Martingale models for quantum state reduction

Abstract

Stochastic models for quantum state reduction give rise to statistical laws
\nthat are in most respects in agreement with those of quantum measurement theory.
\nHere we examine the correspondence of the two theories in detail, making a systematic
\nuse of the methods of martingale theory. An analysis is carried out to determine
\nthe magnitude of the fluctuations experienced by the expectation of the observable
\nduring the course of the reduction process and an upper bound is established for
\nthe ensemble average of the greatest fluctuations incurred. We consider the general
\nprojection postulate of L¨uders applicable in the case of a possibly degenerate eigenvalue spectrum, and derive this result rigorously from the underlying stochastic dynamics for state reduction in the case of both a pure and a mixed initial state. We also analyse the associated Lindblad equation for the evolution of the density matrix, and obtain an exact time-dependent solution for the state reduction that explicitly exhibits the
\ntransition from a general initial density matrix to the L¨uders density matrix. Finally,
\nwe apply Girsanov’s theorem to derive a set of simple formulae for the dynamics of
\nthe state in terms of a family of geometric Brownian motions, thereby constructing an
\nexplicit unravelling of the Lindblad equation.