The capability of performing a reliable communication is an essential asset to any underwater mission. Underwater acoustic communication is a technology that enables the information exchange in a swarm of AUVs. This shared data can also be used for improving the AUV localization accuracy for navigation purposes. This work, which is part of COMET and NEMOSENS projects, aims to improve the communication performance among these underwater vehicles. The overall project objective is to develop a swarm of AUVs, that can be used in distinct underwater missions. The swarm of AUVs composes a network, which has no central hierarchy. As each vehicle broadcasts periodically the information about its location and velocity, the other AUVs can leverage this information for adapting its own transmission parameters. In this work, we take advantage of this data for achieving a better usage of the time-frequency resources. In our proposition, we optimize some physical layer parameters using only the knowledge of the channel geometry. We design a method for optimizing the number of subcarriers and the cyclic-prefix length of the Orthogonal Frequency-Division Multiplexing (OFDM) system. In this method, we seek to minimize the interblock interference (IBI) and the intercarrier interference (ICI) expected at the receiver side. The obtained simulation results show a performance gain in terms of bit error rate (BER) when adopting our parameters optimization approach.