The Analysis and Classification of Random Aperiodic Signals.
Abstract
Two classes of random aperiodic signals were analyzed using one- and two-dimensional Fourier transforms. Low-frequency filtering in the transform domain and the Euclidean distance metric were used to classify signals in one of the two classes. Linear decision boundaries, clustering algorithms, and a training algorithm using a linear categorizer were also used during the analysis. It was found that low-frequency spatial filtering in the two-dimensional Fourier-transform domain gave complete separation of the two classes of signals analyzed. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1971
- Accession Number
- AD0722647
Entities
People
- Charles F. Hall
Organizations
- Air Force Institute of Technology