generate modulation signal samples, whose SNR ranges from -10dB to 10dB
FL-AMC
FedeCNN_xx_Balance is based on class balance trick and here, we use resampling the samples of the class with fewer samples instead of balanced cross entropy (convenient to implement, the two are basically equivalent)
Y. Wang, G. Gui, H. Gacanin, B. Adebisi, H. Sari and F. Adachi, "Federated Learning for Automatic Modulation Classification Under Class Imbalance and Varying Noise Condition," in IEEE Transactions on Cognitive Communications and Networking, vol. 8, no. 1, pp. 86-96, March 2022, doi: 10.1109/TCCN.2021.3089738.