Volume 4, Issue 12, December 2013

Experimental Evaluation of Mobile IPv6 for Linux
Pages: 1-6 (6) | [Full Text] PDF (422K)
Jesus Calle-Cancho, David Cortes-Polo, Javier Carmona-Murillo and Jose-Luis Gonzalez-Sanchez
Department of Computing and Telematics System Engineering, University of Extremadura, Spain
Research, Technological Innovation and Supercomputing Center of Extremadura (CenitS), Trujillo, Spain

Abstract -
With the continuous development of mobile communications and Internet technology, one of the major challenges is to achieve efficient mobility management in wireless networks. Internet protocols do not support mobility and wireless networking does not provide reliable connections to mobile users for real-time multimedia communications. For this reason, the Internet Engineering Task Force (IETF) has developed various protocols for IP mobility management such as Mobile IPv6. In this work we analyze the handover process through handover latency. For this, we have developed a real testbed using Universal Mobile IP for Linux implementation, Cisco routers and Cisco access points. We have also conducted throughput tests with real-time communications such as Real-Time Transport Protocol (RTP), analyzing the behavior of the communications when we introduce delays in the access network.
 Index Terms - Mobile IPv6, Handover, Latency, Multimedia and Linux
Fair Cloud Management System
Pages: 7-10 (4) | [Full Text] PDF (232K)
Cao Thanh Phuong, Dang Thanh Phuc, Mai Xuan Phu and Cao Dang Tan
University of Science Ho Chi Minh City, Viet Nam

Abstract -
Cloud computing offers multiple advantages comparing to "Traditional" computing infrastructures. However, all customer data are stored on a virtual machine (VM) or on a cloud of a service provider (SP). Therefore, service quality and data safety are highly dependent on the SP. It also means that these issues are dependent on SP's privileges on customer's VM. Currently, the rights are not balanced between the SP and the customer on a cloud management system. As a result, the imbalance has granted the SP a number of rights that can cause damage to customer's VM. In such cases, if SP's rights are misused or abused, his customers are highly likely to be adversely affected. To solve this problem, this paper proposed the idea of limiting SP's rights and allowing the customer to prevent the SP from performing dangerous operations on the VM. Our implementation has successfully demonstrated the solution to increase safety for customers while using SP's service. Therefore, the SP is able to acquire competitive advantages in his business.
 Index Terms - Cloud Computing, Virtualization, Balance of Rights and Cloud Management System
OTPK PAKE Protocol with TRNG Based Key Generation
Pages: 11-16 (6) | [Full Text] PDF (332K)
Rajneesh pachouri, Ramratan Ahirwal and Dr. Yogendra Kumar Jain
Department of Computer Science and Engineering, Samrat Ashok Technological Institute, Vidisha (M.P.)

Abstract -
In earlier, two smart card based password authentication key exchange protocols were proposed by lee et al. and Hwang et al. respectively. But neither of them achieves two factor authentication fully since they would become complete insecure once one factor is broken. To overcome these two factor authentication problem in password authentication key exchange protocol (PAKE) proposed a new efficient PAKE protocol with the concept of one time private key (OTPK) concept, which achieves fully two factor authentications and provide forward security of session keys. And to generate more strong session keys using true random number generation method for key generation.
 Index Terms - OTPK, TTP, 2-Factor, Authentication, Key-Exchange and PAKE
A Machine Learning Model for Stock Market Prediction
Pages: 17-23 (7) | [Full Text] PDF (739K)
Prof. Osman Hegazy, Prof. Omar Soliman and Mustafa Abdul Salam
Faculty of Computer and Informatics, Cairo University, Egypt
Higher Technological Institute (H.T.I), 10th of Ramadan City, Egypt

Abstract -
In this paper, Particle Swarm Optimization (PSO), a stochastic population-based evolutionary algorithm for problem solving is presented. PSO algorithm which has been successfully applied in many research and application areas is proposed to optimize least square support vector machine (LS-SVM) to predict the daily stock prices. Stock market prediction is the act of trying to determine the future value of a company stock or other financial instrument traded on a financial exchange. The successful prediction of a stock's future price will maximize investor's gains. The proposed model is based on the study of stocks historical data, technical indicators and optimizing LS-SVM with PSO algorithm. PSO selects best free parameters combination for LS-SVM to avoid over-fitting and local minima problems and improve prediction accuracy. Artificial neural network trained with Levenberg-Marquardt (LM) algorithm is used as a benchmark for comparison with proposed algorithm. Results presented in this paper show the potential of PSO algorithm in optimizing LS-SVM.
 Index Terms - Least Square Support Vector Machine, Particle Swarm Optimization, Technical Indicators and Stock Price Prediction
Classification of Cognitive Load Identification for User's Preference in Web Learning System using Particle Swarm Optimization Techniques
Pages: 24-28 (5) | [Full Text] PDF (443K)
L. Jayasimman and E. George Dharma Prakash Raj
Department of Computer Application, J.J. College of Engineering and Technology, Trichy, India
School of Computer Science and Engineering, Bharathidasan University, Trichy, India

Abstract -
Web based learning systems is a powerful learning environment which provides more benefits to the users compared to traditional learning systems. Web learning systems have effective methods, and adapt a personalized approach based on factors like preferences and emotions. This study emphasizes the system associated with cognitive aspects. Web learning users need innovative methods in the learning system to ensure motivated learning. Web learning system also emphasizes user's cognitive characteristics, experience and demands and so its design requires a thorough understanding of learner's activity to improve the user's cognitive approach based the user's requirements. In this paper user's preferences of web learning system with cognitive load are classified using neural network. To improve the classification accuracy particle swarm optimization is used to optimize the neural network parameters.
 Index Terms - Web Learning, Cognitive Load, Multi Layer Perceptron (MLP) and Particle Swarm Optimization Classification Accuracy