Detection and Prevention of Manet using Hybrid SVM with Ann
Vishal Walia1, Rahul Malhotra2
1Vishal Walia*, PHD Scholar IGKPTU, Jalandhar, India.
2Dr. Rahul Malhotra, Director , Guru Teg Bahadur Khalsa Institute of Engineering & Technology, Chhapianwali, Malout, India
Manuscript received on November 22, 2019. | Revised Manuscript received on December 15, 2019. | Manuscript published on December 30, 2019. | PP: 4463-4469 | Volume-9 Issue-2, December, 2019. | Retrieval Number: B3537129219/2019©BEIESP | DOI: 10.35940/ijeat.B3537.129219
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Mobile Ad hoc Networks (MANET) have been exceptionally vulnerable against attacks because of the dynamic and self-configurable nature of its system foundation. This kind of wireless network is appropriate for temporary communication linked due to its nature of less-foundation and there is no any control of centralized manner. Design a routing mechanism that are security aware with higher QoS parameter is very competetive and the major tasks involved in ad hoc types of network as per the limited power resources and their dynamic routing topology. This paper mainly focused on the design of a secure and trusts based on-demand routing mechanism using Ad-hoc on demand distance vector (AODV) protocol to compute trust-based produces path initialed from source up to destination that will fulfill minimum two end-to-end QoS parameters of network. So here, the generalized AODV routing protocol has been extended from traditional routing mechanism to analyze the performance of this model with combination of artificial intelligence concept. The proposed ad hoc based routing mechanism is used to found possible routes that are prevented through trust adjacent position of security validation protocols and enhanced link optimized route computes on the basis of Artificial Neural Network (ANN) as an artificial intelligence algorithm for well-organized communication in MANET. In addition, this research demonstrates the effectiveness of bio inspired Firefly Algorithm (FFA) as an optimization approach with the consideration of several performance QoS metrics of network. The results have been measured in terms of throughput and PDR with SVM and ANN approach. It has been observed that the throughput and PDR measured using ANN approach is better compared to SVM approach an average of 0.755 PDR value has been obtained using ANN approach.
Keywords: Mobile Ad hoc Networks, d-hoc on demand distance vector, Artificial Neural Network, Firefly Algorithm, SVM, and PDR.