Map-based automatic modulation classification for wireless adaptive OFDM systems

This manuscript addresses the maximum-a-posteriori-(MAP)-based automatic modulation classification (AMC) for wireless orthogonal frequency division multiplexing (OFDM) systems with adaptive coding and modulation (ACM). The proposed classifier for quadrature amplitude modulation (QAM) schemes which utilizes the channel reciprocity in time-division duplex (TDD) systems requires the knowledge about the joint probabilities of the subcarrier-wise bit efficiencies at the transmitter and receiver side. In contrast to prior heuristic approaches, these probabilities are calculated analytically (with some approximations) if the transmitter and receiver apply the same bit loading algorithm on their erroneously estimated channel state information. Numerical results reveal that the precise knowledge of the joint probabilities improves the reliability of the automatic modulation classifier significantly, especially for high signal-to-noise power ratios (SNR).