Blind Identification of Fully Observed Linear Time-Varying Systems via Sparse Recovery

TL;DR

This paper introduces a sparse recovery-based method for identifying fully observed linear time-varying systems without known inputs, with applications in complex dynamical systems analysis such as breast cancer research.

Contribution

It formulates the system identification as a compressive sensing problem accounting for noise and sparsity, providing conditions for unique recovery of system parameters and inputs.

Findings

Method effectively recovers system models in noisy conditions.

Provides practical conditions for experiment design and system identifiability.

Demonstrates robustness through synthetic experiments.

Abstract

Discrete-time linear time-varying (LTV) systems form a powerful class of models to approximate complex dynamical systems with nonlinear dynamics for the purpose of analysis, design and control. Motivated by inference of spatio-temporal dynamics in breast cancer research, we propose a method to efficiently solve an identification problem for a specific class of discrete-time LTV systems, in which the states are fully observed and there is no access to system inputs. In addition, it is assumed that we do not know on which states the inputs act, which can change between time steps, and that the total number of inputs is sparse over all states and over time. The problem is formulated as a compressive sensing problem, which incorporates the effect of measurement noise and which has a solution with a partially sparse support. We derive sufficient conditions for the unique recovery of the system model and input values, which lead to practical conditions on the number of experiments and rank conditions on system outputs. Synthetic experiments analyze the method's sensitivity to noise for randomly generated models.