Understanding interference by cognitive nodes: The underwater case

In this paper we investigate how to rephrase the concepts of cognitive access, originally developed for radio communications, in the framework of underwater acoustic communications. Differently from classical cognitive radio access, where each signal at the receiver is generated by a communication source, in the underwater scenario acoustic signals at the general receive node are likely to be due to communication source as well as natural/artificial acoustic sources (e.g mammals, ship engines and so forth). In order to maximize the access probability for cognitive acoustic nodes, we then focus on understanding the nature of the sensed interference via the tools of image classification, being the image the spectrogram of the acquired signal.