Signal Representation for Symbolic and Numerical Processing.
Abstract
Traditional signal processing by computer relies on numerical methods. This thesis explores and develops the symbolic representation and manipulation of signals and demonstrates new models for the numerical representation and manipulation of signals. It is shown that combined numerical and symbolic representation provides advantages in the computer processing of signals. We describe how symbolic representation and manipulation methods may be incorporated in the numerical processing of signals and how these methods may be used to provide facilities for the design and analysis of signal processing systems. For the numerical representation of signals a model, called the deferred array, that incorporates the functional form of a signal along with a memory and deferred evaluation, is developed. It is shown that the deferred array closely matches the way in which signals are described mathematically and can be used to represent infinite duration signals and many signal processing operations. SPLICE, a software environment for the exploratory development of signal processing algorithms, incorporating our ideas on the numerical representation of signals, is described.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1986
- Accession Number
- ADA178240
Entities
People
- Cory S. Myers
Organizations
- Massachusetts Institute of Technology