Abstract
Conference Title: 2018 IEEE Wireless Communications and Networking Conference (WCNC) Conference Start Date: 2018, April 15 Conference End Date: 2018, April 18 Conference Location: Barcelona, Spain Recently, modulation recognition algorithms have received a large amount of attention in both industry and academia. In addition to their use in the military field, these algorithms found commercial application in many reconfigurable systems, such as cognitive radios. Most previously existing algorithms are focused on recognition of modulations for noncooperative single-input single-output (SISO) and multiple-input multiple-output (MIMO) systems. In this paper, we propose a recognition algorithm for MIMO multi-relay networks under spatially-correlated channels. The proposed algorithm contains two stages. The first stage employs the higher order statistics based features in conjunction with the principal component analysis as a features selection method, while the second stage uses AdaBoost as a classifier. Simulations are provided to evaluate the accuracy of the proposed algorithm using the average probability of correct recognition for modulation schemes. It is shown that the algorithm has the ability to provide a good recognition performance at acceptable signal-to-noise values.