Steady-State Statistical Modeling of Digitally Stabilized Laser Frequency with Markov-State Feedback

TL;DR

This paper introduces a Markov-state model for analyzing digital laser frequency stabilization, capturing stochastic effects and enabling direct stability assessment without extensive simulations.

Contribution

It develops a discrete-time Markov framework that models quantized digital control dynamics in laser stabilization, accounting for noise and sampling effects.

Findings

Exact Markov model for white noise under certain sampling schemes.

Sampling-dependent deviations in actuator behavior with colored noise.

Immediate stability metrics derived from transition matrix eigenvalues.

Abstract

Laser frequency stabilization is conventionally analyzed using continuous-time control theory, which accurately models analog feedback but is insufficient for digital implementations where quantization, sampling, and stochastic noise shape the dynamics. In modern digital laser systems, such as Photonic Integrated Circuit (PIC)-based lasers, finite discriminator and actuator resolution, sampling delays, and measurement noise introduce stochastic behavior that deterministic models do not capture. We present a discrete-time Markov-state framework that models the evolution of the quantized actuator in a digital laser frequency lock, with state-transition probabilities determined by the frequency discriminator response, noise statistics, and implemented digital control logic. The steady-state actuator and locked-laser frequency distributions are obtained directly from the unit-eigenvalue solution of the transition matrix, providing immediate access to key stability metrics without long time-domain simulations. For white frequency noise, we show that the Markov formulation is exact under decorrelated sampling and update schemes, while correlated discriminator sampling introduces a predictable inflation of actuator variance without shifting the operating point. In the presence of colored noise, long-range temporal correlations induce sampling-dependent deviations in both actuator mean and variance, defining the regime of validity of the memoryless Markov description. This framework provides a compact and physically transparent tool for analyzing and optimizing digitally stabilized lasers in integrated photonic systems.