Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/299510
Title: Novel heuristic trust based cloud scheduling architecture
Researcher: Manickam M
Guide(s): Rajagopalan SP
Keywords: Novel heuristic
Cloud scheduling architecture
Different Intrusion Detection Systems
University: Anna University
Completed Date: 2019
Abstract: Cloud computing enables ubiquitous ondemand Internet access to computing resources that can be provisioned with minimal interaction with service providers The cloud delivers the demand of users for near consistent access to their information resources and data Today many organizations are moving their computing services towards the Cloud This makes their computer processing available much more conveniently to users However it also brings new security threats and challenges about safety and reliability Cloud computing is a network of networks over the internet therefore chances of intrusion is more with the erudition of intruder s attacks Cloud computing is distributed in nature hence chances of intrusion is more Analysis of various techniques of intrusion detection and prevention is newlineessential Different Intrusion Detection Systems IDS techniques are used to counter malicious attacks in traditional networks For Cloud computing enormous network access rate, relinquishing the control of data and applications to service providers and distributed attacks vulnerability an efficient, reliable and information transparent IDS is required Cloud computing systems are used by many people therefore they generate huge amount of logs In response to this there is an increase in the usage of IDS as a way to identify attack patterns malicious actions and unauthorized access to an environment To enhance the classification accuracy of IDS a novel clustering and classification method is proposed They are Possibilistic Fuzzy C Means PFCM and Recurrent Neural Network RNN PFCM model characterizes a clustering technique proficiently employed to determine the centroid In the fuzzy clustering data elements are placed in multiple clusters and linked with each and every element with a set of membership levels In the Possibilistic C Means PCM method each cluster is independent of the other cluster which goes a long way in assisting the recognition of the noise points PFCM is competent to proficiently overwhelm the thorny issues of the FCM and PCM techniques and elegantly carries out the clustering procedure The basic corner stone of a RNN are the neurons effectively linked by the synaptic links connections whose synaptic strength is appropriately coded by a weight The request of newlinePFCM clustering techniques enhances the clustering process and accuracy of the IDS Each of the output clusters from the PFCM is trained using N number of RNN classifier. newline newline
Pagination: xvi,161p.
URI: http://hdl.handle.net/10603/299510
Appears in Departments:Faculty of Information and Communication Engineering

Show full item record


Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).

Altmetric Badge: