Optimization of radar scan rate and data quality using phasedarray radar

1
Optimization of Weather Update Time and Data
Quality Using Phased-Array Radar
Tian-You Yu, Marko Orescanin University of
Oklahoma, School of Electrical and Computer
Engineering, Atmospheric Radar Research
Center Christopher Curtis, Dusan Zrnic, and
Douglas Forsyth NOAA/National Severe Storms
Laboratory
Acknowledgement This project is supported
through the NOAA/NSSL NA17RJ1227 and DoD EPSCoR
N00014-06-1-0590
Presenting at 23rd Conference on IIPS 2007
2
Content

• problem statement
• Beam multiplexing (BMX) using the phased-array
  weather radar
• Experimental results using NWRT
• Conclusions

3
Limitation of Scan Rate in Weather Radar
The data update time for operational WSR-88D is
4-10 min, which is not sufficient for
fast-evolving weather
Goal Increase the data update time and maintain
the same data accuracy
4
Conventional Sampling
7
8
5
6
4
1
13
11
12
9
10
M samples
0o
Sample time
1o
Sample time
2o
Sample time
13o
Sample time
5
Conventional Sampling
0o
Sample time
1o
Sample time
2o
Sample time
13o
Sample time
6
Beam Multiplexing (BMX)
13
2
9
11
5
7
1
12
8
10
4
0o
Sample time
7o
Sample time
1o
Sample time
13o
Sample time
7
Beam Multiplexing (BMX)
0o
Sample time
1o
Sample time
2o
Sample time
13o
Sample time
8
Consideration of BMX
9
Improvement Factor
A reduction in data acquisition using BMX when
the same accuracy obtained by the conventional
sampling is maintained.
10
National Weather Radar Testbed (NWRT)
The PAR at NWRT operates at 3.2 GHz with peak
power of 750 kW The SPY-1A is a passive
phased-array with 4352 elements Beam width is
1.5o at broadside and 2.1o at /- 45o off
broadside Antenna is tilted 10o in elevation and
is located 12 m above the ground
11
Experiment at NWRT
12
Experiment of BMX
13
Conclusions

• Beam multiplexing (BMX) is developed to improve
  data acquisition time while the data accuracy is
  still maintained.
• A phased array radar is an ideal instrument to
  implement BMX
• Statistical analysis have shown that BMX can
  provide the mean field of Z and v is consistent
  with those obtained by conventional scanning.
• The statistical variance of radar measurements
  can be reduced using BMX, which can be translated
  to the improvement in data acquisition time.