Use a convolutional neural network (CNN) for modulation classification. You generate synthetic, channel-impaired waveforms. Using the generated waveforms as training data, you train a CNN for modulation classification. You then test the CNN with software-defined radio (SDR) hardware and over-the-air signals.

Transmit and receive data using a nonrectangular 16-ary constellation in the presence of Gaussian noise. Show the scatter plot of the noisy constellation and estimate the symbol error rate (SER) for two different signal-to-noise ratios.

Construct an orthogonal frequency division modulation (OFDM) modulator/demodulator pair and to specify their pilot indices. The OFDM modulator System object enables you to specify pilot subcarrier indices consistent with the constraints described in comm.OFDMModulator.info. In this example, for OFDM transmission over a 3x2 channel, pilot indices are created for each of the three transmit antennas. Additionally, the pilot indices differ between odd and even symbols.