Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Look up keyword
Like this
1Activity
0 of .
Results for:
No results containing your search query
P. 1
Detecting Oceanic Wind Parameers by Using GPS by a Haider

Detecting Oceanic Wind Parameers by Using GPS by a Haider

Ratings: (0)|Views: 6|Likes:
Published by adnanscientist33

More info:

Published by: adnanscientist33 on May 08, 2010
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

07/11/2013

pdf

text

original

 
2
ABSTRACT
Global Positioning System (GPS) signals reflected fromthe ocean surface have potential use for various remotesensing purposes. Some possibilities are measurements of surface roughness characteristics from which wave height,wind speed and direction could be determined. In thispaper, recent reflected GPS measurements collected viaaircraft with a delay mapping GPS receiver, are used toexplore the possibility of determining ocean surface windspeed and direction during Hurricanes Michael and Keithin October, 2000. To interpret the GPS data, a theoreticalmodel is used to describe the correlation power of thereflected GPS signals for different time delays as afunction of geometrical and environmental parameters.Wind direction estimates are based on a multiple satellitenon-linear least squares solution. The estimated windspeed using surface-reflected GPS data collected at avariety of wind speed conditions shows an overallagreement better than 2 m/s with data obtained from
Detecting Oceanic wind parameters by using GPS
A Haider azaidi2@buffalo.edu Research Scientist
Dept of Electrical Engineering State University of New York
 
2
INTRODUCTION
The use of GPS as a forwardscatter remote sensing toolhas come to fruition in the last few years (Katzberg[1996], Garrison et al. [1997, 1998, 1999], Komjathy etal. [1998, 1999], Lin et al. [1999], Armatys [2000]).NASA researchers Stephen J. Katzberg of LaRC andJames L. Garrison now at Purdue University havedetermined the properties of the ocean-reflected signaland have developed a specialized GPS receiver called theDelay-Mapping Receiver (DMR) to measure the reflectedsignals [Garrison et al., 1997].Other investigations of ocean reflected GPS signals arebeing conducted by the Jet Propulsion Laboratory (JPL)and the European Space Agency (ESA), focused primarilyon the application of reflected GPS signal tracking toaltimetry [Martin-Neira, 1993]. These groups haveconducted a number of experiments from static locationsand aircraft, and investigated signals received from aspaceborne antenna [Lowe et al., 2000a, 2000b].Using the DMR and models to predict the interaction of the L1 GPS signal at 1575.42 MHz, researchers at NASALangley Research Center, Purdue University and theUniversity of Colorado at Boulder have been able toestimate wind speeds on the ocean surface with anaccuracy of about 2 m/s. Results to date have advancedthe understanding of reflected GPS signals and provideddirect experimental evidence of their application to oceanremote sensing and mapping.The model employed in this paper
s estimation processwas developed by Zavorotny and Voronovich (Z-V) and isdocumented extensively in Clifford et al. [1998],Zavorotny and Voronovich [2000] and Komjathy et al.[2000]. The Z-V model employs a forwardscatter radarequation with the geometric optics limit of the Kirchhoff Approximation. For backscattering theory andapplications, the interested reader is referred to Ulaby etal. [1986]. The wave spectra used by the Z-V model isthat from Elfouhaily et al. [1997].The Z-V model takes the form:
( ) ( )
∫ 
Λ
422
,
 z2200222ic
q R R4 R)/c]+ R(-[)(  D= ) f Y(
ρ ρ π τ ρ τ 
   
( )
[ ]
ρ ρ 
qqqP f  f S
2 zc D
42
      ×
 
. (1)where
2
,
) f Y(
c
τ 
is the reflected power for any delaybin
τ 
and Doppler offset
c
 f 
;
i
is the integration time inseconds;
is the complex reflectivity of the ocean at L1;
 D
is the antenna gain;
Λ
is the correlation function of the GPS C/A code;
S
is the Doppler sync function;
P
isthe probability density function (PDF) of the surfaceslopes;
q
is the magnitude of the scattering vector
q
 
;
 R
0
is the distance from some point on the surface point to theGPS satellite;
 R
is the distance from the GPS receiver tosome point on the surface;
c
is the speed of light;
 D
 f 
is theDoppler shift at the specular point;
c
 f 
is the compensationfrequency or the Doppler offset to some point
ρ 
 
; and
ρ 
 
is a vector from the specular point to some other pointon the surface. For our aircraft experiments discussed inthis paper, equation (1) can be simplified and
S
set tounity.
INSTRUMENT AND DATA
 
The use of GPS in a bistatic radar configuration tomeasure surface properties relies upon our ability toextract information from the reflected signal. For standardGPS navigation applications, the receiver
s main functionsare to measure the signal delay from the satellite (thepseudorange measurement) and the rate of change of therange (the Doppler measurement) (see e.g., Parkinson etal. [1996]). Conversely, in our remote sensing application,the primary measurement is the received power from areflected signal for a variety of delays and Doppler values.The basis of this measurement and its sensitivity to thesurface conditions is discussed in the followingparagraphs.
 
