Home // AICT 2016, The Twelfth Advanced International Conference on Telecommunications // View article
Authors:
Fernando Almeida
Admilson Ribeiro
Edward Moreno
Carlos Montesco
Keywords: IDS System; IoT Internet of Things; Multilayer Perceptron; Neural Network; Limited Weights;
Abstract:
One way to prevent attacks to security in the Internet of Things (IoT) is the adoption of an Intrusion Detection System (IDS). With the use of an Artificial Neural Network (ANN) it is possible to decrease the limitations of the IDS such as false positive that can compromise the system. In this paper, we evaluate the performance of two ANNs to verify which of both is the more adequate to use in an IDS for the IoT environment. We compare the performance of a Multilayer Perceptron (MLP) with Limited Weights with a Multilayer Perceptron with normal weights. The used Multilayer Perceptron presents ten neurons in the hidden layer. The implementation is in C language and run on an embedded platform with an ARM Cortex-M3 micro-controller. It is possible to consider the ANN training in another platform and to permit the embedded platform receives the trained weights. It is also possible to make the training in real time using the received data one time. We conclude that it is viable to use an Artificial Neural Network Multilayer Perceptron in an Intrusion Detection System for the Internet of Things.
Pages: 82 to 87
Copyright: Copyright (c) IARIA, 2016
Publication date: May 22, 2016
Published in: conference
ISSN: 2308-4030
ISBN: 978-1-61208-473-2
Location: Valencia, Spain
Dates: from May 22, 2016 to May 26, 2016