Interest in underwater acoustic networks has grown rapidly with the desire to monitor the large portion of the world covered by oceans. Fundamental differences between underwater acoustic propagation and terrestrial radio propagation may call for new criteria for the design of networking protocols. In this paper, we focus on some of these fundamental differences, including attenuation and noise, propagation delays, and the dependence of usable bandwidth and transmit power on distance (which has not been extensively considered before in protocol design studies). Furthermore, the relationship between the energy consumptions of acoustic modems in various modes (i.e., transmit, receive, and idle) is different than that of their terrestrial radio counterparts, which also impacts the design of energy-efficient protocols. The main contribution of this work is an in-depth analysis of the impacts of these unique relationships. We present insights that are useful in guiding both protocol design and network deployment. We design a class of energy-efficient routing protocols for underwater sensor networks based on the insights gained in our analysis. These protocols are tested in a number of relevant network scenarios, and shown to significantly outperform other commonly used routing strategies and to provide near optimal total path energy consumption. Finally, we implement in ns2 a detailed model of the underwater acoustic channel, and study the performance of routing choices when used with a simple MAC protocol and a realistic PHY model, with special regard to such issues as interference and medium access

No abstract

No abstract

No abstract

No abstract

In this paper we present a novel framework for ns2 to facilitate the simulation and, in general, the design of beyond 3G networks. The set of libraries we wrote for this purpose is called Multi InteRfAce Cross Layer Extension for ns2 (MIRACLE). They enhance the functionalities offered by the Network Simulator ns2 by providing an efficient and embedded engine for handling cross-layer messages and, at the same time, enabling the coexistence of multiple modules within each layer of the protocol stack. For instance, multiple network, link, MAC or physical layers can be specified and used within the same node. The implications of this are manifold. First of all, the framework facilitates the implementation and the simulation of modern communication systems in ns2. Secondly, due to its modularity, the code will be portable, re-usable and extensible.As an example of the advantages offered by our architecture, we show how the MIRACLE framework can be used to quickly set up protocol architectures for Ambient Networks [1] and evaluate their performance in wireless and multi-technology environments. We stress that, even though the emphasis in the present paper is put on wireless systems, MIRACLE is a general framework which can be used for simulating wired networks as well as a mixture of wired and wireless scenarios. Throughout the paper we also discuss some of the downsides of existing ns2 extensions, which are often programmed in a rather ad hoc manner, according to specific needs or technologies and, as such, are often difficult to extend/re-use. In contrast, our effort aims at providing well defined interfaces and is based on a truly modular architectural design. Our work can be seen as a step toward the definition of a standard framework for the simulation of cross-layer, multi-technology and mobile systems in ns2.

No abstract

No abstract