MICRO AND MACRO SCALE SPATIAL RAIN VARIATION Part 1: Slant path attenuation for EHF systems and cons

1
MICRO AND MACRO SCALE SPATIAL RAIN
VARIATIONPart 1 Slant path attenuation for EHF
systems and considerations for long and medium
range diversity gainPart 2 Dynamic millimeter
wave communications in the presence of rain

• Sarah Callaghan
• University of Portsmouth and
• Radio Communications Research Unit, RAL
• S.A.Callaghan_at_rl.ac.uk

Cristina Enjamio University of Portsmouth and
University of Vigo cristina_at_ee.port.ac.uk
2
Macro and Micro Scales
3
Slant path attenuation for EHF systems and
considerations for long and medium range
diversity gain
Sarah Callaghan University of Portsmouth and
Radio Communications Research Unit,
RAL S.A.Callaghan_at_rl.ac.uk
4
Introduction
Satellite systems operating at EHF frequencies
are very affected by the presence of rain, light
rain and clouds along the slant path.
Attenuation statistics measured in the South of
England for Ku, Ka and V-band
5

• Attenuation is unlikely to be compensated for by
  available fade margin alone.
• Therefore need to design effective fade
  mitigation techniques
• Two major techniques are
• Time diversity
• Site diversity

To effectively design fade mitigation techniques,
it is necessary to accurately model the spatial
and temporal structure of rain.
6
STENTOR Experiment
Artist's impression of STENTOR
Due for launch end 2001. Beacon frequencies 20.7
and 41.4 GHz
Schematic map of locations of beacon receivers
7

• Chilbolton Advanced Meteorological
• Radar.
• 25 m steerable antenna
• 3 GHz Doppler-Polarization radar
• operational range of 100 km
• beam width of 0.25 degrees
• max angular velocity 1 degree / second

CAMRa
8
Storm event Details

• recorded on the 1st May 2001
• 124 near horizontal scans
• measured over an angle of 80 degrees
• interpolated onto a square Cartesian grid, with
  a grid spacing of 300m and a side length of
  56.2km
• data points

9

• Contours of equal (log) rain rate
• used MATLABs predefined contour function at
  specific values of (log) rain rate to determine
  contour lines

10
Box counting method Count number of boxes
required to cover length of each contour
line. Repeat, using boxes of different side
lengths. Plot on graph as ln(1/box size) vs
ln(number of boxes). Slope of best fit line is
box counting dimension.
11
Box counting results for raster 25 (for each
separate contour line having more than 100
vertices)
Sample values for different contour
values Contour values Box counting (log
rain rate) dimension 1
1.12 0.75 1.23 0.5
1.21 0.25 1.15 -0.25
1.21 -0.5 1.15 -0.75
1.18 -1 1.24
12
Box counting results for all contours in raster 25
Contour values Box counting (log rain rate)
dimension 1 1.18 0.75
1.23 0.5 1.25 0.25
1.25 -0.25 1.28 -0.5
1.23 -0.75 1.13 -1 1.15
13
Box counting dimension for all rasters in storm
event.
Contour values Box counting (log rain rate)
dimension 1 1.17 0.75
1.22 0.5 1.25 0.25
1.28 -0.25 1.26 -0.5
1.20 -0.75 1.15 -1 1.17
14
Radar picture from CAMRa taken along slant path
to ITALSAT
Corresponding attenuation time series experienced
by beacon on ITALSAT
15
Concluding remarks

• Slant path systems operating at EHF suffer from
  attenuation that cannot be compensated for by
  available fade margin alone.
• Need to further understand and accurately model
  spatial and temporal distribution of rain.
• Fractal nature of rain rate contours has been
  established, confirming other work done on the
  fractal nature of the spatial variation of rain.
• Work ties in with others working on different
  scales, working towards a global understanding
  covering micro and macro scales.

16
References

• Lovejoy, S., Area-Perimeter Relation for Rain
  and Cloud Areas, Science, Vol 216, 185-187, April
  1982
• Rys, F.S., Waldvogel, A. Fractal Shape of Hail
  Clouds, Physical Review Letters, Vol. 56, Number
  7, 784-787, February 1986
• Klinkenberg, B., A fractal analysis of shadowed
  and sunlit areas, Int. Jnl. Remote Sensing, Vol.
  15, No. 5, 967-977, 1994