Mobile sensor networks enable the monitoring of remote and hostile environments without requiring human supervision. Several approaches have been proposed in the literature to let mobile sensors self-deploy over a region of interest. In this paper we study, for the first time, the vulnerabilities of one of the most referenced approaches to mobile sensor deployment, namely the Voronoi-based approach. We show that, by compromising a small number of sensors, an attacker can influence the sensor deployment causing a significant reduction of the monitoring capability of the network. We propose a secure deployment algorithm called SecureVOR. We formally prove that SecureVOR has guaranteed termination and that it allows legitimate sensors to detect the malicious behavior of compromised nodes. We also show by extensive simulations that SecureVOR is able to fulfill the network monitoring goals even in presence of an attack, at the expense of a small performance overhead. © 2014 IEEE.
Voronoi-based deployment of mobile sensors in the face of adversaries / Bartolini, Novella; Bongiovanni, Giancarlo; T., La Porta; S., Silvestri; F., Vincenti. - STAMPA. - (2014), pp. 532-537. (Intervento presentato al convegno 2014 1st IEEE International Conference on Communications, ICC 2014 tenutosi a Sydney, NSW nel 10 June 2014 through 14 June 2014) [10.1109/icc.2014.6883373].
Voronoi-based deployment of mobile sensors in the face of adversaries
BARTOLINI, NOVELLA;BONGIOVANNI, Giancarlo;
2014
Abstract
Mobile sensor networks enable the monitoring of remote and hostile environments without requiring human supervision. Several approaches have been proposed in the literature to let mobile sensors self-deploy over a region of interest. In this paper we study, for the first time, the vulnerabilities of one of the most referenced approaches to mobile sensor deployment, namely the Voronoi-based approach. We show that, by compromising a small number of sensors, an attacker can influence the sensor deployment causing a significant reduction of the monitoring capability of the network. We propose a secure deployment algorithm called SecureVOR. We formally prove that SecureVOR has guaranteed termination and that it allows legitimate sensors to detect the malicious behavior of compromised nodes. We also show by extensive simulations that SecureVOR is able to fulfill the network monitoring goals even in presence of an attack, at the expense of a small performance overhead. © 2014 IEEE.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
