The recent years increasing interest for research on underwater communication and networking has sparked the interest for a software defined modem (SDM). This concept has been realized on a sophisticated hardware platform, which is provided in form of the commercially available develogic Hydro-Acoustic Modem HAM.Base. This is a small sized, energy efficient, and deep water approved underwater communication modem. It provides widespread flexibility especially in the physical layer, but also broad accessibility to higher layers. Such an SDM is especially suitable for scientific research in underwater communication networks, where the implemented software architecture may be a result and not a prerequisite. This paper is about the tools, on how, and to what extent reprogramming and reconfiguration of this SDM are performable. As an example it is shown how this SDM is programmed to be used in the framework of TUWACS (Transient Underwater Acoustic Communication System) used in the robust acoustic communications in underwater networks (RACUN) EDA project.
Security in underwater networks is a crucial aspect to maintain correct network operations, but has only partially been investigated so far. Most of the applications of an underwater network are related to military or public safety scenarios, which should exhibit a high robustness to attacks and failures in order to avoid disastrous consequences even with simple Denial-of-Service attacks. The defense mechanisms and countermeasures are usually tailored to specific types of attacks. Albeit this procedure allows to obtain very effective defense mechanisms, it requires the development of different countermeasures for each possible attack. Another possible solution is to use reputation systems to identify the attackers; hence, in this paper we design a trust model able to tackle a wide range of attacks. The idea of the trust model is to observe the behavior of the nodes overhearing the neighbors’ transmissions and use this information in a subjective logic framework to assign a trust metric to each of the neighbors. In addition, in order to take into account the unique characteristics of the underwater acoustic channel, which alternates periods of high packet loss to periods with low errors, we include the channel state in the trust model to avoid to erroneously mark as not trustworthy nodes that correctly forwarded the packets, but could not be overheard due to an unreliable channel.