Nonlinear multifunctional sensor signal reconstruction based on local least squares support vector machines

Abstract

Least squares support vector machines (LSSVM), as a recently reported least squares version support vector machines (SVM), involves equality constraints instead of inequality constraints and adopts least squares cost function, therefore it expresses the training by solving a set of linear equations instead of the quadratic programming problem which greatly reduces computational cost. In this paper, we combine LSSVM with a local approach in order to obtain accurate estimations of multifunctional sensor signals. For the simulation model of multifunctional sensor, the reconstruction accuracies of input signals are 1.07% and 1.27%, respectively. The experimental results demonstrate the higher reliability and accuracy of proposed method for multifunctional sensor signal reconstruction than original LSSVM algorithm, and verify the feasibility of proposed method.