Networking Protocols, experimentation, at-sea trials. We are involved in several research activities for underwater mobile networking. There, our focus is centered around the design of networking protocol (channel access, routing, data gathering, etc). Various experimental activities are also on the way, including simulation, lab and field test of network protocols and at-sea testing.
Underwater networking poses serious challenges to network designers, as no straightforward translation exists between protocols for the terrestrial wireless radio environment and their underwater counterparts. The main reason behind this is the different nature of the underwater acoustic channel: long propagation delays (the average propagation speed of an acoustic wave is about 1500 m/s underwater, nearly 200’000 times smaller than radio waves in the air), strong multipath effects, long-term channel variations, and last but not least a much smaller bandwidth available (due to the use of acoustic frequencies, i.e., in the tens of kHz range), which turns into lower transmit bit rates. While radio and optical waves may also be an option for underwater communications, acoustic waves still foster a lot of interest, as they are currently the only means to reach distances longer than a few hundred meters. With the proper configuration of transmission hardware parameters, acoustic waves can travel up to tens of kilometers, making long range communication feasible, albeit at a possibly very low bit rate.
As there is no widely agreed upon model for the channel, we seek new models that are sufficiently compact yet simple enough to be plugged into a network simulator. This activity is carried out in cooperation with institutions and research centers that can provide real data from undersea measurement campaigns.
MAC protocol design and evaluation
We are currently analyzing a number of MAC solutions by means of simulations and stochastic models, in order to discover which features make one protocol perform better than others, with the final objective to create a novel protocol encompassing the best behaviors seen in other approaches.
Routing and broadcasting
Multihop underwater networks will require delay-tolerant routing protocols, that work well in the presence of very long propagation delays. We are currently analyzing the relevant routing tradeoffs that allow, e.g., to save energy by choosing wisely which nodes will relay messages. We are also designing efficient broadcasting techniques based on Hybrid ARQ.
We collaborate with research institutions to define joint sea trial activities which allow for network protocol evaluation as well as channel characterization; this feeds novel insight and provides directions to all above tasks.
DESERT underwater simulator
DESERT Underwater is a complete set of public C++ libraries that extend the NS-MIRACLE simulator to support the design and implementation of underwater network protocols.
- CLAM (CoLlAborative eMbedded networks for submarine surveillance, 2010-2013)
- ONR research grant
- Woods Hole Oceanographic Institution, Woods Hole, MA, USA
- NATO Undersea Research Centre, La Spezia
- Naval Postgraduate School, Monterey, CA, USA
- Italian Ministry of Defense
- University of Twente, Enschede, The Nederlands
- University of Rome “La Sapienza”