Joint Signal Recovery and Graph Learning from Incomplete Time-Series
This addresses a domain-specific issue in graph signal processing for scenarios with incomplete data, representing an incremental improvement.
The paper tackles the problem of learning a graph from incomplete time-series data, proposing a BSUM-based algorithm for joint signal recovery and graph inference, with simulation results demonstrating its performance.
Learning a graph from data is the key to taking advantage of graph signal processing tools. Most of the conventional algorithms for graph learning require complete data statistics, which might not be available in some scenarios. In this work, we aim to learn a graph from incomplete time-series observations. From another viewpoint, we consider the problem of semi-blind recovery of time-varying graph signals where the underlying graph model is unknown. We propose an algorithm based on the method of block successive upperbound minimization (BSUM), for simultaneous inference of the signal and the graph from incomplete data. Simulation results on synthetic and real time-series demonstrate the performance of the proposed method for graph learning and signal recovery.