Transcript Slide 1

Characterization of Atmospheric
Noise in the Loran-C Band
Presented to the International Loran Association (ILA-32)
November 6, 2003 Boulder, CO
Manish Lad
Frank van Graas, Ph.D.
David Diggle, Ph.D.
Curtis Cutright
Avionics Engineering Center
Outline
• Data Processing Overview
• Flight Test Results
 “Quiet (normal conditions)” data
collected in Ohio
“Thunderstorm” data collected in Florida
• Conclusions
Avionics Engineering Center
Flight Data Collection Equipment
• King Air C-90 B Aircraft
• LORADD-DS DataGrabber
• Novatel OEM4 GPS
receiver
• WX-500 StormScope
• Apollo 618 (Loran receiver)
• Data collection PC
King Air C-90B
Avionics Engineering Center
Wire vs. Loop Antenna Gain
• Theoretically, the dual-loop antenna has a 3-dB gain
advantage over the wire antenna
• Exact gain difference depends on the antenna installation
Gain = 0 dB
Gain = 3 dB
Note: SNR results are determined at the output of the
antenna.
Avionics Engineering Center
Processing Collected Data
•
•
•
•
•
•
•
Read Loran-C data
Identify and remove CW interference
Remove Thunderstorm bursts
Read GPS data
Identify and remove Loran-C chains
Calculate signal-to-noise ratio
Characterize atmospheric noise
Avionics Engineering Center
Loran-C Data Processing Overview
Collected data
GPS Time,
Position
Find filter
coefficients
Loran Processor
Bandstop filters
Track transmitters
Remove pulses that
are above noise floor
Noise power
Signal power
Calculate SNR
Noise Sequence
Noise Characterization
Noise distribution
Avionics Engineering Center
Removal of CW and Thunderstorm Bursts
Sampled data at
400 kSamples/sec
2-sec data
block
Bank of band-pass
filters
x
2
i
1-500 Hz
501-1000 Hz
1001-1500 Hz
199.5-200KHz
Integrate PCI’s for
identifying different
chains
x
x
2
i
2
i
Detect
CW
x
2
i
Remove bins with
Energy above a
set threshold
Avionics Engineering Center
Calculate
bandstop filter
coefficients
Filter the CW
from 2 second
data block
Calculate Energy
in bins
Loran processor
Signal after removal of CWs and Thunderstorm Bursts (if present)
Chain
information
Integrate signal as per PCI of chain
Identify Master and secondaries
Antenna
position
Compute signal power for each station
Remove Loran-C pulses that are above noise floor
Compute average noise power obtained after removal of pulses
Calculate the noise distribution
Calculate SNR for master and secondaries
Avionics Engineering Center
Flight Test Data
• Results obtained for different data sets
under diverse atmospheric conditions
(clear and Thunderstorm)
• Data collected in Ohio: NEUS chain
• Data collected in Florida: SEUS chain
Avionics Engineering Center
Flight Test Results
Normal conditions
Athens, Ohio
Collected on August 13, 2003
Avionics Engineering Center
Ground Path (Athens, Ohio)
Flight test trajectory
near Athens, Ohio
Avionics Engineering Center
E–field Data Example: Time Domain
Note: Signal Amplitude is in A/D levels
Avionics Engineering Center
E–field Data Example (Cont’d)
Before and after processing the 2-second data chunk
Avionics Engineering Center
E-field Noise Statistics (Athens, Ohio)
Noise Distribution
Number of
samples
 10 to 10
Avionics Engineering Center
E-field Noise Statistics (Cont’d)
Calculated and Gaussian cdf
cdf
(1-cdf)
 10 to 0
0 to 10
Cumulative
probability
Avionics Engineering Center
H–field Data Example (Athens, Ohio)
Avionics Engineering Center
H–field Data Example (Cont’d)
Before and After Processing the 2-second data chunk
Avionics Engineering Center
H-field Noise Statistics (Athens, Ohio)
Noise Distribution
Number of
samples
 10 to 10
Avionics Engineering Center
H-field Noise Statistics (Cont’d)
