On restricted diagonalization

Abstract: Let $\mathcal{H}$ be a separable infinite-dimensional complex Hilbert space, $\mathcal{B}(\mathcal{H})$ the algebra of bounded linear operators acting on $\mathcal{H}$ and $\mathcal{J}$ a proper two-sided ideal of $\mathcal{B}(\mathcal{H})$. Denote by $\mathcal{U}\mathcal{J}(\mathcal{H})$ the group of all unitary operators of the form $I+\mathcal{J}$. Recall that an operator $A \in \mathcal{B}(\mathcal{H})$ is diagonalizable if there exists a unitary operator $U$ such that $UAU^*$ is diagonal with respect to some orthonormal basis. A more restrictive notion of diagonalization can be formulated with respect to a fixed orthonormal basis $\mathrm{e}={ e_n}{n\geq 1}$ and a proper operator ideal $\mathcal{J}$ as follows: $A \in \mathcal{B}(\mathcal{H})$ is called restricted diagonalizable if there exists $U\in \mathcal{U}_\mathcal{J}(\mathcal{H})$ such that $UAU^*$ is diagonal with respect to $\mathrm{e}$. In this work we give necessary and sufficient conditions for a diagonalizable operator to be restricted diagonalizable. Our conditions become a characterization of those diagonalizable operators which are restricted diagonalizable when the ideal is arithmetic mean closed. Then we obtain results on the structure of the set of all restricted diagonalizable operators. In this way we answer several open problems recently raised by Belti\c{t}$\breve{\text{a}}$, Patnaik and Weiss.