Generalizing the Kantorovich metric to projection valued measures (Q904273)

Here \((X, d)\) is a compact metric space, \(\mathcal{B}(X)\) denotes all Borel subsets of \(X\), \(M(X)\) all probability measures on \(X\) and \(C_{R}(X)\) all real-valued continuous functions on \(X\). It is well-known that, with Kantorovich metric, \(M(X)\) is a compact metric space and the topology is the same as the topology of pointwise convergence on \(C_{R}(X)\). In this paper the author gets some extensions of this result to more general setting. Now \(\mathcal{H}\) is a Hilbert space and \(\mathcal{B}(\mathcal{H})\) all bounded linear operators on \(\mathcal{H}\) with operator norm. Then the Borel measures on \(X\) take values in \(\mathcal{B}(\mathcal{H})\) which are projections on the Hilbert space \(\mathcal{H}\). He defines a new metric \(\rho\) on this new set of measures and calls it a generalized Kantorovich metric and proves that the topology of this metric is the same as the topology of pointwise convergence on \(C_{R}(X)\). It is also proved that this metric is complete but not necessarily compact. He uses the contraction mapping theorem and this complete metric space to give a new proof of a result in the paper: \textit{P. E. T. Jorgensen} [``Use of operator algebras in the analysis of measures from wavelets and iterated function systems'', Contemp. Math. 414, 13--26 (2006; Zbl 1125.28010)]. Some additional comments about some generalizations of the results he has obtained are made.