Informatics Publishing Limited

M. Durairaj ¹, A. Persia ¹

Affiliations
1 School of Computer Science, Engineering & Applications, Bharathidasan University, Tamilnadu, IN

Abstract

The major cause of threat circling around the wireless local area infrastructure network is Denial of Service (DoS) attack. The anticipated users will be denied from availing the resources by spoofing the Medium Access Control (MAC) address of the legitimate client/AP. Less protection in MAC address led to get easy spoofing. Since, the management frame has been unencrypted; adversary sends the management frame to the victim using the spoofed MAC address. This prerequisite goaded to offer an effective prevention mechanism for a DoS attack. Many security mechanisms are obtainable which have their own drawback in providing complete prevention of DoS attack. This paper proposes ThreV mechanism which detects and prevents the intruder in an efficient manner. The solutions are simulated using NS2, to analyze how far it prevents the DoS attacks. The simulation result of the proposed algorithm measures the packet delivery ratio, control overhead, normalized routing overhead, throughput, delay and packet loss. The experimental result proved the effectiveness of proposed ThreV algorithm in detecting and preventing MAC spoof DoS attacks.

Keywords

ANM, DoS Attack, MAC Spoof, ThreV, TpatLetEn, WLANn

Full Text

   

M. Durairaj ¹, A. Manimaran ¹

Affiliations
1 School of Computer Science Engineering & Applications, Bharathidasan University, Trichy-620023, IN

Abstract

Cloud based E-Learning is the method to reduce cost and complexity of data accessing, which are controlled by third party services. Traditional E-Learning methods are incorporated with cloud computing technology to provide massive advantages to the academic users but it compromises in security aspects. Proposed methodology ensures data availability and provides solution to protect indispensable data from the attackers. This study identifies different security issues in cloud service delivery model with an aim to suggest a solution in the form of security measures related to the cloud based e-learning. Different types of attacks in service delivery models of e-learning proposed by different researchers are discussed. Threats, security requirements, and challenges involved are also taken into consideration. This study of e-Learning models advocates users to access their data in the cloud through a secured layer using the internet.

Keywords

Cloud Security, Cloud Service Delivery Models, Cloud Security Attacks, E-Learning, Data Availability, Cloud Security Challenges and Threats.

Full Text

   

V. V. Nivargi ¹, M. Durairaj ¹, C. Makhale ¹

Affiliations
1 Ruby Hall Clinic, Pune, IN

Abstract

A 42 year old lady with a diagnosis of hyperreactive airway disease since 4 years, treated with bronchodilators presented with monomorphic ventricular tachycardia with syncope requiring direct current (DC) cardioversion. 2 dimensional echocardiography with cardiac colour doppler examination and magnetic resonance Imaging (MRI) revealed left ventricular dilatation, systolic dysfunction with patchy areas of thickening and regional wall motion abnormality not specific to any particular vascular territory and advanced diastolic dysfunction.

Review of records revealed persistently raised erythrocyte sedimentation rate which was previously ignored.

Gadolinium contrast cardiac MRI revealed late gadolinium enhancement suggestive of myocarditis. Thoracic MRI revealed a mediastinal lymphnode mass. A computed tomography (C.T.) directed lymphnode biopsy confirmed Sarcoidosis. She was offered implantable cardioverter-defibrillator device and immunosuppression therapy with prednisolone 1 mg/kg/day.

Keywords

Aborted Sudden Death, Cardiac Sarcoidosis.

Full Text

     

Sivagowry Shathesh ¹, M. Durairaj ¹

Affiliations
1 School of Computer Science, Engineering and Applications, Bharathidasan University, Trichy – 620024, Tamil Nadu, IN

Abstract

Background/Objectives: The objective is to generate rules to predict the severity of Cardiac disease by applying Rough Set Theory, Particle Swarm Optimization (PSO) with Artificial Neural Network (ANN) and Fuzzy Logic (FL). Methods/ Statistical Analysis: In order to achieve the aim, the Rough Set Theory is hybrided with PSO to give Intelligent Hybrid Quick Reduct Particle Swarm Optimization (IHQRPSO) Algorithm which explore the search space and find the optimal data. The optimal data set is then classified by using ANN as disease and undiseased patients and finally the rules are framed using FL to predict the severity of disease. Findings: By using the IHQRPSO Algorithm, the attribute number has been reduced from 13 to 5. The optimal data set is then trained by ANN which used Back propagation Learning algorithm for classification. The classified unhealthy patients are then categorized to risk level such as mild, low and severe based on the rules obtained from Fuzzy Inference System. A total of 63 rules is framed. The whole Hybrid mechanism is then compared with the existing work which used Genetic Algorithm and Fuzzy Temporal Rule Mining Algorithm for prediction. While comparing the proposed hybrid mechanism with existing one, the prediction accuracy has been increased to a considerable level. Applications/Improvements: The Performance of the algorithm is evaluated in terms of accuracy. The accuracy was 95.25% while using the proposed algorithm where it was around 85 while using the existing work.

Keywords

Cardio Vascular Disease, Hybrid Mechanism, Intelligent, Prediction, Risk

Full Text