TY - JOUR

T1 - Randomized Subspace Newton Convex Method Applied to Data-Driven Sensor Selection Problem

AU - Nonomura, Taku

AU - Ono, Shunsuke

AU - Nakai, Kumi

AU - Saito, Yuji

N1 - Funding Information:
Manuscript received December 3, 2020; accepted January 3, 2021. Date of publication January 13, 2021; date of current version February 9, 2021. This work was supported by JST CREST under Grant JPMJCR1763 and by ACT-X under Grant JPMJAX20AD. The associate editor coordinating the review of this manuscript and approving it for publication was Dr. Arash Mohammadi. (Corresponding author: Taku Nonomura.) Taku Nonomura, Kumi Nakai, and Yuji Saito are with the Department of Aerospace Engineering, Tohoku University, Sendai 980-8579, Japan (e-mail: nonomura@aero.mech.tohoku.ac.jp; nakai@aero.mech.tohoku.ac.jp; saito@aero.mech.tohoku.ac.jp).

PY - 2021

Y1 - 2021

N2 - The randomized subspace Newton convex methods for the sensor selection problem are proposed. The randomized subspace Newton algorithm is straightforwardly applied to the convex formulation, and the customized method in which the part of the update variables are selected to be the present best sensor candidates is also considered. In the converged solution, almost the same results are obtained by original and randomized-subspace-Newton convex methods. As expected, the randomized-subspace-Newton methods require more computational steps while they reduce the total amount of the computational time because the computational time for one step is significantly reduced by the cubic of the ratio of numbers of randomly updating variables to all the variables. The customized method shows superior performance to the straightforward implementation in terms of the quality of sensors and the computational time.

AB - The randomized subspace Newton convex methods for the sensor selection problem are proposed. The randomized subspace Newton algorithm is straightforwardly applied to the convex formulation, and the customized method in which the part of the update variables are selected to be the present best sensor candidates is also considered. In the converged solution, almost the same results are obtained by original and randomized-subspace-Newton convex methods. As expected, the randomized-subspace-Newton methods require more computational steps while they reduce the total amount of the computational time because the computational time for one step is significantly reduced by the cubic of the ratio of numbers of randomly updating variables to all the variables. The customized method shows superior performance to the straightforward implementation in terms of the quality of sensors and the computational time.

KW - Computational efficiency

KW - Convergence

KW - Convex sensor selection problem

KW - Data-driven sensor selection

KW - Heuristic algorithms

KW - Linear programming

KW - Newton method

KW - Randomized subspace Newton algorithm

KW - Signal processing algorithms

KW - Sparse matrices

UR - http://www.scopus.com/inward/record.url?scp=85099529581&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85099529581&partnerID=8YFLogxK

U2 - 10.1109/LSP.2021.3050708

DO - 10.1109/LSP.2021.3050708

M3 - Article

AN - SCOPUS:85099529581

JO - IEEE Signal Processing Letters

JF - IEEE Signal Processing Letters

SN - 1070-9908

ER -