Encrypted signal processing for privacy protection: conveying the utility of homomorphic encryption and multiparty computation

In recent years, signal processing applications that deal with user-related data have aroused privacy concerns. For instance, face recog- nition and personalized recommendations rely on privacy-sensitive infor- mation that can be abused if the signal processing is executed on remote servers or in the cloud. In this tutorial article, we introduce the fusion of signal processing and cryptography as an emerging paradigm to protect the privacy of users. While service providers cannot access directly the con- tent of the encrypted signals, the data can still be processed in encrypted form to perform the required signal processing task. The solutions for pro- cessing encrypted data are designed using cryptographic primitives like homomorphic cryptosystems and secure multiparty computation. We in- clude four boxes that provide introductory material on these cryptographic primitives. We first introduce encrypted signal processing for privacy pro- tection using a toy example. We then discuss the application of crypto- graphic primitives to typical signal processing operations. Finally, we fo- cus on prototypical solutions of increasing difficulty for privacy-preserving signal processing, namely privacy-preserving face recognition, secure clus- tering and content recommendation.