Multimodal underwater networks

The recent level of maturity reached by broadband underwater non-acoustic communication technologies paves the way to the development of new applications, such as wireless remote control for underwater vehicles and the possibility to retrieve a massive quantity of data from underwater sensor networks. Indeed, an optical link can support the transmission of high-traffic demanding data (e.g., video streams) in real time, but its reach can hardly exceed 100 meters. Also radio frequency electromagnetic communications can provide a high transmission rate, however, in salty sea-waters their maximum range is less than 7 meters. Therefore, when considering either optical or radio frequency communications, a low-rate long-range acoustic link still has to be employed. Although this backup link cannot be used to transmit hight data traffic, it can still keep the minimal quality of service needed to monitor the status of the underwater network. This thesis presents how optical and acoustic communications can be combined in the so-called multimodal networks, by testing such solutions with DESERT Underwater, a simulation and experimentation framework for underwater networks. Optimal routing and data-link layers for multimodal networks are analyzed, as well as a switching algorithm that decides which technology to employ for transmitting the data packets, depending on the type of data and the channel quality. These protocols are evaluated via both simulation and field experiments.