A Study on Proposed Distributed Attack Detection Algorithm Using Experimental and Simulation Analysis
Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2016, Vol 18, Issue 3
Abstract
Abstract: The attack detection rates are higher for large node deployment densities, as compared to the low density counterparts. This is because the attack detection scheme relies on both the individual traffic observations by the detector nodes, as well as the subsequent verification and reconstruction of observed traffic sub patterns by the detector nodes as well as the mGN nodes, for classification purposes. In this work, we evaluated the performance of our proposed distributed attack detection scheme, through experiments andsimulation analysis. The scheme proposed, performs distributed denial of service attack detection, in the presence of injected nodes, inclusive of laptop-class nodes, in the network. As part of the detection process, detector nodes monitor network traffic flow towards a set of victim nodes, and further, reconstruct patterns of observed network traffic, to facilitate attack decision making. The performance of the scheme is affected by several algorithmic as well as network parameters, which need to be defined, at network initialization time. We studied the effect of variation of these algorithmic and network-level parameters on the outcomes of the proposed attack detection scheme. We quantified the results obtained for variations in these parameter values,based on simulation experiments. The metrics that are compared and analyzed as part of the simulation experiment are: attack detection rate, false positive rate, false negative rate and detector/mGN node energy utilization rates. We also performed a comparative analysis of the acquired experimental results for the proposed scheme, with corresponding results obtained from the simulation of a centralized Self Organizing Map-based attack detection scheme. The purpose of this comparison is to establish the superiority of the proposed distributed pattern recognition approach over other centralized techniques for detection of distributed denial of service attacks in wireless sensor networks.
Authors and Affiliations
Siva Balaji Yadav. C , R. Seshadri,PhD
Keywords
Related Articles
- EP ID EP123183
- DOI -
- Views 134
- Downloads 0
Reducing location error in Wireless Sensor Networks
Abstract: Localization is the basic problem in the wireless sensor networks. All the result data get wasted if the node is placed in the wrong position. Range based and Range free algorithms are used for locating t...
Identity Based Replicated Node Detection and Localization in Wireless Network
Abstract: Wireless networks are more prone towards identity-based attacks, especially the spoofing attacks. These spoofing attacks can further cause many other forms of attacks such as rouge access point attack and...
A Mobile Surveillance Robot Over The Wifi Network Using Atmega 8
Technology is currently growing very rapidly, and evolved changes, both robots using media technology cable or Bluetooth robot, which is in use is limited by distance. Smart computers are growing very rapidly including S...
A Portable Content Management System
Abstract: This report describes the design, implementation and usage of an efficient Content Management System.The aimof content management systems is to store and organize files, and provide version-controlled and...
Efficient Refining Of Why-Not Questions on Top-K Queries
Abstract: After decades of effort working on database performance, the quality and the usability of databasesystems have received more attention in recent years. In particular, answering the why-not questions after asear...