Industrial Networks and Intelligent Systems. 5th EAI International Conference, INISCOM 2019, Ho Chi Minh City, Vietnam, August 19, 2019, Proceedings

Research Article

Detecting Signalling DoS Attacks on LTE Networks

Download
164 downloads
  • @INPROCEEDINGS{10.1007/978-3-030-30149-1_23,
        author={Gin\^{e}s Escudero-Andreu and Konstantinos Kyriakopoulos and James Flint and Sangarapillai Lambotharan},
        title={Detecting Signalling DoS Attacks on LTE Networks},
        proceedings={Industrial Networks and Intelligent Systems. 5th EAI International Conference, INISCOM 2019, Ho Chi Minh City, Vietnam, August 19, 2019, Proceedings},
        proceedings_a={INISCOM},
        year={2019},
        month={9},
        keywords={LTE security DoS attacks RRC signaling attack Dempster-Shafer},
        doi={10.1007/978-3-030-30149-1_23}
    }
    
  • Ginés Escudero-Andreu
    Konstantinos Kyriakopoulos
    James Flint
    Sangarapillai Lambotharan
    Year: 2019
    Detecting Signalling DoS Attacks on LTE Networks
    INISCOM
    Springer
    DOI: 10.1007/978-3-030-30149-1_23
Ginés Escudero-Andreu1,*, Konstantinos Kyriakopoulos,*, James Flint1,*, Sangarapillai Lambotharan1,*
  • 1: Loughborough University
*Contact email: g.escudero-andreu@lboro.ac.uk, k.kyriakopoulos@lboro.ac.uk, j.a.flint@lboro.ac.uk, s.lambotharan@lboro.ac.uk

Abstract

As mobile communications increase their presence in our life, service availability becomes a crucial player for the next generation of cellular networks. However, both 4G and 5G systems lack of full protection against Denial-of-Service (DoS) attacks, due to the need of designing radio-access protocols focused on providing seamless connectivity. This paper presents a new method to detect a DoS attack over the Radio Resource Control (RRC) layer, offering three original metrics to identify such attack in a live Intrusion Detection System (IDS). The proposed metrics evaluate the connection release rate, the average session establishment and the session success rate to identify the attack. The presented results provide an average detection rate above 96%, with an average false positive rate below 3.8%.