PROJECT TITLE :

Empirical Geometry-Based Random-Cluster Model for High-Speed-Train Channels in UMTS Networks

ABSTRACT:

During this paper, a recently conducted measurement campaign for high-speed-train (HST) channels is introduced, where the downlink signals of an in-service Universal Mobile Terrestrial System (UMTS) deployed along an HST railway between Beijing and Shanghai were acquired. The channel impulse responses (CIRs) are extracted from the info received in the common pilot channels (CPICHs). At intervals 1318 km, fourteen4 base stations (BSs) were detected. Multipath parts (MPCs) estimated from the CIRs are clustered and associated across the time slots. The results show that, limited by the sounding bandwidth of three.84 MHz, most of the channels contain a single line-of-sight (LoS) cluster, and the remainder consists of several LoS clusters due to distributed antennas, leaking cable, or neighboring BSs sharing the identical CPICH. A replacement geometry-primarily based random-cluster model is established for the clusters' behavior in delay and Doppler domains. Different from conventional models, the time-evolving behaviors of clusters are characterised by random geometrical parameters, i.e., the relative position of BS to railway, and also the train speed. The distributions of these parameters, and the per-cluster path loss, shadowing, delay, and Doppler spreads, are extracted from the measurement knowledge.