Analytical controls for dispersive multi-qubit interactions in mediator-coupled quantum registers

Abstract:Fast control of long-lived quantum registers often relies on an auxiliary degree of freedom, a mediator, but real mediator excitation can introduce decoherence, leakage, or loss when the mediator is not the coherence-optimized component. We introduce dispersive interaction via analytical linear inversion (DIAL), an analytical control construction for excitation-suppressed interaction engineering in mediator-coupled quantum registers. In the dispersive regime, we show that off-resonant driving of an effective two-level mediator transition yields a diagonal Hamiltonian whose Pauli-\(Z\) interaction rates depend linearly on the drive intensities through a transfer matrix determined by the register-resolved transition spectrum. This establishes a direct linear map from experimentally accessible controls to induced multi-qubit interactions, enabling control design without iterative propagation of the full driven dynamics. We turn this structure into the target-aware DIAL algorithm, which selects favorable off-resonant tones and determines their drive intensities for a prescribed target interaction. We benchmark the resulting controls across 100 register realizations for each register size \(n=2,3,4,5\), targeting the highest-order interaction \(Z_1\cdots Z_n\) with an interaction phase of magnitude \(\pi/4\), which generates an \(n\)-qubit GHZ state up to local equivalence. Propagation under the full driven mediator-register dynamics yields typical gate fidelities above \(0.997\), while the maximum transient mediator excitation remains near \(1\%\) on average, far below what is typically possible with resonant Rabi drives. The manuscript is accompanied by an open-source Python package implementing the target-aware DIAL algorithm and the associated numerical validation workflow.

arXiv:2610.03064 [quant-ph]

(or arXiv:2610.03064v1 [quant-ph] for this version)

https://doi.org/10.48550/arXiv.2610.03064 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.