Abstract
The use of hydrodynamic modeling to optimize the placement of sensor nodes with respect to reduction of uncertainty in sea current modeling is demonstrated. This is done under the additional constraint that the nodes should constitute a connected underwater acoustic communication network. For the connectivity analysis a sound propagation model is employed, based on detailed sound speed data produced by the hydrodynamic model. It is found that such a detailed sound speed field is very attractive for obtaining sufficient precision in acoustic modeling. This is particularly important in regions of significant variability in bathymetry and sea currents. In such regions it would be very demanding to obtain sufficiently detailed data by physical measurement, and the use of a hydrodynamic model is virtually the only alternative. In general the model based approach is important for up-scaling the geographical extent of a network, and has a large impact on the economy of its deployment and operation.