Suthaharan , Shan

UNCG

There are 7 item/s.

TitleDateViewsBrief Description
An analysis of a sparse linearization attack on the advanced encryption standard 2006 3204 "Since Rijndael was accepted as the new Advanced Encryption Standard by the NIST, several techniques have been developed to attack it. One of the more controversial techniques is a relatively new mathematically based attack known as Extended Sparse L...
Stabilizing RED queue oscillation using the logistic map in AutoRED mechanism 2009 1812 Active queue management (AQM) is one of the ways to control congestion at Internet Routers. One of the widely used AQM's is the random early detection (RED) scheme. The RED scheme suffers from chaotic queue oscillation problem particularly in a highl...
eChirp: Measuring Available Bandwidth for the Internet Using Multiple Chirp Packet Trains 2008 2441 Measuring available bandwidth over a network path in the Internet is a challenging research problem. In this thesis we have studied this problem and developed a new technique called "eChirp". First, the effectiveness of pathChirp [1] is studied in te...
Applying hybrid cloud systems to solve challenges posed by the big data problem 2013 1269 The problem of Big Data poses challenges to traditional compute systems used for Machine Learning (ML) techniques that extract, analyze and visualize important information. New and creative solutions for processing data must be explored in order to o...
Wi-Fi 802.11 based mobile robotics positioning system 2007 1870 "This thesis provides a method for finding a location of a mobile robot based on the signal strengths obtained from the IEEE 802.11 standard wireless Access Points. In this method a set of eight signal loss functions is proposed to enable the robot t...
Evaluation of the performance of deep learning techniques over tampered dataset 2015 4781 The reduction of classification error over supervised data sets is the main goal in Deep Learning (DL) approaches. However, tampered data is a serious problem in machine learning techniques. One of the recent interests to the machine learning communi...
Feature extraction and feature reduction for spoken letter recognition 2016 3559 The complexity of finding the relevant features for the classification of spoken letters is due to the phonetic similarities between letters and their high dimensionality. Spoken letter classification in machine learning literature has often led to v...