Volume 11, Issue 9, September 2019

 

 

A Collaborative SYN Flooding Detection Approach using NFV
 

A Collaborative SYN Flooding Detection Approach using NFV
Pages: 186-196 (11) | [Full Text] PDF (1.29 MB)
T Alharbi, A Aljuhani, B Taylor
The Catholic University of America, United States, University of Jeddah, Saudi Arabia
The Catholic University of America, United States, University of Tabuk, Saudi Arabia
The Catholic University of America, United States

Abstract - Detection and mitigation of Distributed Denial of Service (DDoS) attacks require immediate and automatic response in real-time to prevent or minimize the system downtime. We argue that Network Functions Virtualization (NFV) opens up the door to new solutions to these challenges. This paper focuses on SYN flooding as a common and powerful form of DDoS attack. We present an NFV-based solution to detect and mitigate SYN flooding attacks without human intervention. The proposed solution provides an internally collaborative detection framework to ensure that the detection system resists during the attacks. Our defense system aggregates and analyzes multiple reported SYN patterns to generate customized blocking rulesets based on the detected patterns. Simulated attack testing scenarios demonstrate the effectiveness of our approach.
 
Index Terms - Distributed Denial of Service (DDoS), SYN Flooding, Network Functions Virtualization (NFV), Detection, Firewall

Citation - T Alharbi, A Aljuhani, B Taylor. "A Collaborative SYN Flooding Detection Approach using NFV." International Journal of Computer Engineering and Information Technology 11, no. 9 (2019): 186-196.
An Improved Hybrid Evolutionary Clustering Algorithm to Mitigate Empty Clustering Problem
 

An Improved Hybrid Evolutionary Clustering Algorithm to Mitigate Empty Clustering Problem
Pages: 197-203 (7) | [Full Text] PDF (551 KB)
YA Joarder, F Naznin, MA Awal, MZ Islam
Department of Computer Science and Engineering (CSE), World University of Bangladesh (WUB), Dhaka, Bangladesh
Department of Computer Science and Engineering (CSE), Green University of Bangladesh (GUB), Dhaka, Bangladesh
Department of Computer Science and Engineering (CSE),International University of Business Agriculture & Technology(IUBAT), Dhaka, Bangladesh
Department ofInformation and Communication Technology, Islamic University (IU),Kushtia, Bangladesh

Abstract - Clustering algorithm try to get groups or clusters of data points that belong together. The main aim of this research is to improve the K-MEANS’ clustering quality by eliminating empty clustering issue using the proposed hybrid partitioning algorithm titled Improved Hybrid Evolutionary Clustering with Empty Clustering Solution (IH(EC)2S ) and to do comparison of advanced experimental results among three best performing clustering algorithm: K-MEANS, H(EC)2S and IH(EC)2S respectively. Though, K-MEANS converges fairly quickly, achieving a decent solution is not guaranteed. The clustering quality is very dependent on the choice of the initial centroid selection; once the number of clusters increases, it starts to suffer from Empty Clustering issue. We have improved a hybrid partitioning algorithm that was proposed previously to remove empty cluster. Our proposed algorithm has employed Scalable K-MEANS++ algorithm in place of K-MEANS algorithm as it is efficient than the former one. Firstly, it clusters the whole data set. Secondly, it detects the empty cluster. Finally, it removes the empty clustering issue. Our analysis portion justifies the usages of Scalable K-MEANS++ as the best algorithm among the clustering algorithms in terms of performance and time complexity. Also, we have shown that IH(EC)2S gives the better performance than both K-MEANS and H(EC)2S.
 
Index Terms - hybrid partitioning algorithm, big data, data clustering, K-MEANS, evolutionary algorithm, Scalable K-MEANS++, cuckoo search algorithm, enhanced fireworks algorithm

Citation - YA Joarder, F Naznin, MA Awal, MZ Islam. "An Improved Hybrid Evolutionary Clustering Algorithm to Mitigate Empty Clustering Problem." International Journal of Computer Engineering and Information Technology 11, no. 9 (2019): 197-203.
Implementation of the Bio Heat Transfer Equation on BEECube FPGA Platform
 

Implementation of the Bio Heat Transfer Equation on BEECube FPGA Platform
Pages: 204-208 (5) | [Full Text] PDF (482 KB)
I Mellal, A Oukaira, E Kengne, A Lakhssassi
LIMA Laboratory, University of Quebec at Outaouais, Gatineau, Quebec, 18X 3X7, Canada

Abstract - The Hardware Implementation of the physical models offers an outstanding opportunity for engineers in computational computing techniques. Contrary to the software implementation, the physical hardware implementations present the advantage of speed up the computation with inexpensive and practical way. The Finite Difference Method (FDM) is one the most common numerical method used to solve Electromagnetic and Heat transfer problems. In this paper, we present a FPGA-based implementation of the Bio Heat Equation (BHT). We then show the architecture of the proposed model to be implemented in the FPGA platform BEECube. The results of the implementation and simulation are reported and discussed.
 
Index Terms - FPGA, BEECube, Bio Heat Equation, FDM

Citation - I Mellal, A Oukaira, E Kengne, A Lakhssassi. "Implementation of the Bio Heat Transfer Equation on BEECube FPGA Platform." International Journal of Computer Engineering and Information Technology 11, no. 9 (2019): 204-208.