Wavelets in subspaces (Q1916872)

Let \(T\) and \(D\) be the translation and unitary dilation operators on \(L^2(\mathbb{R})\) given by \((Tf)(t)= f(t- 1)\) and \((Df)(t)=\sqrt 2f(2t)\). An orthogonal wavelet for a subspace \(X\) of \(L^2(\mathbb{R})\) is a unit vector \(\psi\in X\) such that \(\{D^nT^m\psi: n,m\in\mathbb{Z}\}\) is an orthonormal basis of \(X\). The authors use the unitaries \(T\) and \(D\) and their commutation relation, \(DT^2= TD\), as a starting point to study the connection between wavelet theory and operator theory, continuing the program of \textit{X. Dai} and \textit{D. R. Larson} [``Wandering vectors for unitary systems and orthogonal wavelets'', Mem. Am. Math. Soc. (to appear)]. One result is that for a given multiresolution analysis the set of scaling functions is path connected in the norm topology. Another is a parameterization of the set of all multiresolution analyses by a set of unitary operators which satisfy certain (local) commutation relations. There is also an interesting analysis of which subspaces of \(L^2(\mathbb{R})\) can possess orthogonal wavelets.