Differential Privacy Trajectory Protection Method Based on Spatiotemporal Correlation

• 作者: Jian Liu Kangkang Dou
• 来源期刊: 国际计算机前沿大会会议论文集
• 学科: 工学
• 关键词: Location-Based Services Trajectory privacy protection Spatiotemporal correlation Cross-correlation constraints Noise candidate set
• 年，卷（期）: 2020,（2）
• 页码范围: 151-166
• 页数: 16页
• 分类号: TN9

摘要

Location-based services provide service and convenience,while causing the leakage of track privacy.The existing trajectory privacy protection methods lack the consideration of the correlation between the noise sequence,the user’s original trajectory sequence,and the published trajectory sequence.And they are susceptible to noise filtering attacks using filtering methods.In view of this problem,a differential privacy trajectory protection method based on spatiotemporal correlation is proposed in this paper.With this method,the concept of correlation function was introduced to establish the correlation constraint of release track sequence,and the least square method was used to fit the user’s original track and the overall direction of noise sequence to construct noise candidate set.It ensured that the added noise sequence has spatiotemporal correlation with the user’s original track sequence and release track sequence.Also,it effectively resists attackers’denoising attacks,and reduces the risk of trajectory privacy leakage.Finally,comparative experiments were carried out on the real data sets.The experimental results show that this method effectively improves the privacy protection effect and the data availability of the release track,and it also has better practicability.