Abstract—This letter introduces a new gradient-based adaptive filtering algorithm based on a cost function that is constructed by combining two robust cost functions, which are a new tanh-type cost function and Vega’s cost function. Through the approach to combine robust cost functions, the robustness of the proposed algorithm outperforms that of other adaptive algorithms. Since the proposed algorithm is derived by combining two robust cost functions, it leads to an excellent transient and steady-state behavior in high probability of impulsive measurement noise. The proposed algorithm is tested in different probability of impulsive measurement noise.
Index Terms—Adaptive filter, robust filtering, impulsive noise, normalized least-mean-square (NLMS) algorithm.
The authors are with the Division of Department of Electrical and Computer Engineering and IT Convergence Engineering, Pohang University of Science and Technology, Pohang, Gyungbuk, Republic of Korea (e-mail: weedsis@postech.ac.kr, wilee@postech.ac.kr, kwunnam@postech.ac.kr, ppg@postech.ac.kr)
[PDF]
Cite: Insun Song, Won Il Lee, Nam Kyu Kwon, and Poo Gyeon Park, "A Robust Variable Step-Size NLMS Algorithm through a Combination of Robust Cost Functions," International Journal of Information and Electronics Engineering vol. 2, no. 6, pp. 936-939, 2012.
