On Privacy of Quantized Sensor Measurements through Additive Noise

TL;DR

This paper investigates how adding carefully designed random noise to quantized sensor data can enhance privacy by reducing information leakage to eavesdroppers, while maintaining acceptable data distortion levels.

Contribution

It introduces a method for designing noise distributions that minimize mutual information between noisy measurements and original data under distortion constraints.

Findings

The proposed noise addition effectively reduces mutual information.

Simulations demonstrate the privacy-utility trade-off.

The method maintains data utility while enhancing privacy.

Abstract

We study the problem of maximizing privacy of quantized sensor measurements by adding random variables. In particular, we consider the setting where information about the state of a process is obtained using noisy sensor measurements. This information is quantized and sent to a remote station through an unsecured communication network. It is desired to keep the state of the process private; however, because the network is not secure, adversaries might have access to sensor information, which could be used to estimate the process state. To avoid an accurate state estimation, we add random numbers to the quantized sensor measurements and send the sum to the remote station instead. The distribution of these random variables is designed to minimize the mutual information between the sum and the quantized sensor measurements for a desired level of distortion -- how different the sum and the quantized sensor measurements are allowed to be. Simulations are presented to illustrate our results.