Evaluation of features for the automatic recognition of OFDM signals in monitoring or cognitive receivers

TL;DRAbstract

The automatic recognition of signal types is an important task of monitoring receivers and also cognitive receivers. Several modulation recognition or classification procedures exist for single channel signal types while a simple robust procedure for automatic recognition of OFDM signals is lacking because of its numerous frequency channelslying close together. The task considered in this paper is the discrimination between OFDM (or multi-channel) signals and other signal types. The number of frequency channels of the OFDM signals is assumed to be unknown a priori. So, together with the automatic OFDM detection the estimation of the number of frequency channels is treated. Several discrimination features have been examined and the most promising ones are described: measures of the variation, of the skewness, of the kurtosis, and of the specific picket-fence shape of the spectrum which is typical for many OFDM signals. For a number of real-world OFDM samples, recorded from the high freq

Chat with Paper

AI Agents for this Paper