Affiliations: [a] Department of Human and Information Systems
Engineering, Gifu University, Gifu, Japan | [b] Department of Mechanical and Systems Engineering, Gifu
University, Gifu, Japan
Correspondence:
[*]
Corresponding author. Department of Human and Information Systems
Engineering, Gifu University, Gifu, Japan 1-1 Yanagido, Gifu, Japan. Tel.: +81 58 293 2541; Fax: +81 58 230
1892; E-mail: [email protected]
Abstract: In this paper an active random noise control using adaptive learning
rate neural networks with an immune feedback law is presented. The adaptive
learning rate strategy increases the learning rate by a small constant if the
current partial derivative of the objective function with respect to the weight
and the exponential average of the previous derivatives have the same sign,
otherwise the learning rate is decreased in proportion to its value. The use of
an adaptive learning rate attempts to keep the learning step size as large as
possible without inducing oscillation. In the proposed method, because an
immune feedback law is changing the learning rate of the neural networks
individually and adaptively, it is expected that the neural cost function will
reach its minimum rapidly, resulting in a reduced training time. Numerical
simulations and experiments of active random noise control with the transfer
function of the error path will be performed to validate the convergence
properties of the method. Control results show that the adaptive learning rate
neural network control structure can outperform linear controllers and
conventional neural network controllers for active random noise control.
Keywords: Active noise control, random noise, immune feedback law, adaptive filtering algorithm
