James A Garrison

US Patent:

7358895, Apr 15, 2008

Filed:

Aug 25, 2005

Appl. No.:

11/212125

Inventors:

Nesreen I. Ziedan - Oxford OH, US
James L. Garrison - Lafayette IN, US

Assignee:

Purdue Research Foundation - West Lafayette IN

International Classification:

G01S 1/00
H04B 7/185

US Classification:

34235712

Abstract:

A method of acquiring a weak signal includes selecting a first predetection integration time (PIT) interval for processing a digital intermediate frequency (IF) signal, selecting a number of data bit edges within the PIT, computing a plurality of coherent integrations for each data bit edge, selecting a coherent integration, which corresponds to a most likely data combination for each data bit edge, from the plurality of coherent integrations, updating an incoherent integration total for each data bit edge with the coherent integration corresponding to the selected most likely data bit combination for each data bit edge, comparing the incoherent integration total for each data bit edge with a threshold after a predetermined number of steps of the coherent integration computations and incoherent integration total updates, and identifying a code delay from the incoherent integration. The method may also include determining a code delay from an estimated Doppler shift corresponding to the identified code delay. In this method, the coherent integrations are computed using a circular correlation technique that is performed in a manner so the coherent integrations are not independent from one another in successive steps.

US Patent:

7545894, Jun 9, 2009

Filed:

Mar 21, 2005

Appl. No.:

11/087065

Inventors:

Nesreen I. Ziedan - Oxford OH, US
James L. Garrison - Lafayette IN, US

Assignee:

Purdue Research Foundation - West Lafayette IN

International Classification:

H04B 7/10

US Classification:

375347, 34042619, 34053913, 34235705, 34235706, 34235712, 701213

Abstract:

A method acquires a location system signal that includes second data signals superimposed on the first data signals. The method includes iterating Doppler frequency offset values and time offsets to generate error signals, each representing a difference between summed consecutive correlation values and a corresponding estimated ideal value. Each error signal corresponds to a different time offset and an associated time period. The second data signal has a bit edge that occurs during the time period of each of the sums of consecutive correlation values. The presence of a bit edge during an associated time period increases the error signal. A measure of the minimum error is used to identify a time offset value that is used to develop a first determined Doppler frequency offset. A Kalman filter refines the first determined Doppler frequency offset.

US Patent:

7679550, Mar 16, 2010

Filed:

Feb 21, 2007

Appl. No.:

11/709309

Inventors:

James L. Garrison - Lafayette IN, US
Brenda E. Eichel - Pittsburgh PA, US

International Classification:

G01S 1/02

US Classification:

34235701, 342352

Abstract:

A method for propagating ephemeris data for a satellite in Earth orbit is provided. The method includes the steps of receiving orbital positional data for a first time period of a satellite's Earth orbit, propagating orbital positional data for the satellite's Earth orbit during a second time period extending beyond the first time period, fitting a Keplerian ephemeris model to the propagated orbital positional data to estimate model coefficients, and sending the estimated model coefficients to receivers for determination of receiver position at a time during the second time period.

US Patent:

7920093, Apr 5, 2011

Filed:

Sep 27, 2007

Appl. No.:

11/904600

Inventors:

Gregory W. Heckler - Greenbelt MD, US
James L. Garrison - Lafayette IN, US

Assignee:

Purdue Research Foundation - West Lafayette IN

International Classification:

G01S 19/37

US Classification:

34235777

Abstract:

A method reduces consumption of computational resources in a satellite signal receiver. The method includes segmenting a sample of a received global positioning base band vector, segmenting a sample of a replica code/acquisition signal into segments corresponding in length to the segments of the base band vector, circularly correlating segments of the base band vector to zero padded segments of the code/acquisition signal to form a time domain correlation vector, inserting a portion of the time domain correlation vector in a matrix, continuing to correlate circularly the segments of the received signal sample with zero padded segments of the code signal sample until all of the segments in received signal sample have been circularly correlated, dividing rows in the matrix into blocks that are equal to a smallest increment corresponding to a bit edge, sorting the blocks from the divided rows into zero padded blocks, transforming the zero padded blocks into frequency domain columns, summing the frequency domain columns in different combinations to form a plurality of coherent integration matrices for testing possible bit edge locations and bit values.

US Patent:

20060115022, Jun 1, 2006

Filed:

Nov 30, 2005

Appl. No.:

11/290982

Inventors:

Nesreen Ziedan - Oxford OH, US
James Garrison - Lafayette IN, US

International Classification:

H04L 27/06
G01S 5/14

US Classification:

375340000, 342357150

Abstract:

A method of acquiring a signal in a high dynamic environment includes generating a first set of acquired signal power values, the first set comprising an acquired signal power value for each Doppler bin in a plurality of Doppler bins over a first predetection integration time (PIT) interval for processing a digital signal, and generating a second set of acquired signal power values, the second set comprising an acquired signal power value for each Doppler bin in the plurality of Doppler bins over a second subsequent predetection integration time (PIT) interval for processing the digital signal. The first and the second sets are used to generate at least three additional sets of acquired signal power values. A new set of acquired power values is generated by selecting the maximum of the summations for each Doppler bin from the three additional sets, and a Doppler shift is identified from the Doppler bin having the maximum summation value that is greater than a threshold value. The output of this method may be used to initialize a first and a second extended Kalman filters for tracking a carrier signal and a code signal, respectively.

US Patent:

20110071808, Mar 24, 2011

Filed:

Sep 23, 2010

Appl. No.:

12/889351

Inventors:

James Levi Garrison - Lafayette IN, US
Michael Allen Walker - Newburgh IN, US

Assignee:

PURDUE RESEARCH FOUNDATION - West Lafayette IN

International Classification:

G06F 17/14

US Classification:

703 2

Abstract:

A method for providing an extended propagation ephemeris model for a satellite in Earth orbit, the method includes obtaining a satellite's orbital position over a first period of time, applying a least square estimation filter to determine coefficients defining osculating Keplarian orbital elements and harmonic perturbation parameters associated with a coordinate system defining an extended propagation ephemeris model that can be used to estimate the satellite's position during the first period, wherein the osculating Keplarian orbital elements include semi-major axis of the satellite (a), eccentricity of the satellite (e), inclination of the satellite (i), right ascension of ascending node of the satellite (Ω), true anomaly (θ*), and argument of periapsis (ω), applying the least square estimation filter to determine a dominant frequency of the true anomaly, and applying a Fourier transform to determine dominant frequencies of the harmonic perturbation parameters.