Classification of advance malware for autonomous vehicles by using stochastic logic

Alsadat tabatabaei, S, Saraee, MH ORCID: https://orcid.org/0000-0002-3283-1912 and Dehghantanha, A ORCID: https://orcid.org/0000-0002-9294-7554 2018, Classification of advance malware for autonomous vehicles by using stochastic logic , in: 11th IEEE International Conference on Developments in eSystems Engineering DeSE2018, 3-5 September 2018, Cambridge, UK.

Full text not available from this repository. (Request a copy)

Abstract

Connectivity of vehicles allows the seamless power of communication over the internet but is not without its cyber risks. Many IoT communication systems - such as vehicle-to-vehicle or vehicle-to-roadside - may require latencies below a few tens of milliseconds to cope with arbitrary and open-ended circumstances. These systems have severely limited resources. An autonomous vehicle, for example, can generate more tens of megabytes of data per second. Therefore, many resource-constrained IoT devices will rely on cloud services. Accordingly, the term “cloud-to-things” was born. In such an interaction between IoT devices and cloud services, taking a system offline for any reason - such as if a connected car were to be hacked - has significant consequences for the safety and privacy of passengers and other citizens. These factors currently fall far outside what mainstream IT security services can address: autonomous vehicles' safety systems should be able to withstand attacks and continue to function. To solve the issue, the operational requirements of safety features where close interactions between cyber systems and physical systems occur need to be carefully designed. The aim of this research is to propose an implementation of stochastic SVM and ANN classifiers. This approach gives a machine the ability to make predictive judgments about the effects of its actions as is shown in Fig.1. The system will train machine learning models both through both supervised and unsupervised algorithms. Next, by applying the cognitive intelligence to that system, the appropriate decisions on what to do about any detected situation will be performed.

Item Type: Conference or Workshop Item (Paper)
Schools: Schools > School of Computing, Science and Engineering > Salford Innovation Research Centre
Journal or Publication Title: Proceedings of the 11th IEEE International Conference on Developments in eSystems Engineering (DeSE2018)
Publisher: IEEE
Related URLs:
Depositing User: Prof. Mo Saraee
Date Deposited: 04 Feb 2019 14:33
Last Modified: 04 Feb 2019 14:33
URI: http://usir.salford.ac.uk/id/eprint/49936

Actions (login required)

Edit record (repository staff only) Edit record (repository staff only)