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 the A-optimal design of experiment using the alternating direction method of multipliers (ADMM) algorithm. The performance of the proposed method was evaluated with a random sensor problem and compared with previously proposed methods, such as the greedy and convex relaxation methods. The performance of the proposed method is better than the existing greedy and convex relaxation methods in terms of the A-optimality criterion. Although, the proposed method requires a longer computational time than the greedy method, it is quite shorter than that of convex relaxation method in large-scale problems. Then the proposed method was applied to the data-driven sparse-sensor-selection problem. The dataset adopted was the National Oceanic and Atmospheric Administration optimum interpolation sea surface temperature dataset. At a number of sensors larger than that of the latent variables, the proposed method showed similar and better performance compared with previously proposed methods in terms of the A-optimality criterion and reconstruction error.

Automated summary

This study proposes a sensor selection method based on the proximal splitting algorithm and the ADMM algorithm, showing better performance than existing greedy and convex relaxation methods in terms of the A-optimality criterion. The proposed method requires longer computational time than the greedy method but is shorter than the convex relaxation method in large-scale problems.