随着移动互联网的普及，位置查询已经成为用户享受服务的重要方式.然而在位置查询中，服务器可获取用户位置及轨迹信息这一特点会威胁到用户隐私.为此，提出了一种利用改进的长短期记忆网络来保护用户轨迹隐私的方法.首先可信第三方利用改进的长短期记忆网络预测用户未来时刻的位置点信息，之后第三方在构造匿名区域时，将其预测的位置点信息放入当前时刻进行请求，以打乱轨迹中的时序信息，并模糊服务器端的用户轨迹序列，从而有效保护用户轨迹隐私.同时，将未来时刻点放入当前时刻进行请求这一操作也克服了位置隐私保护中的匿名区域不合理问题，提高了系统服务的有效性.实验结果表明，该方法在用户轨迹的隐私性方面优于现有方法.
With the popularity of the mobile Internet, location query has become an important way for users to enjoy services. However, in the location query, the server can obtain the user location and trajectory information, which threaten the user's privacy. In order to improve the privacy of user trajectory, this paper proposes a method to protect user trajectory privacy by using improved long short-term memory (LSTM) networks. First, the trusted third party uses the improved LSTM network to predict the later position of the user. While constructing the anonymous area, the third party places the predicted location point information into the current moment to make a request. It can disrupt the timing information in the trajectory and obscure the user trajectory sequence on the server side, and therefore protects the privacy of user trajectory. Meanwhile, by placing the future moments into the current moment, the unreasonable problem of the anonymous area in the location privacy protection is overcome. Thus, the effectiveness of the system service is ensured. Experimental results show that the privacy of the user trajectory in this method is better than existing method.
[1] 张建明,赵玉娟,江浩斌,等. 车辆自组网的位置隐私保护技术研究[J]. 通信学报,2012, 33(8):180-189. Zhang J M, Zhao Y J, Jiang H B, et al. Research on protection technology for location privacy in VANET[J]. Journal on Communications, 2012, 33(8):180-189. (in Chinese)
[2] 王璐,孟小峰. 位置大数据隐私保护研究综述[J]. 软件学报,2014, 25(4):693-712. Wang L, Meng X F. Location privacy preservation in big data era:a survey[J]. Journal of Software, 2014, 25(4):693-712. (in Chinese)
[3] 潘晓,肖珍,孟小峰. 位置隐私研究综述[J]. 计算机科学与探索,2007, 1(3):268-281. Pan X, Xiao Z, Meng X F. Survey of location privacy-preserving[J]. Journal of Frontiers of Computer Science and Technology, 2007, 1(3):268-281. (in Chinese)
[4] Hu H B, Xu J L. 2PASS bandwidth-optimized location cloaking for anonymous location based services[J]. IEEE Transactions on Parallel and Distributed Systems, 2010, 21(10):1458-1472.
[5] Ghinita G, Kalnis P, Skiadopoulos S. Mobihide:a mobile peer to peer system for anonymous location-based queries[C]//Proceedings of the 10th International Conference on Advances in Spatial and Temporal Databases (SSTD 2007) Boston, MA, USA, 2007. 221-238.
[6] Palanisamyb, Liu L. MobiMix:protecting location privacy with mix-zones over road networks[C]//IEEE 27th International Conference on Data Engineering (ICDE 2011), 2011:494-505.
[7] Jang M Y, Chang J W. A new cloaking method based on weighted adjacency graph for preserving user location privacy in LBS[J]. Computer Science and Its Applications, 2012, 203:129-138.
[8] Mokbel M F, Chow C Y, Aref W G. The new casper:query processing for location services without compromising privacy[C]//VLDB'06:Proc. International Conference on Very Large Data Bases, New York, USA. 2006:763-774.
[9] Gao S, Ma J, Sun C, Li X. Balancing trajectory privacy and data utility using a personalized anonymization model[J]. Journal of Network and Computer Applications, 2014, 38(1):125-134.
[10] 杨洋,王汝传. 增强现实中基于LBS的矩形区域K-匿名位置隐私保护方法[J]. 南京师大学报(自然科学版),2016, 39(4):44-49. Yang Y, Wang R C. Rectangular region K-anonymity location privacy protection based on LBS in augmented reality[J]. Journal of Nanjing Normal University (Natural Science Edition), 2016, 39(4):44-49. (in Chinese)
[11] Peng T, Liu Q, Meng D, et al. Collaborative trajectory privacy preserving scheme in locationbased services[J]. Information Sciences, 2017, 387:165-179.
[12] 任智慧,徐浩煜,封松林. 基于LSTM网络的序列标注中文分词法[J]. 计算机应用研究,2017, 34(5):1321-1324. Ren Z H, Xu H Y, Feng S L. Sequence labeling Chinese word segmentation method based on LSTM networks[J] Application Research of Computers, 2017, 34(5):1321-1324. (in Chinese)
[13] Park J, Son H, Lee J, et al. Driving assistant companion with voice interface using long short-term memory networks[J]. IEEE Transactions on Industrial Informatics, 2019, 15(1):582-590.
[14] Nammous M K, Saeed K. Natural language processing:speaker, language, and gender identification with LSTM[M]. Advanced Computing and Systems for Security. Springer, Singapore, 2019.
[15] Brinkhoff T. A framework for generating network-based moving objects[J]. GeoInformatica, 2002, 6(2):153-180.
