A Novel Wavelet-Based Approach to Enhance the Performance of Mobile-Business Systems

Lin, Wilfred W.K. and Wong, Allan K.Y. and Dillon, Tharam S. and Chang, Elizabeth. 2010. A Novel Wavelet-Based Approach to Enhance the Performance of Mobile-Business Systems, in Ismail, L. and Chang, E. and Karduck, A.P. (ed), IEEE international conference on digital ecosystems and technologies (DEST 2010), Apr 12 2010, pp. 112-117. Dubai, United Arab Emirates: IEEE.

Source Title

Proceedings of the IEEE international conference on digital ecosystems and technologies (DEST 2010)

Source Conference

IEEE international conference on digital ecosystems and technologies (DEST 2010)

DOI

10.1109/DEST.2010.5610660

ISBN

9781424455515

School

Digital Ecosystems and Business Intelligence Institute (DEBII)

Remarks

Copyright © 2010 IEEE This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

URI

http://hdl.handle.net/20.500.11937/3166

Collection

Curtin Research Publications

Abstract

We propose a novel wavelet-based approach (WBA) that accurately decides if the Internet traffic is currently SRD (short-range dependence) or LRD (long-range dependence) on the fly. As a result, it helps a real-time dynamic cache size tuner such as the MACSC (Model for Adaptive Cache Size Control) maintain the given cache hit ratio under all Internet traffic conditions. The WBA mechanism utilizes the following unique characteristics of wavelet coefficients (WC) for stationary signals f(x) : i) WC values by discrete wavelet transform (DWT), irrespective to if f(x) is SRD or LRD, are always SRD; ii) if f(x) is LRD then the WC variances scale divergently with the dilation factor; and iii) WC variances of SRD signals, however, converge to a constant. The DWT process is Dj,k = (f(x) , (J)j,k) or += Tji2 f f(x)w(Tj X - k)dx formally, where Dj,k arethe WC values, (J)j.k the mother wavelet, j the dilation factor/scale, and k the translation factor.