Abstract
Modulation recognition plays a critical role in the development of reconfigurable radios which intelligently interact with their environments to maximize resources utilization. This research topic has been intensively investigated in the literature for different communication systems during the last decades. However, to the best of the authors' knowledge, no work has been reported to take advantage of channel decoders, which are typically implemented in most practical systems, in designing modulation recognition algorithms. In this contribution, for the first time in the literature, we show how the output of a channel decoder can be exploited to iteratively estimate the maximum-likelihood value of the transmit modulation format. The proposed algorithm is implemented via an expectation-maximization procedure. The proposed algorithm operates with any channel code as long as its decoder can provide the a posteriori probabilities of the transmitted bits. Monte Carlo simulations indicate that the proposed algorithm achieves a high recognition performance over a wide range of signal-to-noise ratios compared to the traditional ones.