HR: 1340h
AN: SF53A-0725    [Abstracts]
TI: Computationally Efficient Inversion of Scattered GPS Signals for the Retrieval of Surface Winds or Roughness from an Airborne Receiver
AU: * Garrison, J L
EM: jgarriso@ecn.purdue.edu
AF: Purdue University, 315 N. Grant Street, West Lafayette, IN 47907-2023
AB: An approximation is presented for the forward model relating the correlation waveform of a reflected GPS signal to the statistics of the reflecting surface slopes. This approximation eliminates the need to numerically integrate over the reflecting surface to generate waveforms for assumed surface conditions. The inverse problem, estimation of the surface slope statistics from measurements of the waveform, is substantially faster through the use of this approximate model and its gradient. Prior work has shown that the bistatic scattering model can be formulated as a convolution between the GPS code autocorrelation function and the distribution of scattered power with delay. A change of variables in the surface integral, from a Cartesian system, to delay and an angular coordinate, the scattered waveform can be computed as a single integral over the angular coordinate, and a convolution in delay. If Doppler spreading is ignored, a good approximation at aircraft speeds, the second integral is shown to scale with altitude, thereby allowing a single calculation to be performed, with the result scaled to the appropriate aircraft altitude. Finally, an exponential series, with numerically-determined coefficients, is found to approximate this integral very well, and can be computed with substantially lower processing time. This reduced model is demonstrated by processing a long series of aircraft data using a recursive least squares method, to generate estimates of the mean square slope of the ocean surface. A number of empirical models can then be applied to produce a wind speed estimate from this mean square slope. The processing time for this demonstration is low enough to justify its application to an onboard real time wind speed sensor, which could have operational uses in weather monitoring, or search and rescue.
DE: 6959 Radio oceanography
DE: 6969 Remote sensing
DE: 0629 Inverse scattering
DE: 0694 Instrumentation and techniques
DE: 0930 Oceanic structures
SC: Special Focus: Advances in Data Acquisition, Management, Analysis and Display [SF]
MN: 2004 AGU Fall Meeting