Calculated and Gaussian cdf
cdf
(1-cdf)
Cumulative
probability
 10 to 0
0 to 10
Avionics Engineering Center
Results (Athens, Ohio)
SNR measurements (average of 326 seconds) at the
output of the antenna for NEUS chain
Antenna
SNR M (avg)
Seneca, NY
SNR Z (avg)
Dana , IN
Wire
(E-field)
10.5
13.3
Loop
(H-field)
12.4
13.2
Avionics Engineering Center
Flight Test Results
Daytona Beach, Florida
Collected in the Vicinity of
Thunderstorms on August 14, 2003
Avionics Engineering Center
Ground Path (Daytona Beach, FL)
Flight test trajectory
near Daytona
Beach, FL
Avionics Engineering Center
E-field Data Example (Daytona Beach, FL)
Signal
Amplitude
Number of samples (2 seconds of data)
Note: Dynamic range of data collection equipment is 96 dB (16 bits)
Avionics Engineering Center
E-field Data Example (Cont’d)
Before and After Processing the 2-second data chunk
Avionics Engineering Center
E-field Data Example (Cont’d)
Signal
Amplitude
Number of samples (2 seconds of data)
Avionics Engineering Center
E-field Data Example (Cont’d)
Before and After Processing the 2-second data chunk
Avionics Engineering Center
E-field Data Example (Cont’d)
Avionics Engineering Center
E–field Noise Statistics (Daytona Beach, FL)
Noise Distribution
Number of
samples
 10 to 10
Avionics Engineering Center
E–field Noise Statistics (Cont’d)
Calculated and Gaussian cdf
cdf
(1-cdf)
 10 to 0
0 to 10
Cumulative
probability
Avionics Engineering Center
H–field Noise Statistics (Daytona Beach, FL)
Noise Distribution
Number of
samples
 10 to 10
Avionics Engineering Center
H–field Noise Statistics (Cont’d)
Calculated and Gaussian cdf
cdf
(1-cdf)
Cumulative
probability
 10 to 0
0 to 10
Avionics Engineering Center
Results (Daytona Beach, FL)
SNR (average of 326 seconds) measurements at the output
of the antenna for SEUS chain
Antenna
SNR M (avg)
Malone, FL
SNR Y (avg)
Jupiter, FL
Wire
(E-field)
7.7
8.7
Loop
(H-field)
9.2
12.5
Avionics Engineering Center
Flight Test Results
Palm Coast, Florida
Collected in the Vicinity of
Thunderstorms on August 14, 2003
Avionics Engineering Center
Ground Path (Palm Coast, FL)
Flight test trajectory
near Palm coast, FL
Avionics Engineering Center
E–field Noise Statistics (Palm Coast, FL)
Noise Distribution
Number of
samples
 10 to 10
Avionics Engineering Center
E–field Noise Statistics (Palm Coast, FL)
Calculated and Gaussian cdf
cdf
(1-cdf)
Cumulative
probability
 10 to 0
0 to 10
Avionics Engineering Center
H–field Noise Statistics (Palm Coast, FL)
Noise Distribution
Number of
samples
 10 to 10
Avionics Engineering Center
H–field Noise Statistics (Palm Coast, FL)
Calculated and Gaussian cdf
cdf
(1-cdf)
Cumulative
probability
 10 to 0
0 to 10
Avionics Engineering Center
Results (Palm Coast, FL)
SNR (average of 326 seconds) measurements at the output
of the antenna for SEUS chain
Antenna
SNR M (avg)
Malone, FL
SNR Y (avg)
Jupiter, FL
Wire
(E-field)
12.6
15.2
Loop
(H-field)
13.8
18.4
Avionics Engineering Center
Conclusions
• SNR at the output of Loop (H-field) antenna is
generally greater than the SNR at the output of
Wire (E-field) antenna by 2-3 dB
• Noise distribution
Core of distribution looks Gaussian
Tail probabilities are much larger than Gaussian
with an equivalent rms value (looks like 3sigma)
• Data collected from both the antennas closely
match in the calculated cdf of the noise
Avionics Engineering Center
Acknowledgements
• Federal Aviation Administration (FAA)
Mitch Narins (Loran Program Manager)
• Reelektronika B.V.
Dr. Durk van Willigen, Wouter Pelgrum
• King Air Crew
Bryan Branham, Jay Clark
Avionics Engineering Center
Questions ?
Avionics Engineering Center