Sequential Change Detection for Multiple Data Streams with Differential Privacy
It addresses the need for privacy-preserving change detection in multi-stream settings, which is relevant for applications like IoT security where data privacy is critical.
The paper introduces DP-SUM-CUSUM, a differentially private method for detecting synchronized changes in multiple data streams, and provides theoretical bounds on false alarm rate and detection delay, demonstrating the privacy-efficiency tradeoff. Experiments on an IoT botnet dataset validate the approach.
Sequential change-point detection seeks to rapidly identify distributional changes in streaming data while controlling false alarms. Existing multi-stream detection methods typically rely on non-private access to raw observations or intermediate statistics, limiting their usage in privacy-sensitive settings. We study sequential change-point detection for multiple data streams under differential privacy constraints. We consider multiple independent streams undergoing a synchronized change at an unknown time and in an unknown subset of streams, and propose DP-SUM-CUSUM, a differentially private detection procedure based on the summation of per-stream CUSUM statistics with calibrated Laplace noise injection. We show that DP-SUM-CUSUM satisfies sequential $\varepsilon$-differential privacy and derive bounds on the average run length to false alarm and the worst-case average detection delay, explicitly characterizing the privacy--efficiency tradeoff. A truncation-based extension is also presented to handle distributional shifts with unbounded log-likelihood ratios. Simulations and experiments on an Internet of Things (IoT) botnet dataset validate the proposed approach.