Detailed Analysis of Directional Radio Channel Measurements

Physical models describe the radio channel by characterizing the multipath propagation. Also, measurements are frequently employed in the establishment of such models, or for tuning the existing ones. This paper presents a systematic analysis of double-directional radio channel data, obtained by combining two single-input multiple-output measurements (forward and reverse), each, containing the channel frequency responses at the locations of a virtual rectangular antenna array. The measurement setup uses a vector network analyzer, which, together with the adopted data processing, represents an affordable, but very consistent, alternative to more sophisticated equipment, given that, a physical analysis of the identified channel contributions provided sensible results. The study comprised three main steps: 1) the characterization of each single-directional channel by employing a high-resolution algorithm to obtain the parameters of a chosen number of multipath components (MPCs); 2) the classification of the obtained MPCs in a twofold way, first, by grouping MPCs using clustering tools, and, second, by classifying these clusters according to their type of interaction with the propagation scenario; and 3) the extraction of typical values for the number of clusters, the number of MPCs per cluster, rules for power decay, and statistical distributions for inter and intracluster delays and azimuths.