Privacy-Preserving Federated Learning based on Differential Privacy and Momentum Gradient Descent | IEEE Conference Publication | IEEE Xplore