The Delay Mapping Receiver (DMR) is a softwareconfigurable GEC Plessey (now Mitel Semiconductor)GPS Builder-2 receiver modified to observe reflectedLHCP signals from two GPS satellites and to recordcorrelator power at 10 consecutive half-chip intervals. Thehalf-chip intervals are analogous to range bins in a radarreceiver and are used to isolate power reflecting from aspecific region on the ocean surface. Signals reflectedfrom the ocean surface originate from a glistening zone(see Figure 1a) surrounding a nominal specular reflectionpoint. The size and shape of the glistening zone arefunctions of the roughness of the ocean surface. Tomeasure the reflected power from this glistening zone, thereceiver-generated pseudorandom noise codes are delayedin time with respect to directly received, line-of-sightsignals. This isolates power originating from the region
 
3surrounding the specular reflection point. The shape of theresulting waveform of power-versus-delay is dependentupon the roughness of the ocean surface (see Figure 1b).This roughness is in turn a function of the surface windspeed and direction, and therefore provides a means toretrieve these geophysical parameters from the GPSreflected signal power measurement.Figure 1a. Illustration for ocean-reflected GPS signals.
+1-10
   C  o  r  r  e   l  a   t   i  o  n   P  o  w  e  r
Delay Relative to Specular Reflection Point(Code Chips)Peak PowerTrailing EdgeIdeal AutocorrelationFunction
Figure 1b. Correlation function shapes for ideal directGPS signal and for reflected signals from rough surfaces.
METHOD FOR WIND SPEED AND WINDDIRECTION RETRIEVAL
Preprocessing the Data
. Before the estimator can makeuse of the reflected GPS data, they must be preprocessed.Preprocessing takes place in several steps. First, the noisefloor is computed for each data set. This is done bycomputing the mean of all the points before the firstcorrelation peak of the reflected signal. After computingthe noise floor, it is subtracted from all data points. Thesereflected data points are then normalized by dividing bythe total reflected power. Normalization is necessary toremove effects of uncalibrated receiver gains. The totalreflected power is computed by summing all of thecorrelation measurements for one waveform, essentiallyintegrating the correlation waveform. Total reflectedpower is chosen for normalization because it should benearly constant due to conservation of energy. Finally, thedata is broken into one-minute segments for use by theestimator.An estimate of the path delay,
τ 
, is computed using post-processed positions of the satellite and receiver. Thesatellite positions are interpolated from International GPSService
s (IGS) 15-minute precise positions, and thereceiver positions are interpolated using the receiver
snavigation solution. Using these positions and the WGS-84 ellipsoid, an estimate of the specular reflecting pointcoordinates on the earth surface is computed. The pathdelay is then estimated from these three positions.Because the delay variable computed from the receiverand satellite geometry with respect to the WGS-84ellipsoid may contain errors, a shifting parameter isintroduced. A scaling parameter is also introduced whichcompensates for errors in the assumptions made duringnormalization of the measured power. Because the totalpower measured over the 10 delay bins fails to includepower over the same range of delay as the modeledwaveform, inclusion of a scale factor accounts for thisdiscrepancy during normalization. During preprocessing,we also quality-check the data and eliminate outliers bycomputing the mean and standard deviation of thereflected power in each delay bin.
 Main Processor 
. The state for the estimation processcontains wind speed, wind direction and as an option, pathdelay error estimates, and scaling parameters can also besimultaneously estimated. For routine data processing, thesoftware is able to estimate path delay errors by aligningthe waveform leading edges.The basis for wind direction determination is that the PDFof surface slopes is wider in the direction of the wind.This produces an asymmetry in the glistening zone. Withdelay measurements from a single satellite, it is notpossible to unambiguously identify this asymmetrydirection because the integration over a delay bin orannulus tends to obscure the uneven distribution, creatingan ambiguity with respect to the asymmetry direction.Recovery is possible with multiple satellites under thecondition that the glistening zones are due to the samesurface wind conditions. Because the annuli for the two ormore satellites are not mutually concentric, thesemeasurements provide the necessary conditions forobserving the PDF asymmetry.In the latest version of our algorithm, we implemented theoption to process any number of satellites in a single batchto fit the measured to the modeled waveforms using a non-

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->