Towards a Lightweight Continuous Authentication Protocol for Device-to-Device Communication
Authors:
Syed W. Shah,
Naeem F. Syed,
Arash Shaghaghi,
Adnan Anwar,
Zubair Baig,
Robin Doss
Abstract:
Continuous Authentication (CA) has been proposed as a potential solution to counter complex cybersecurity attacks that exploit conventional static authentication mechanisms that authenticate users only at an ingress point. However, widely researched human user characteristics-based CA mechanisms cannot be extended to continuously authenticate Internet of Things (IoT) devices. The challenges are ex…
▽ More
Continuous Authentication (CA) has been proposed as a potential solution to counter complex cybersecurity attacks that exploit conventional static authentication mechanisms that authenticate users only at an ingress point. However, widely researched human user characteristics-based CA mechanisms cannot be extended to continuously authenticate Internet of Things (IoT) devices. The challenges are exacerbated with increased adoption of device-to-device (d2d) communication in critical infrastructures. Existing d2d authentication protocols proposed in the literature are either prone to subversion or are computationally infeasible to be deployed on constrained IoT devices. In view of these challenges, we propose a novel, lightweight, and secure CA protocol that leverages communication channel properties and a tunable mathematical function to generate dynamically changing session keys. Our preliminary informal protocol analysis suggests that the proposed protocol is resistant to known attack vectors and thus has strong potential for deployment in securing critical and resource-constrained d2d communication.
△ Less
Submitted 10 October, 2020;
originally announced October 2020.
Toward a Deep Learning-Driven Intrusion Detection Approach for Internet of Things
Authors:
Mengmeng Ge,
Naeem Firdous Syed,
Xi** Fu,
Zubair Baig,
Antonio Robles-Kelly
Abstract:
Internet of Things (IoT) has brought along immense benefits to our daily lives encompassing a diverse range of application domains that we regularly interact with, ranging from healthcare automation to transport and smart environments. However, due to the limitation of constrained resources and computational capabilities, IoT networks are prone to various cyber attacks. Thus, defending the IoT net…
▽ More
Internet of Things (IoT) has brought along immense benefits to our daily lives encompassing a diverse range of application domains that we regularly interact with, ranging from healthcare automation to transport and smart environments. However, due to the limitation of constrained resources and computational capabilities, IoT networks are prone to various cyber attacks. Thus, defending the IoT network against adversarial attacks is of vital importance. In this paper, we present a novel intrusion detection approach for IoT networks through the application of a deep learning technique. We adopt a cutting-edge IoT dataset comprising IoT traces and realistic attack traffic, including denial of service, distributed denial of service, reconnaissance and information theft attacks. We utilise the header field information in individual packets as generic features to capture general network behaviours, and develop a feed-forward neural networks model with embedding layers (to encode high-dimensional categorical features) for multi-class classification. The concept of transfer learning is subsequently adopted to encode high-dimensional categorical features to build a binary classifier. Results obtained through the evaluation of the proposed approach demonstrate a high classification accuracy for both binary and multi-class classifiers.
△ Less
Submitted 18 July, 2020;
originally announced July 2020.
