Secure Communication in Mobile Underwater Wireless Sensor Networks Anjana P Das College of Engineering, Trivandrum, India LBS Centre for Science and Technology Trivandrum, India Sabu M Thampi Indian Institute of Information Technology and Management-Kerala India Abstract—Underwater sensor networks (UWSN) provide a promising solution for several aquatic applications. Mobile UWSNs are vulnerable to various threats because of their harsh environment, acoustic communication channel, bandwidth limitation, high propagation delay, multipath effect, and mobility. Therefore, appropriate security mechanisms should be included in networking services, such as localization, to facilitate time critical applications. In this paper, we discuss various Denial-of- Service attacks in mobile UWSNs. We simulated a flooding attack and an out-of-coverage problem caused by mobility by using Aqua-Sim, and analyzed their impact on UWSN performance. In addition, we surveyed UWSN localization techniques and secure localization schemes of terrestrial wireless sensor net- works (WSN). Simulation experiments were conducted to analyze the performance variation between mobile UWSNs and mobile WSNs. The study concluded that the variation in the network performance of mobile UWSNs varies from that of WSNs. Therefore, the security mechanisms proposed for terrestrial WSNs are unsuitable for UWSNs. Index Terms: Denial of Service (DoS); localization; un- derwater sensor networks (UWSNs); wireless sensor networks (WSNs) I. I NTRODUCTION Underwater sensor networks (UWSNs) constitute sensor nodes or underwater vehicles that perform collaborative mon- itoring tasks in a marine environment. UWSNs have a wide range of applications including oceanographic data collection, pollution monitoring, offshore exploration, tactical surveil- lance, and disaster detection and measurement. An under- water network with moving sensor nodes constitutes mobile UWSNs. UWSNs and terrestrial wireless sensor networks (WSNs) have few common properties, such as high node density and limited energy supplies. However, they differ notably from terrestrial sensor networks in many aspects, such as acoustic channels, high latency, node mobility, high error probability, and three-dimensional network topology. These conditions introduce several challenges in designing UWSNs. A comparison between WSNs and UWSNs is shown in Table I. UWSNs are mobile because sensor nodes move with water flow. Underwater objects move at a speed of 2–3 knots (or 36 km/hr). This mobility pattern results in a highly dynamic network topology. Acoustic channels are typically preferred in UWSNs be- cause of high attenuation of radio signaling. The propagation speed of acoustic signals in water is approximately 1.5 × 10 3 m/s, whereas the propagation speed of a radio frequency signal in air is 3 × 10 8 m/s. In general, UWSNs are established by the cooperation among several sensor nodes, which upholds the network through bidirectional acoustic channels. The ar- chitecture of a typical UWSN is shown in Fig. 1. It consists of wireless sensor nodes deployed in an underwater environment and multiple base stations floating on the water surface with aquatic and RF modems. The base stations communicate with terrestrial wireless stations by using radio channels. UWSNs are highly error-prone because the acoustic communication channels are significantly affected by factors such as signal attenuation, noise, multipath effect, Doppler spread, and water temperature. These factors cause high bit-error rate and delay variance. Moreover, sensor nodes in UWSNs have a higher node failure rate and packet loss probability than those in terrestrial WSNs [1]–[6]. Fig. 1. Underwater sensor network (UWSN) Because of the special characteristics of underwater chan- nels and the mobility of sensor nodes, UWSNs are vulnerable to various attacks, and a clear line of defense has not been proposed so far. Denial of Service (DoS) is an attempt to make a network resource or service unavailable to its intended users. In an UWSN environment, DoS is a low-cost attack, which disrupts the communication among sensor nodes causing im- mense degradation of data availability, thus affecting data freshness. Because networking services, such as localization and routing highly depend on the availability and freshness of data, a malicious attacker can easily distort the entire network. An acoustic wave travels faster and longer in water than in air. An adversary can easily intercept or deny communication messages through this open acoustic channel. In this paper, we discuss various security threats against mobile UWSNs. The main contributions of this study are as 2164 978-1-4799-8792-4/15/$31.00 c 2015 IEEE