We investigate expansions of periodic functions with respect to wavelet bases. Direct and inverse theorems for wavelet approximation in C and L p norms are proved. For the functions possessing local regularity we study the rate of pointwise convergence of wavelet Fourier series. We also define and i
Wavelets Associated with Periodic Basis Functions
β Scribed by Francis J. Narcowich; Joseph D. Ward
- Publisher
- Elsevier Science
- Year
- 1996
- Tongue
- English
- Weight
- 348 KB
- Volume
- 3
- Category
- Article
- ISSN
- 1063-5203
No coin nor oath required. For personal study only.
β¦ Synopsis
In this paper, we investigate a class of nonstationary, orthogonal periodic scaling functions and wavelets generated by continuously differentiable periodic functions with positive Fourier coefficients; such functions are termed periodic basis functions. For this class of wavelets, the decomposition and reconstruction coefficients can be computed in terms of the discrete Fourier transform, so that FFT methods apply for their evaluation. In addition, decomposition at the nth level only involves 2 terms from the higher level. Similar remarks apply for reconstruction. We apply a periodic "uncertainty principle" to obtain an angle/frequency uncertainty "window" for these wavelets, and we show that for many wavelets in this class the angle/frequency localization is good.
π SIMILAR VOLUMES
## Abstract A novel technique for the solution of electromagnetics problems in the time domain is proposed. In this technique, waveletβlike functions are used as the basis set for the field components. The timeβdomain algorithm is derived and examples using the technique are presented. Comparisons
We present some necessary and sufficient conditions for a frame multiresolution analysis (FMRA) to admit a frame wavelet whose dyadic dilations and integer translates generate a frame for L 2 (R) and propose a construction of a wavelet, if it exists, which reduces to the classical orthonormal wavele
We construct a basis in the spaces of Whitney functions E (K) for two model cases, where KβIR is a sequence of closed intervals tending to a point. In the proof we use a convolution property for the coefficients of scaling Chebyshev polynomials. 0.