An Efficient Privacy-Preserving Algorithm based on Randomized Response in IoT-based Smart Grid

In this paper, we propose a new randomized response algorithm that can achieve differential-privacy and utility guarantees for consumer's behaviors, and process a batch of data at each time. Firstly, differing from traditional differential private approaches, we add randomized response noise into the behavior signatures matrix to achieve an acceptable utility-privacy tradeoff. Secondly, a behavior signature modeling method based on sparse coding is proposed. After some lightweight trainings using the energy consumption data, the dictionary will be associated with the behavior characteristics of the electric appliances. At last, through the experimental results verification, we find that our Algorithm can preserve consumer's privacy without comprising utility.