Multiresolution Signal and System Analysis and the Analysis and Control of Discrete Event Dynamic Systems.
Abstract
In this report we summarize the following research accomplishments: (a) combining our research on inverse problems with our work on the development of multiresolution stochastic models in order to develop novel and very efficient methods for the fusion and inversion of heterogeneous and multiresolution sensor data; (b) developing a significant extension of the use of our scale recursive multiresolution models for the modeling of spatial phenomena based both on a novel application of the concept of canonical correlations in statistics and on relaxing the relationship between variables in our multiresolution representations and the spatial variables they represent; (c) developing multiresolution models for SAR imagery and the use of these models as thesis for new and very effective likelihood feature for the discrimination of man made objects and natural clutter; and (d) developing a new method for signal approximation and feature extraction known as high resolution pursuit that produces stable and physically meaningful features and the application of this method to high range resolution radar data.
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 1996
- Accession Number
- ADA305529
Entities
People
- Alan S. Willsky
Organizations
- Massachusetts Institute of Technology