G Tsao, P Iyamu, L Petropoulakis, R Atkinson, I Andonovic, I A Glover
Department of Electronic and Electrical Engineering
University of Strathclyde
Glasgow G1 1XW, UK
e’mail: gavin.tsao@strath.ac.uk and osayi.iyamu@gmail.com and {l.petropoulakis, r.atkinson, i.andonovic,
ian.glover.}@eee.strath.ac.uk
!! "
# " $
% !! &
’ ’ ((
I. INTRODUCTION
Demand for higher wireless data’rates in applications
such as multimedia streaming within the home environment
continues to increase [1]. Ultra’wideband (UWB) and
Multiple’Input Multiple’Output (MIMO) technologies can
help in addressing this demand [1].
MIMO systems employ the use of spatial multiplexing,
taking advantage of the multipath wireless channel [2].
UWB uses large bandwidth to provide high channel
capacity. The large spectral occupancy of UWB signals
means their power density must be low, however, and this is
enforced by constraining UWB signals to a spectral mask
defined by the appropriate radio regulatory authority [3], [4].
Fig. 1 shows an example of a Federal Communications
Commission mask used in the USA. Fig. 2 shows a similar
mask defined by the Electronic Communication Committee
(ECC) for Europe.
Figure 1. FCC indoor spectrum mask for commercial systems
Figure 2. ECC indoor spectrum mask for commercial systems
It is possible to combine both UWB and MIMO
technologies to achieve higher data’rates than either
technology is capable of supporting alone. As with any
wireless communication system, the channel characteristics
determine the degree to which the theoretical channel
capacity can be realized [5–7].
The experiment described here measures channel
characteristics for a 22 , 23 , and 24 MIMO’UWB
systems in which one ‘terminal’ is at a fixed, elevated,
location and the other ‘terminal’ is moved systematically
over a large horizontal surface (a laboratory bench). The
intention is to emulate a ceiling’mounted access point and
portable, table’top, device. The maximum theoretical
channel capacity is then calculated for these particular, and
realistic, channels.
II. METHODOLOGY
A. Measurements
The channel frequency response (or S
21
scattering
parameter) was measured using a pair of omnidirectional
UWB antennas and an Agilent programmable network
analyzer (PNA) N5230A [8]. The PNA was set to sweep
between 1 GHz and 6 GHz with 16001 sampling points
giving a frequency resolution of 312 kHz. 20 frequency
response measurements were averaged to increase the
measurement signal to noise ratio (SNR). The environment
was static during these measurements.
!!" ! #$%
&
’
( !!" !
( #$%
() #$%
978-1-4673-4455-5/12/$31.00 ©2012 IEEE