is an indispensable and important means for communication system design and performance evaluation and test.However,the traditional air-to-ground channel model cannot accurately represent the characteristics of E-band aviation channel.Based on this,the research on E-band low-altitude air-to-ground channel modeling in this paper has important theoretical significance and practical value.(1)The basic characteristics and important parameters of the wireless channel are summarized,and the advantages and problems of the E-band in the wireless mobile communication scenario are analyzed.Traditional channel modeling methods and the channel modeling methods based on machine learning are studied in this paper,and the basic principle and construction ideas are discussed in detail,which lays a foundation for the research in the following chapters of this paper.(2)The rainfall attenuation,large scale fading and small scale fading of E band in aviation channel scenario are discussed in detail.ITU-R rain decline model is used to predict the rain decline in various urban areas in China,and it is proved that rainfall attenuation is an important factor affecting the E-band low altitude communication.Secondly,in the absence of channel measurement data,large-scale and small-scale wireless channel modeling is carried out based on the two-line model to fully explore the channel characteristics of E-band in the low-altitude scene.Based on the simulation analysis of the traditional Clarke model and Jakes model,an improved simulation model is proposed.The randomness of the Angle of arrival formula makes the autocorrelation function curve of the whole model more consistent with the theoretical model.Finally,starting from the three states in the process of flight,respectively discusses the navigational state,stop state and take-off/landing of large scale fading and small scale fading way of modeling,and the envelope curve of channel and the probability density curve of the simulation and comparison analysis,the results verify the correctness and effectiveness.(3)A large-scale fading prediction model of low altitude channel based on machine learning algorithm is proposed,and the attenuation data is calculated through the electromagnetic simulation software Wireless Insite in the Ottwa urban area.The coordinate values,distances and path losses between all the receiver paths and the transmitting points in the modeling are taken as the data sets.The parameters on the Route2 path were selected as the testing set,and then regression algorithms such as KNN and random forest were used for training to complete the prediction analysis of the large-scale fading of the urban area.The simulation results verify the superior performance of machine learning algorithm in the large-scale fading prediction of channel.