A Statistical Multi-Cluster Indoor Channel Model in an Office Environment for Ultra-Wideband Signals
This thesis presents a new statistical multi-cluster indoor channel model for ultrawideband (UWB) communications between 0.1 to 2 GHz. The proposed model has been developed based on 924 measured channel impulse responses conducted at 24 different locations over a single floor in a typical modem o...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English |
| Published: |
2005
|
| Subjects: | |
| Online Access: | http://psasir.upm.edu.my/id/eprint/6049/1/FK_2005_43.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This thesis presents a new statistical multi-cluster indoor channel model for ultrawideband
(UWB) communications between 0.1 to 2 GHz. The proposed model has
been developed based on 924 measured channel impulse responses conducted at 24
different locations over a single floor in a typical modem office environment.
The multi-cluster channel characteristics are modeled according to 10 independent
channel parameters extracted from the large-scale and small-scale statistical
analysis of the received energy delay profiles (EDP). Three new channel parameters
were introduced to better describe the time and energy dispersion statistics of the
UWB signal. They are the maximum excess delay (MED), decay factor, y within
the MED and energy gains variations between adjacent bins, X. The pulse propagation characteristics were further modeled according to several
path topogaphies. In addition, two new path loss prediction models were developed
to take into account the attenuation factor (AF) due to gypsum and brick wall, the
two main obstructing walls in this experiment.
The proposed multi-cluster channel model improves the prediction accuracy of the
UWB pulse characteristics for various path topographies due to its detailed
parameterization of the channel statistics. In comparison, the channel model of
Cassioli et a]. in 2001 was developed using only three independent lognormal
parameters, i.e. shadowing, o, decay factor, y and the energy gains ratio, r.
Furthermore, Cassioli et al. proposed a single cluster channel model without
topographical classification. The channel model of Alvarez et al. in 2003 however
was developed based on five channel parameters extracted from analysis conducted
in the frequency domain. This method does not accurately capture the time delay
statistics of the transmitted pulse.
The experimental results indicate the U WB signal's high
immunity against fades
and its attractiveness for propagation in tunnels. In addition the UWB channel was
observed to exhibits a wide-sense-stationary-uncorrelated-scattering (WSSUS)
characteristic which simplifies the transfer function between the time and frequency
domains via a single Fourier transform. |
|---|