Node Intrusion Tendency Recognition Using Network Level Features Based Deep Learning Approach

Abstract

Adhoc network is highly susceptible for intrusion attacks due to the simplified access
control and compacted network stack. Malicious node recognition in Mobile adhoc
network (MANET) is challengeable due to nodes mobility and limited coverage
of nodes. Thus, link may keep fluctuating throughout the communication period.
In this paper, deep analytic model is made for extracting attacker node behaviors
from networking point of view. Attributed such as link durations, re-healing time
and number of received packets (by attacker) was the main features of this work.
Later, deep learning paradigm is integrated to perform attacker node recognition. Data
obtained from network analytical model is used to train three different models namely
Feed forward neural network (FFNN), Cascade backpropagation neural network
(CBPNN) and Convolutional neural network (CNN). Attacker node recognition
accuracy of 85.5