Title :
Cluster analysis of wireless channel impulse responses
Author_Institution :
Inst. Commun. & Wave Propagation, Graz Univ. of Technol., Austria
Abstract :
This paper introduces a novel wireless channel clustering technique based on the Saleh-Valenzuela channel model. The impulse response is regarded as a single realization of the statistical channel model based on which the prior density functions of cluster arrival times are derived. Cluster analysis is done by means of extending the original Saleh-Valenzuela model by allowing different cluster arrival rates. It is shown that the estimation of cluster arrival rates is essentially equivalent to clustering the impulse response and estimating the expected cluster centers. Having defined the clusters, other parameters of the Saleh-Valenzuela model can be estimated from a single realization. The proposed method has been applied to simulated as well as measured channel impulse responses.
Keywords :
pattern clustering; statistical analysis; telecommunication channels; transient response; Saleh-Valenzuela channel model; cluster analysis; cluster arrival times; cluster center; density function; statistical channel model; wireless channel impulse response; Baseband; Clustering algorithms; Data mining; Delay; Density functional theory; Fading; Paper technology; Probability density function; Transceivers; Wideband;
Conference_Titel :
Communications, 2004 International Zurich Seminar on
Print_ISBN :
0-7803-8329-X
DOI :
10.1109/IZS.2004.1287404