Splitting Probability in Monitored Quantum Dynamics
π arXiv:2610.29547 Β· π₯ PDF Β· 2026-10-21 Β· cond-mat.stat-mech
Authors: Prashant Singh [arXiv Β· scholar] , Sabine Tornow [arXiv Β· scholar] , David A Kessler [arXiv Β· scholar] , Eli Barkai [arXiv Β· scholar]
π Abstract
We consider a continuous-time quantum walk stroboscopically subjected to projective measurements on two distinct targets with sampling time period $Ο$. We construct a general renewal framework for the first-detection amplitude at each target, from which we obtain the associated splitting probabilities. For Hamiltonians with reflection symmetry, we further show that each dual-target first-detection amplitude can be viewed as an interference effect between a pair of single-target first-detection amplitudes. This formulation yields exact expressions for the splitting probabilities and gives special sampling times at which they undergo pointwise discontinuous jumps due to a form of ergodicity breaking associated with dark states. As an illustrative example, we apply the framework to a one-dimensional tight-binding lattice chain with nearest-neighbor hopping, and study the splitting probabilities as functions of the initial position and sampling time. Below a critical sampling time, $0<Ο< Ο_c$ where $Ο_c = 2 Ο/ ΞE$ with $ΞE$ being the energy bandwidth, the splitting probabilities for large system sizes become equal to $1/2$, independent of the initial position, as long as this position does not coincide with target sites. Above the critical sampling time, $Ο\geq Ο_c$, the splitting probabilities deviate from $1/2$ and develop oscillations as a function of initial position. The wavenumber associated with these oscillations is $Q=2\sin^{-1}(rΟ_c/Ο)$, where $r$ is a positive integer. We also explore the effects of relative phase and spatial separation between different components of a coherent initial state on the splitting probabilities. Finally, we study how a finite number of measurements modifies the splitting probabilities and broadens the resonances at special sampling times.