Summary of Abstract Submission

Abstract Submission No. IO50-04-0032PresentationPoster

IMPLEMENTATION OF ROBUST DE-NOISING USING WILCOXON BASED ADAPTIVE FILTERING TECHNIQUES IN UNDERWATER APPLICATION

Ms P.Devi Sowjanya*1, Dr Kusuma Kumari Cheepurupalli1, Mr. A.Vamsee Krishna2

1 GVP College of Engineering (A), Visakhapatnam, India
2 Vignan institute of technology, Visakhapatanam, India

ABSTRACT :

The real time signals are used for information extraction and decision making in various applications. These real time signals are generally corrupted by various noises in the environment. The main concern in signal processing is to remove the noise and extract the information from the signal. Signal de-noising can be effectively done if the noise properties are known prior. But in real scenario it is difficult to know the properties prior. Hence various adaptive algorithms are being developed in the literature such as LMS NLMS, RLS, Kalman etc., which are used to achieve better accuracy and good output SNR values. Non-parametric techniques such as Wilcoxon test etc., can be considered as robust learning algorithm that provides greater accuracy, when combined with adaptive techniques. In this paper, wilcoxon norm is used in combination with least square and recursive least square algorithms to enhance the performance of these adaptive algorithms. The performance measure includes the RMSE and Output SNR calculation in this paper. For various input SNR conditions and fitter orders, the adaptive techniques are compared and observed that the performances of these algorithms are enhanced.