Prevention of Location-related Attacks in Wireless Networks

Abstract

Like many other technological developments, wireless networks have moved away from the military arena that spawned them and are now applied to mainstream civilian activities. These networks are becoming increasingly popular due to features such as easy setup and absence of network cables. Location awareness is an essential characteristic of various types of wireless networks. It can provide geographic services and maximize network utilization. However, location awareness can also be exploited by nefarious individuals whose aim it is to compromise networks. Therefore, in this study, we focus on the prevention of location-related attacks in various types of wireless networks. First, we address the security of underwater vehicular ad hoc networks (VANETs) for the first time. A VANET allows for communications among nearby vehicles and between vehicles and stationary roadside base stations. Mobile nodes of traditional VANETs are typically thought of as road vehicles that follow paved roads and VANETs are expected to utilize radio frequency (RF) technologies such as WiFi, cellular, satellite, and WiMAX. However, with the emergence of sQuba, a new type of car that can be driven both on land and under water, the paradigm has shifted. Hence, it is time that the scope of VANETs be extended to encompass underwater communication among submersible cars such as sQuba. There are important physical and technological differences between terrestrial and underwater VANETs. The high rate of absorption of electromagnetic signals in water limits the applicability of RF technologies, and hence, acoustic communication is considered to be the only candidate technology for underwater VANETs. As conventional sonar techniques are not appropriate for submersible cars, we propose and present a novel sonar algorithm geared toward ascertaining the position of target vehicles in underwater VANETs. Based on this proposed sonar method, we describe countermeasures against various location-related security threats such as distance fraud attacks, Sybil attacks, and wormhole attacks. Second, we propose two unilateral lightweight distance bounding protocols. The first protocol is applied bidirectional challenges without a final confirmation message. In relay attacks, this protocol reduces the adversary's success probability to (5/16)^n = (1/3.2)^n. A distance bounding protocol provides an upper bound on the distance between communicating parties by measuring the round-trip time between challenges and responses. It is usually designed to defend against relay attacks and distance fraud attacks. The adversary's success probability of previous distance bounding protocols without a final confirmation message, such as a digital signature or a message authentication code, is at least (3/8)^n = (1/2.67)^n. The second proposed protocol focuses only on the relay attack because there are applications for which distance fraud attacks are not critical threats. This protocol has a message authentication code, is more efficient than previous protocols that utilize a final confirmation message and has a low false acceptance rate under relay attacks. Previous distance bounding protocols generally have three passes in the slow exchange phase: two passes are used to exchange random numbers between the parties, and one pass is used to send a final confirmation message. To reduce computational complexity, in the first of our two proposed protocols, one pass is removed from the slow exchange phase (there is no final confirmation message)

in the second proposed protocol, the two passes used to exchange random numbers between the verifier and the prover are removed.