A novel online adaptive time delay identification technique

Abstract

Time delay is a phenomenon which is common in signal processing, communication, control applications, etc. The special feature of time delay that makes it attractive is that it is a commonly faced problem in many systems. A literature search on time-delay identification highlights the fact that most studies focused on numerical solutions. In this study, a novel online adaptive time-delay identification technique is proposed. This technique is based on an adaptive update law through a minimum–maximum strategy which is firstly applied to time-delay identification. In the design of the adaptive identification law, Lyapunov-based stability analysis techniques are utilised. Several numerical simulations were conducted with Matlab/Simulink to evaluate the performance of the proposed technique. It is numerically demonstrated that the proposed technique works efficiently in identifying both constant and disturbed time delays, and is also robust to measurement noise.

Keywords:

• adaptive identification
• time delay
• signal processing

Notes

1. Preliminary results of this work were published in Bayrak and Tatlicioglu (2011).

2. A simplex in ${\mathbb{R}}^n$ is a convex polytope with n + 1 vertices.

3. Although the derivations are very similar to that of Annaswamy et al. (1998), we presented them for the sake of completeness.

4. In examples, phase shifted form was used to illustrate the delayed sinusoidal signals. The phase shift in radians can be, easily, converted to time delay by a basic linear transformation.

Additional information

Notes on contributors

Alper Bayrak

Alper Bayrak was born in Ankara, Turkey, on 19 October 1980. He graduated from the Electrical and Electronics Engineering Department in Blacksea Technical University, Trabzon, Turkey, in 2004. In 2005, he attended the MSc programme at the Electrical and Electronics Engineering Department in Gazi University and graduated in 2007. In 2013, he acquired his Ph.D. degree in electronics and communications engineering from Izmir Institute of Technology, Izmir, Turkey. He worked as a research assistant at the Department of Electrical and Electronics Engineering at Abant Izzet Baysal University, Bolu, Turkey, from 2005 to 2007, and at the Department of Electrical and Electronics Engineering in Izmir Institute of Technology from 2008 to 2013. He has been working as an assistant professor at the Department of Electrical and Electronics Engineering in Abant Izzet Baysal University since January, 2014. His current fields of research are control, identification, nonlinear systems, and robotics.

Enver Tatlicioglu

Enver Tatlicioglu received his BSc degree in electrical and electronics engineering from Dokuz Eylul University, Izmir, Turkey, and his Ph.D. degree in electrical and computer engineering from Clemson University, Clemson, SC, USA, in 1999 and 2007, respectively. Upon completion of his Ph.D. degree, he worked as a post-doctoral research fellow in the Department of Electrical and Computer Engineering at Clemson University and then joined the Department of Electrical & Electronics Engineering at Izmir Institute of Technology, Izmir, Turkey, where he is currently working as an associate professor. His research interests include control and identification of time-delay systems, dynamic modelling of extensible continuum robot manipulators, nonlinear control techniques for kinematically redundant robot manipulators, partial state feedback and output feedback control, haptic systems and teleoperation; learning, robust, and adaptive control of nonlinear systems.