Engineering topics
Publications and source records attributed to Pang, S. S..
We summarize 24 years (1978??2) of ice export estimates and examine, over a 9-year record, the associated variability in the time-varying upward-looking sonar (ULS) thickness distributions of the Fram Strait.
We summarize 24 years of ice export estimates and examine, over a 9-year record, the associated variability in the time-varying upward-looking sonar (ULS) thickness distributions of the Fram Strait. A more thorough assessment of the PMW (passive microwave) ice motion with 5 years of synthetic aperture radar (SAR)observations shows the uncertainties to be consistent with that found by Kwok and Rothrock [1999], giving greater confidence to the record of ice flux calculations.
The spectral analysis algorithm is a viable technique for processing synthetic aperture radar (SAR) data in near real time throughput rates by trading the image resolution. One major challenge of the spectral analysis algorithm is that the output image, often referred to as the range-Doppler image, is represented in the iso-range and iso-Doppler lines, a curved grid format. This phenomenon is known to be the fanshape effect. Therefore, resampling is required to convert the range-Doppler image into a rectangular grid format before the individual images can be overlaid together to form seamless multi-look strip imagery. An efficient algorithm for geometric rectification of the range-Doppler image is presented. The proposed algorithm, realized in two one-dimensional resampling steps, takes into consideration the fanshape phenomenon of the range-Doppler image as well as the high squint angle and updates of the cross-track and along-track Doppler parameters. No ground reference points are required.
Errors in the determination of the platform state vector (position and velocity) limit the accuracy with which individual pixels in a SAR (synthetic aperture radar) image can be located. A method of updating the state vector based on the SAR data is described. The algorithm combines reference targets at known locations with the estimates available from the mission control center, utilizing a maximum likelihood estimator. The targets can be located at calibration sites or be easily recognizable targets of opportunity. The accuracy of the update depends on the number of targets, their geometry, and the accuracy of their position determination. The algorithm is tested using Seasat SAR data. High-accuracy state vectors were perturbed by Gaussian distributed errors with standard deviations and cross-correlations based on worst-case expectations for SIR-C (Shuttle Imaging Radar). Tests using ground targets located in topographic maps (1:24,000) showed that state vector position errors on the order of 300-400 m and velocity accuracies of 2-3 m/s are readily achievable.
Digital image-processing system assembles maplike mosaic images of large areas of Earth from overlapping images of smaller areas generated by spaceborne synthetic-aperture radar (SAR). Automatically registers and blends smaller images with each other. Eliminates need for time-consuming, tedious manual processing. Modifiable to accept images from spaceborne optical sensors.
During the past thirty years, a tremendous amount of research was done on the development of crazing in polymers. The phenomenon of crazing was recognized as an unusual deformation behavior associated with a process of molecular orientation in a solid to resist failure. The craze absorbs a fairly large amount of energy during the crazing process. When a craze does occur the surrounding bulk material is usually stretched to several hundred percent of its original dimension and creates a new phase. The total energy absorbed by a craze during the crazing process in creep was calculated analytically with the help of some experimental measurements. A comparison of the energy absorption by the new phase and that by the original bulk uncrazed medium is made.