Differentially Private GANs for Generating Synthetic Indoor Location Data

TL;DR

Indoor location services raise privacy concerns due to sensitive movement data. The paper proposes a privacy-preserving framework based on differentially private GANs (DPWGAN and DPCGAN) to generate synthetic indoor location data that retain utility for fingerprinting localization. Empirical results on a real CRI dataset show that DPWGAN can maintain localization accuracy in location-based tasks under reasonable privacy budgets, while DPCGAN offers stronger privacy gains in zone-based localization, albeit with some utility trade-offs. The work demonstrates how synthetic, DP-protected data can support privacy-aware indoor localization training and evaluation, with practical guidance on selecting DPGAN variants for different localization modes. The findings highlight the potential of DP-based synthetic data to enable privacy-preserving yet effective indoor LBS deployments.

Abstract

The advent of location-based services has led to the widespread adoption of indoor localization systems, which enable location tracking of individuals within enclosed spaces such as buildings. While these systems provide numerous benefits such as improved security and personalized services, they also raise concerns regarding privacy violations. As such, there is a growing need for privacy-preserving solutions that can protect users' sensitive location information while still enabling the functionality of indoor localization systems. In recent years, Differentially Private Generative Adversarial Networks (DPGANs) have emerged as a powerful methodology that aims to protect the privacy of individual data points while generating realistic synthetic data similar to original data. DPGANs combine the power of generative adversarial networks (GANs) with the privacy-preserving technique of differential privacy (DP). In this paper, we introduce an indoor localization framework employing DPGANs in order to generate privacy-preserving indoor location data. We evaluate the performance of our framework on a real-world indoor localization dataset and demonstrate its effectiveness in preserving privacy while maintaining the accuracy of the localization system.

Figures (10)

• Figure 1: The general overview of indoor location fingerprinting learning and localization processes.
• Figure 2: The schematics of a GAN
• Figure 3: The proposed indoor location DPGAN framework.
• Figure 4: The floor plan of the CRI testbed.
• Figure 5: Correlation results for location-based DPWGAN