Mathematical Methods and Algorithms for Real-Time Applications
Abstract
The notion of super splines and vertex splines is introduced and studied. Quasi-interpolation formulas for real-time applications are constructed. The method of noncommutative blending of quasi-interpolation and vertex spline interpolation is introduced to yield interpolation schemes which are local, flexible, and of optimal approximation orders. These formulas can be applied to real-time interpolation by means of table-look-up or FIR implementation. Applications to engineering problems such as parallel implementation of the extended Kalman filter and Hankel-norm frequency domain methods are studied. Wavelets are constructed by applying cardinal splines, and hence, they are readily available for real-time interpolation and orthogonal wavelet decompositions and reconstructions.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 25, 1990
- Accession Number
- ADA228889
Entities
People
- Charles K. Chui
Organizations
- Texas A&M University