A measurement campaign to assess sources of error in microwave link rainfall estimation
<p>We present a measurement campaign to address several error sources associated with rainfall estimates from microwave links in cellular communication networks. The core of the experiment is provided by three co-located microwave links installed between two major buildings on opposite side...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2018-08-01
|
| Series: | Atmospheric Measurement Techniques |
| Online Access: | https://www.atmos-meas-tech.net/11/4645/2018/amt-11-4645-2018.pdf |
| Summary: | <p>We present a measurement campaign to address several error sources associated
with rainfall estimates from microwave links in cellular communication
networks. The core of the experiment is provided by three co-located
microwave links installed between two major buildings on opposite sides of
the small town of Wageningen, approximately 2 km apart: a 38 GHz formerly
commercial microwave link, as well as 26 and 38 GHz (dual-polarization)
research microwave links. Transmitting and receiving antennas have been
attached to masts installed on the roofs of the two buildings, about 30 m
above the ground. This setup was complemented with an infrared large-aperture
scintillometer, installed over the same path, as well as five laser
disdrometers positioned at several locations along the path and an automated
rain gauge. Temporal sampling of the received signals was performed at a rate
of 20 Hz. The setup was monitored by time-lapse cameras to assess the state
of the antennas as well as the atmosphere. The experiment was active between
August 2014 and December 2015. Data from an existing automated weather
station situated just outside Wageningen was further used to compare and to
interpret the findings. In addition to presenting the experiment, we also
conduct a preliminary global analysis and show several cases highlighting the
different phenomena affecting received signal levels: rainfall, solid
precipitation, temperature, fog, antenna wetting due to rain or dew, and
clutter. We also briefly explore cases where several phenomena play a role. A
rainfall intensity (<i>R</i>) – specific attenuation (<i>k</i>) relationship was
derived from the disdrometer data. We find that a basic rainfall retrieval
algorithm without corrections already provides a reasonable correlation to
rainfall as measured by the disdrometers. However, there are strong
systematic overestimations (factors of 1.2–2.1) which cannot be attributed
to the <i>R</i>–<i>k</i> relationship. We observe attenuations in the order of 3 dB
due to antenna wetting under fog or dew conditions. We also observe
fluctuations of a similar magnitude related to changes in temperature. The
response of different makes of microwave antennas to many of these phenomena
is significantly different even under the exact same operating conditions and
configurations.</p> |
|---|---|
| ISSN: | 1867-1381 1867-8548 |