Privacy Preserving of Digital Data Using Complex Non Linear Adaptive Filtering
In the Field of Signal Processing in Encrypted Domain (SPED) has emerged in order to provide efficient and secure solutions for preserving privacy of signals that are processed by untrusted agents. In this work, privacy problem of adaptive filtering is one of most important and ubiquitous blocks in signal processing today. It present several use cases for adaptive signal processing, studying the privacy characteristics, constraints, and requirements, which differ in several aspects from those of already tackled linear filtering and classification problems. It shows the impossibility of using a strategy based solely on current homomorphic encryption systems. Fast protocols that already have used are almost as robust as original LMS algorithm with respect to quantization errors, while presenting the low computational and communication complexity.