Hough-Transform-Based Cluster Identification and Modeling for V2V Channels Based on Measurements

Xuesong Cai,Bile Peng,Xuefeng Yin,Antonio Perez Yuste
DOI: https://doi.org/10.1109/tvt.2017.2787731
IF: 6.8
2018-01-01
IEEE Transactions on Vehicular Technology
Abstract:In this paper, a recently conducted measurement campaign for vehicle-to-vehicle (V2V) propagation channel characterization is introduced. Two vehicles carrying a transmitter and a receiver, respectively, have been driven along an eight-lane road with heavy traffic. The measurement was conducted with 100 MHz signal bandwidth at a carrier frequency of 5.9 GHz. Channels are observed consisting of two kinds of channel components, i.e., time-evolving clusters and clutter paths. A novel approach based on Hough transform is proposed to identify the clusters. Based on the cluster identification results, channel characteristics in composite, intracluster, and time-variant levels are analyzed. The parameters investigated include the composite root-mean-square (RMS) delay spreads and power decay versus delay behaviors of clusters and clutter paths, cluster RMS delay spread, cluster RMS Doppler frequency spread, correlations of cluster parameters, and coherence time of parameters of interest. The statistics constitute an empirical stochastic clustered-delay-line channel model focusing on the wideband characteristics observed in the realistic time-variant V2V propagation scenario.
What problem does this paper attempt to address?