Data-Driven Sparse Sensor Selection Based on A-Optimal Design of Experiment With ADMM

The present study proposes a sensor selection method based on the proximal splitting algorithm and A-optimal design of experiment using alternating direction multipliers (ADMM) algorithm. performance proposed was evaluated with random problem compared previously methods such as greedy convex relaxation. is better than an existing in terms A-optimality criterion. In addition, requires longer computational time but it quite shorter relaxation large-scale problems. applied to data-driven sparse-sensor-selection problem. A data set adopted NOAA OISST V2 mean sea surface temperature set. At number sensors larger that latent state variables, showed similar performances criterion reconstruction error.