Performance of AdaBoost classifier in recognition of superposed modulations for MIMO TWRC with physical-layer network coding

TLDR: An algorithm for recognition of sources modulations for MIMO TWRC with PLNC is proposed and simulation results show the ability of the proposed algorithm to provide a good recognition performance at acceptable signal-to-noise values.

Abstract: Modulation recognition algorithms have recently received a great deal of attention in academia and industry. In addition to their application in the military field, these algorithms found civilian use in reconfigurable systems, such as cognitive radios. Most previously existing algorithms are focused on recognition of a single modulation. However, a multiple-input multiple-output two-way relaying channel (MIMO TWRC) with physical-layer network coding (PLNC) requires the recognition of the pair of sources modulations from the superposed constellation at the relay. In this paper, we propose an algorithm for recognition of sources modulations for MIMO TWRC with PLNC. The proposed algorithm is divided in two steps. The first step uses the higher order statistics based features in conjunction with genetic algorithm as a features selection method, while the second step employs AdaBoost as a classifier. Simulation results show the ability of the proposed algorithm to provide a good recognition performance at acceptable signal-to-noise values.