Experimental Work & Field Testing
 Much
of the company's work is concerned with the mathematical modelling
and simulation of radio systems, and the interaction (e.g.
interference) between them. Generally, this simulation work can make
use of established models and algorithms, such as those contained in
the ITU-R Recommendations.
It is sometimes the case, however, that no suitable models exist,
or that the existing models do not cover the required scenarios
(frequency, environments, etc.). In such cases it may be necessary
to gather new data as the basis for developing new models.
A recent example of such a measurement campaign was a study,
funded under the British National Space Centre SATCOM programme,
that sought to determine the degree to which radio signals to or
from satellites are attenuated as they pass in or out of typical
buildings. This information is important both for the designers of
satellite radio systems that may be used indoors (e.g. satellite DAB
or mobile communications) and in estimating the interference to
satellite uplink receivers from terrestrial transmitters that share the same band (such as
wireless LAN systems). To explore the
dependence of such building penetration losses on the path elevation
angle, a tethered helium balloon was used, and this was raised and
lowered to explore the required geometries.
Measurements at 1.3 GHz, 2.4 GHz and 5.7 GHz were made using
small transmitters carried by the balloon. The power received inside
the buildings was sampled over a range of several wavelengths using
a computer-controlled antenna positioner. Full details of the
results of this study may be found in a paper given at ICAP.
 |
|
 |
| Correlating Receiver |
|
Typical channel
impulse response (outdoor-to-indoor, 2.4 GHz, x-axis 0-60 ns) |
A further project for BNSC investigated the wideband characteristics of the outdoor-indoor
radio channel at 2.4 GHz. The sliding correlator constructed for these measurements made use
of pseudo-noise code sequences, clocked at 100 MHz. These sequences have excellent autocorrelation
characteristics, and offer an efficient approach to channel sounding. The reference sequence
is clocked at a slightly lower rate than the transmitted sequence, and the two codes therefore
drift past each other. The received signal is thus 'explored' by the reference sequence, and
the impulse response of the channel determined. The sliding correlator architecture avoids the
need for high-speed data capture, as the correlator output is dilated by the ratio of slip rate
to clock rate. This correlator system was used to obtain statistics relevant to typical UK
domestic building types, at high temporal resolution.
 |
|
More recently, on behalf of Ofcom, we have undertaken a campaign of outdoor measurements,
for which a lower resolution is appropriate. This system, clocked at 10 MHz, operated at three
frequencies, 237 MHz, 1477 MHz and 3430 MHz. Because of the lower clock rate, it is straightforward
to sample the received signal directly at IF, and to perform the de-correlation in software. This
dramatically reduces the time taken to sample each delay profile, allowing mobile measurements to
be carried out. |
| DSP-based channel sounder showing receiver (above) and 3-band transmitter (below) |
|
When the power delay profiles are plotted, then discrete reflections can be seen, with relative delays that vary along the route. These
delay values can be used to construct a series of ellipses, allowing the reflection points to be determined.
In the example shown below, one of these reflection points is a large office block ("Brown & Root") to the north-west
of the measurement area; a second reflection point could not be firmly identified, but appears to come
from part of a hospital complex ("Ingleby House"); a reflection from a nearby gasometer is also apparent.
 |
|
 |
| Power delay profile measured along 3.3 km route (1477 MHz) |
|
Delay ellipses from mobile measurement |
Other recent measurement work has involved the mobile recording of field strength at VHF, UHF and
microwave frequencies to inform the development and verification of propagation models. For these
measurements, position is logged by GPS, while samples are triggered at regular distance intervals
by a transducer linked to the vehicle speedometer drive. A typical received data series is shown below,
and compared with the prediction from a model currently under development.
 |
|
 |
| Transmitting aerials for 3-band propagation measurements/i> |
|
Field strength comparison |
To discuss how Aegis can help your organisation to plan and
implement its measurement and field testing requirements, contact us
at enquiry-2008@aegis-systems.co.uk.
See also:
|
|
|
|
|
