Title: SNTAM: secure neuro-fuzzy-based trusted neighbour selection and attackers detection in multipath ad-hoc network
Authors: Sandeep Patil; Lalita Admuthe; Meenakshi Patil
Addresses: Department of Electronics & Telecommunication Engineering, DKTE'S Textile & Engineering Institute, Ichalkaranji, Maharashtra, India ' Department of Electronics & Telecommunication Engineering, DKTE'S Textile & Engineering Institute, Ichalkaranji, Maharashtra, India ' Department of Electronics and Communication Engineering, CMR Institute of Technology, Bengaluru, Karnataka, India
Abstract: An ad-hoc network comprises a set of wireless nodes with no pre-defined infrastructure. The most well-known attribute of a wireless network is dynamic. They also have inadequate channel strength, security belief issues and functioning with narrowed battery power. This study explores the capabilities of working with different attacks like a vampire attack. This attack is forever hindrance to the networks by reducing the node's bandwidth and energy. In addition, the protocol detects a replica attack that can capture the neighbour's address and act as a genuine. These two attacks can spoil the entire system activities and use network abilities. The proposed Secure Neuro-fuzzy Based Trusted Neighbour Selection and Attackers Detection in Multipath Ad-Hoc Network (SNTAM) protocol focused to eliminate these attackers from the network to form the communication path by an optimised way of neuro-fuzzy techniques. The results show that SNTAM gives effective results in securing the network in all the parameters and attributes especially packet delivery ratio (PDR) which is almost 100% and residual energy of nodes which is more than compared protocol. Furthermore, SNTAM protocol can avoid sampled malicious nodes effectively.
Keywords: MANET; channel capacity; energy; attackers; vampire attack; replica attack; trust; reliability.
International Journal of Internet Protocol Technology, 2023 Vol.16 No.2, pp.119 - 130
Accepted: 14 May 2022
Published online: 06 Jun 2023 *