Hybrid intrusion detection in connected self-driving vehicles

Emerging self-driving vehicles are vulnerable to different attacks due to the principle and the type of communication systems that are used in these vehicles. These vehicles are increasingly relying on external communication via vehicular ad hoc networks (VANETs). VANETs add new threats to self-driving vehicles that contribute to substantial challenges in autonomous systems. These communication systems render self-driving vehicles vulnerable to many types of malicious attacks, such as Sybil attacks, Denial of Service (DoS), black hole, grey hole and wormhole attacks. In this paper, we propose an intelligent security system designed to secure external communications for self-driving and semi self-driving cars. The proposed scheme is based on Proportional Overlapping Score (POS) to decrease the number of features found in the Kyoto benchmark dataset. The hybrid detection system relies on the Back Propagation neural networks (BP), to detect a common type of attack in VANETs: Denial-of-Service (DoS). The experimental results show that the proposed BP-IDS is capable of identifying malicious vehicles in self-driving and semi self-driving vehicles.

[1]  Khattab M. Ali Alheeti,et al.  An intrusion detection system against malicious attacks on the communication network of driverless cars , 2015, 2015 12th Annual IEEE Consumer Communications and Networking Conference (CCNC).

[2]  Klaus D. McDonald-Maier,et al.  On the detection of grey hole and rushing attacks in self-driving vehicular networks , 2015, 2015 7th Computer Science and Electronic Engineering Conference (CEEC).

[3]  Khattab M. Ali Alheeti,et al.  An Intrusion Detection System against Black Hole Attacks on the Communication Network of Self-Driving Cars , 2015, 2015 Sixth International Conference on Emerging Security Technologies (EST).

[4]  Hirofumi Yamaki,et al.  Unknown Attacks Detection Using Feature Extraction from Anomaly-Based IDS Alerts , 2012, 2012 IEEE/IPSJ 12th International Symposium on Applications and the Internet.

[5]  J. Ramkumar,et al.  Fuzzy Logic Approach for Detecting Black Hole Attack in Hybrid Wireless Mesh Network , 2015 .

[6]  J.Ramkumar,et al.  Fuzzy Logic Approach for Detecting Black HoleAttack in Hybrid Wireless Mesh Network , 2014 .

[7]  Mohammed Erritali,et al.  A review and classification of various VANET Intrusion Detection Systems , 2013, 2013 National Security Days (JNS3).

[8]  Venus W. Samawi,et al.  The affect of fuzzification on neural networks intrusion detection system , 2009, 2009 4th IEEE Conference on Industrial Electronics and Applications.

[9]  Andrew Harrison,et al.  A feature selection method for classification within functional genomics experiments based on the proportional overlapping score , 2014, BMC Bioinformatics.

[10]  Los Angeles,et al.  Doctor of Philosophy in Computer Science , 2007 .

[11]  Lie-Liang Yang,et al.  Reduced-Rank Adaptive Least Bit Error-Rate Detection in Hybrid Direct-Sequence Time-Hopping Ultrawide Bandwidth Systems , 2009, 2009 IEEE International Conference on Communications.

[12]  Xiaohui Liang,et al.  Sybil Attacks and Their Defenses in the Internet of Things , 2014, IEEE Internet of Things Journal.

[13]  Zied Choukair,et al.  Link analysis approach to improve detection of fragmentation attacks in Misuse IDS , 2009, 2009 First International Conference on Communications and Networking.

[14]  Klaus D. McDonald-Maier,et al.  Securing MEMS Based Sensor Nodes in the Internet of Things , 2015, 2015 Sixth International Conference on Emerging Security Technologies (EST).

[15]  Steven E. Shladover,et al.  Potential Cyberattacks on Automated Vehicles , 2015, IEEE Transactions on Intelligent Transportation Systems.

[16]  Frank Kargl,et al.  Pseudonym Schemes in Vehicular Networks: A Survey , 2015, IEEE Communications Surveys & Tutorials.

[17]  Hiroki Takakura,et al.  Statistical analysis of honeypot data and building of Kyoto 2006+ dataset for NIDS evaluation , 2011, BADGERS '11.

[18]  Ramesh Sharda,et al.  Identifying significant predictors of injury severity in traffic accidents using a series of artificial neural networks. , 2006, Accident; analysis and prevention.

[19]  Laith Al-Jobouri,et al.  Increasing the rate of intrusion detection based on a hybrid technique , 2013, 2013 5th Computer Science and Electronic Engineering Conference (CEEC).