Spectral models for orthonormal wavelets and multiresolution analysis of \(L^{2}(\mathbb R)\) (Q532503)

The authors introduce an abstract operator-theoretic approach to the construction of orthonormal wavelet bases. To this end, they consider two different ways of representing unitary operators, i.e. via the so-called \textit{spectral model} (provided by the spectral theorem) and the so-called \textit{shift model}. Both representations are derived for the (dyadic) dilation and shift operators underlying the orthonormal wavelet bases. Using these representations, the authors derive necessary and sufficient conditions for scaling functions and associated wavelets. The new approach provides rather neat and compact criteria, which are however closely related to preexisting, similar approaches due to, amongst others, Hernández, Wang, Weiss, Ron, and Shen. This is duely pointed out in the final section of the paper.