Eigenfilter -based Automatic Modulation Classification with Offsets for Distributed Antenna Systems

Automatic modulation classification (AMC) schemes are crucial for cognitive radio applications in 5G systems as well as for military scenarios. Current AMC schemes do not exploit the redundant information of distributed antenna systems (DAS). Moreover, in case of DAS, an offset should be taken into account. In this paper, we propose an eigenfilter and higher- order cumulants based AMC scheme with offsets. By including different offsets during the training phase of the artificial neural network (ANN), we can exploit all the cumulants achieving a very high accuracy. Moreover, by combining the captured signals by the DAS with the eigenfilter, we can successfully exploit the array gain.