Showing papers by "Balamurugan Balusamy published in 2018"

A Provably Secure Three-Factor Session Initiation Protocol for Multimedia Big Data Communications

Abstract: The session initiation protocol (SIP) is an IP-based telephony authentication mechanism for multimedia big data communications over the Internet. It is used to set up, and control voice and video calls, as well as for instant messaging. One of the concerns of this kind of open-text-based protocol is the security for user authentication. The HTTP digest-based challenge-response authentication process is used in the original SIP. However, this kind of authentication procedure is insecure and a pre-existing user configuration on the remote server is required. According to the literature, several authentication mechanisms for SIP are already devised, but none of these SIPs are robust against existing security attacks. Therefore, we design a three-factor SIP (TF-SIP) for multimedia big data communications, which is robust and flexible against existing known security issues. We show that our TF-SIP is provably secure in the random oracle model. We formally verify the mutual authentication and the freshness of the agreed session key between the user and the remote server using the BAN logic analysis. We found that the communication and computation costs are low, but the storage cost is slightly higher for our TF-SIP in comparison with other SIPs.

A framework for smart traffic management using hybrid clustering techniques

Abstract: Due to increase in traffic in cities and on major roads, it has become a necessity to have an efficient traffic management system to handle such scenarios. Present traffic management system performs mere traffic monitoring and event handling which cannot be a viable system for highly populous country like India and China. In this paper, we propose a system that will predict the densely populated roads based on the present and past traffic congestion. This system also suggests the alternate paths for the given source and destination. A simulation of live stream of online data is performed on legacy traffic data set which is processed incrementally. Density based clustering is applied after Fuzzification of data to assign weightage for the densely congested path on the route map. The weightage for the path on the given time helps to decide the best route form the source to destination. Floyd’s algorithm is also applied to find the shortest set of alternate path for the given source to destination.

Survey of energy drink consumption and adverse health effects in Lebanon

Abstract: Consumption of energy is a national and international phenomenon that showed increase in market spread and profits from 1990 and made the emergence of many brands. Energy drinks are aggressively marketed with the claim that these products give an energy boost to improve physical and cognitive performance. However, studies supporting these claims are limited. The study examines the new phenomena of energy drinks among university students in Lebanon, based on the participants’ personnel characteristics, university grade and the impact on health status. The study also determined whether high frequency of consumption was correlated with negative physical health symptoms. A cross-sectional study survey was undertaken on students aged between 18 and 30 years in private university over three branches (Beirut, Tripoli and Saida). A self-administered questionnaire was used inquiring about socio-demographic characteristics, consumption patterns and side effect of energy drinks. Data was analyzed using SPSS 24. Findings showed a serious concern exists for the health and safety of the most at risk students who engaged in daily energy drink usage when two-thirds of these reported difficulties sleeping, more than one experienced heart palpitation and blood pressure; one-third had anxiety, nervousness and feeling thirsty, and one fifth indicated tiredness and headache. Such symptoms are reported with excessive consumption of caffeine that had adverse health effect on the body.

A Fuzzy Based Hybrid Firefly Optimization Technique for Load Balancing in Cloud Datacenters

Abstract: Cloud computing technology has massive inferences with the use of virtualization technologies. Most of the organizations have incorporated to practice the virtualization strategies to create and operate an effective dynamic data center. The growing maturity of the technologies and utilities of the cloud make the users hasten the adoption of the cloud. The dynamic demanding nature of cloud resources leads to an imbalance in virtual machine utilization and radically increases the energy consumption and operating cost of the data center. In this paper, we propose a fuzzy based hybrid load balancing algorithm for the optimal utilization of virtual machines. The proposed algorithm aim is to reduce the makespan, response time and cost with minimal energy usage and resource wastage. The fuzzy based hybrid optimization approach unveils better performance than existing metaheuristic load balancing algorithms.