Abstract:Location data is collected from users continuously to understand their mobility patterns. Releasing the user trajectories may compromise user privacy. Therefore, the general practice is to release aggregated location datasets. However, private information may still be inferred from an aggregated version of location trajectories. Differential privacy (DP) protects the query output against inference attacks regardless of background knowledge. This paper presents a differential privacy-based privacy model that protects the user's origins and destinations from being inferred from aggregated mobility datasets. This is achieved by injecting Planar Laplace noise to the user origin and destination GPS points. The noisy GPS points are then transformed into a link representation using a link-matching algorithm. Finally, the link trajectories form an aggregated mobility network. The injected noise level is selected using the Sparse Vector Mechanism. This DP selection mechanism considers the link density of the location and the functional category of the localized links. Compared to the different baseline models, including a k-anonymity method, our differential privacy-based aggregation model offers query responses that are close to the raw data in terms of aggregate statistics at both the network and trajectory-levels with maximum 9% deviation from the baseline in terms of network length.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to balance privacy protection and data utility in aggregated mobile networks. Specifically, the paper focuses on how to prevent the release of users' mobile trajectory data from leaking users' sensitive information, such as home addresses, workplaces, etc., while maintaining the usefulness of the data so that urban planners and researchers can use these data for effective analysis and decision - making. ### Background of the Paper With the popularization of smart phones, GPS devices and connected vehicles, geospatial - temporal data sets are gradually increasing, and these data sets record the movement of devices. However, these data sets contain details of the movement patterns of device owners, which raises privacy issues. Therefore, organizations usually take anonymization processing or only release aggregated data sets when releasing these data. Nevertheless, even in the aggregated data sets, personal privacy information may still be inferred. For example, by analyzing frequently repeated trajectories, unique movement patterns can be identified and thus associated with specific individuals. ### Solution The paper proposes a method based on Differential Privacy (DP), which protects privacy by injecting planar Laplace noise into the users' starting and ending GPS points. The specific steps are as follows: 1. **Noise Injection**: Select an appropriate noise level according to the density of the road network and perturb the users' starting and ending GPS points. 2. **Link Matching**: Use a link - matching algorithm to convert the perturbed GPS points into a link representation. 3. **Aggregate Network Generation**: Finally, these linked trajectories form an aggregated mobile network. ### Key Technologies - **Differential Privacy**: Differential privacy is a statistical privacy - protection technique, aiming to minimize the leakage of personal information while retaining the characteristic patterns of data. By controlling the noise injection level, differential privacy can ensure that attackers cannot extract private information from the data set even with background knowledge. - **Adaptive Noise Injection**: Use the Sparse Vector Mechanism (SVM) to select adaptive noise range parameters. This method takes into account the link density of the location and the functional category of the local link to ensure an appropriate noise level in different areas. - **Link - Matching Algorithm**: Match the perturbed GPS points with new links to ensure the geospatial constraints of the data. ### Experimental Results The paper verifies the effectiveness of this method through experiments. Compared with baseline models (such as the k - anonymization method), the proposed Differential Privacy Adaptive Noise Injection (DP - ANI) model performs excellently in maintaining the accuracy of aggregated statistical data, with a maximum deviation of the network length of only 9%. ### Main Contributions 1. Propose a Differential Privacy - based Adaptive Noise Injection (DP - ANI) model, which perturbs the starting and ending GPS points in an adaptive manner according to the density of the road network. 2. Use the Sparse Vector Technique (SVM) to select adaptive range parameters to ensure the privacy of noise injection. 3. Evaluate the impact of the noise injection model on the aggregated mobile network and verify its ability to protect privacy while maintaining data utility. ### Conclusion The paper proposes an effective method that preserves the data utility of the aggregated mobile network while protecting users' privacy. This method provides important support for urban planning and traffic research and helps promote the formulation and implementation of relevant policies.

Differential Privacy in Aggregated Mobility Networks: Balancing Privacy and Utility

An Efficient and Differential Privacy-Based Scheme for Aggregating Mobility Datasets

Novel trajectory privacy-preserving method based on clustering using differential privacy

Real-time trajectory privacy protection based on improved differential privacy method and deep learning model

Real-World Trajectory Sharing with Local Differential Privacy

OPTDP: Towards optimal personalized trajectory differential privacy for trajectory data publishing

Spatio-temporal Trajectory Dataset Privacy Based on Network Traffic Control

Protecting Personalized Trajectory with Differential Privacy under Temporal Correlations

Privacy Preservation for Trajectory Publication Based on Differential Privacy

Differential Privacy Location Protection Method Based on the Markov Model

Frequency-based Randomization for Guaranteeing Differential Privacy in Spatial Trajectories

DP-LTOD: Differential Privacy Latent Trajectory Community Discovering Services over Location-Based Social Networks

May the privacy be with us: Correlated differential privacy in location data for ITS

Personalized Differential Privacy Preservation Method for Trajectory Based on Regional Density Analysis

Effective Route Recommendation Leveraging Differentially Private Location Data

Anonymization and De-Anonymization of Mobility Trajectories: Dissecting the Gaps Between Theory and Practice

What Does The Crowd Say About You? Evaluating Aggregation-based Location Privacy

Releasing Large-Scale Human Mobility Histograms with Differential Privacy

An Efficient Differential Privacy-Based Method for Location Privacy Protection in Location-Based Services

Travel Trajectory Frequent Pattern Mining Based on Differential Privacy Protection

Optimization of Privacy Budget Allocation In Differential Privacy-Based Public Transit Trajectory Data Publishing for Smart Mobility Applications