ShieldFL: Mitigating Model Poisoning Attacks in Privacy-Preserving Federated Learning | IEEE Journals & Magazine | IEEE Xplore