Author :
Guerra, Federico ; Casari, Paolo ; Zorzi, Michele
Author_Institution :
Dept. of Inf. Eng., Univ. of Padova, Padova, Italy
Abstract :
Large-sized underwater testbed deployments still pose a number of challenges, first of all the cost of the nodes and of the deploying ships. Therefore, simulation becomes an important tool for the assessment of network performance, and a precious aid to protocol, topology, and deployment design. However, simulations are significant only if the reproduction of acoustic propagation is accurate. To this end, we have joined two well known tools for simulation, namely Bellhop (for acoustic propagation modeling through ray tracing) and ns2-MIRACLE (an event-based network simulator). These tools, together, provide a flexible and customizable environment, fostering more realistic reproduction of propagation, PHY-level behavior, as well as the detailed specification of medium access control, routing, and higher-level protocols. In this paper, we describe our tool, and give an example of its employment in the comparison of three MAC protocols for underwater networks over different kinds of physical layers. The protocols have been specifically chosen to shed some light on the relationship between the complexity of a protocol and the amount of coordination it enables among nodes: in other words, our results show when it is better to rely on plain random access, and when on some form of handshaking (despite the usually greater complexity of handshake-based protocols).
Keywords :
access protocols; ray tracing; underwater acoustic communication; Bellhop; MAC protocols; acoustic propagation modeling; event-based network simulator; network performance; ns2-MIRACLE; physical layers; ray tracing; realistic channel simulator; underwater networks; underwater testbed deployments; Access protocols; Acoustic propagation; Costs; Discrete event simulation; Marine vehicles; Media Access Protocol; Network topology; Ray tracing; Routing protocols; Testing;