Comments on "Design of Asymmetric Shift Operators for Efficient Decentralized Subspace Projection"
arXiv:2107.11111 · doi:10.1109/TSP.2022.3182161
Abstract
This correspondence disproves the main results in the paper "Design of Asymmetric Shift Operators for Efficient Decentralized Subspace Projection" by S. Mollaebrahim and B. Beferull-Lozano. Counterexamples and counterproofs are provided when applicable. A correction is suggested for some of the flaws. However, it does not seem possible to amend most of the flaws since the overall approach based on a Schur decomposition of the shift matrix does not appear to be helpful to solve the desired problem.