On the performance of unsynchronized distributed MAC protocols in deep water acoustic networks

Recent simulation software such as WOSS [1] and similar packages make it possible to perform accurate simulations of underwater acoustic networks by taking into account the actual behavior of sound propagation. In this paper, we focus on deep-water acoustic networks, and employ WOSS to compare the performance of three MAC protocols based on random access. The final goal of this effort is to test several network scenarios and baseline protocol approaches, and to understand which is more likely to offer top performance in underwater networks. In fact, it is still a partly open question whether or not there is a medium access control (MAC) protocol that is "good" in most if not all scenarios and network arrangements; or, at least, if any MAC protocol would be able to emerge as the de facto choice for most applications and network topologies. In this paper, we perform a comparison of MAC protocols in deep-water scenarios, in order to extend previous analyses, which targeted mainly shallow water networks. We will consider three types of medium access control: random access with instantaneous carrier sensing [1], transmitter-side tone-driven contention [2], and 3-way Request-To-Send/Clear-To-Send (RTS/CTS)-based access [3].