Abstract—Conventional error based cost function provides unsatisfactory weight update of an adaptive system when outliers are present in the training signal. To alleviate this problem in this paper a hybrid approach using differential evolution (DE) and Wilcoxon norm is proposed to provide robust training in identification of complex nonlinear systems. Exhaustive simulation study shows superior performance of the new method compared to the conventional square error based minimization method.
Index Terms—Differential evolution, robust system identification, wilcoxon norm.
B. Majhi is with the Dept. of CS & IT, G. G. Vishwavidyalaya, Central University, Bilaspur, India (e-mail: firstname.lastname@example.org).
H. P. Thethi is with School of Electronics, KIITS University, Bhubaneswar, India (e-mail: email@example.com).
J. Satpathy is with Dept. of Electronics Engg, at Govt. Polytechnic, Bhubaneswar, India (e-mail: firstname.lastname@example.org).
Cite: Babita Majhi, H. Pal Thethi, and Jeetamitra Satpathy, "Identification of Nonlinear Systems in Presence of Outliers Using Robust Norm and Differential Evolution," International Journal of Modeling and Optimization vol. 2, no. 4, pp. 508-512, 2012.