Fuhrmeister, Kay ChristopherKay ChristopherFuhrmeisterCui, HaoHaoCuiYaroshchuk, ArtemArtemYaroshchukKöllmer, ThomasThomasKöllmer2024-12-052024-12-052024-09-30https://publica.fraunhofer.de/handle/publica/47969810.1109/IS262782.2024.10704107In this paper, we introduce FLCrypt 1 1, a library designed to enhance Federated Learning with additional privacy guarantees by applying Fully Homomorphic Encryption to the model aggregation stage, thereby preventing the aggregator from accessing the unencrypted model parameters of the training participants. We evaluate our approach by comparing its accuracy against an unencrypted baseline using an audio classification task aimed at distinguishing metal balls with different surfaces, which serves as a proxy for fault detection in industrial sound analysis. Our findings indicate a marginal decrease in accuracy due to applying Fully Homomorphic Encryption, alongside a significant increase in both runtime and memory demands. Our analysis also concludes that runtime increases linearly with the number of model parameters. Our results lead us to affirm the viability of FLCrypt for Federated Learning applications in acoustic sensor networks with elevated security requirements such as sound classification.enTrainingAnalytical modelsPrivacyRuntimeAccuracyFederated learningScalabilityMetalsSide-channel attacksHomomorphic encryptionFederated LearningSecurityPrivacyAudio Event DetectionAcoustic Sensor NetworksTrustworthy AIconference paper