Research On Privacy Protection Method Of Trajectory Based On (k,l)-Anonymity

In recent years,with the development of mobile intelligent terminals and wireless positioning technologies,location service providers have collected a large amount of trajectory data while providing location-based services to mobile objects.The trajectory data contains rich spatio-temporal information.The analysis and mining of trajectory data have important value in scientific research,economic construction and social development.However,if the location service provider is not trusted,it may lead to the disclosure of personal information such as personal habits,health conditions and hobbies.Therefore,the issue of trajectory privacy protection has received more and more attention from people,and it is also a research hotspot in recent years.In the trajectory privacy protection method of data release,k-anonymous model and ldiversity model have become an important research method in the field of data release trajectory privacy protection because of their high privacy protection level and good algorithm portability.The key of the l-diversity model lies in the similarity between the trajectories.The similarity of the trajectory spatial distribution is an important aspect of trajectory similarity.However,many current metrics for the similarity of trajectory spatial distribution can not reasonably reflect the spatial distribution characteristics of trajectory,which leads to the low level of privacy protection of the algorithm.This paper proposes a method to measure the similarity of the trajectory spatial distribution by using the trajectory fitting rate,that is,the proportion of the pair of sampling points whose distance is less than a given threshold is used to measure the degree of fit between the trajectories.Based on this,a(k,l)-anonymity model based on trajectory fit ratio for trajectory data release is proposed,and the formal definition of(k,l)-anonymous model is given from the perspective of set.The key of this model is constructs a fully dissimilar subset of an anonymous set,that is,a subset of trajectory that are not similar between the two in an anonymous set.Based on the model,the KLTPP algorithm for privacy protection of data release trajectory is proposed.This algorithm constructs(k,l)anonymous sets by clustering method based on greedy strategy.By trajectory expansion in the equivalence class division stage,merging equivalence classes can avoid Loss of information caused by deleting sample points.This paper defines two main indicators for evaluating the performance of the algorithm,namely privacy disclosure risk and information distortion.The effects of changes in the internal parameters of the model on the performance of the algorithm are measured experimentally,and the performance of KLTPP and NWA,GTT in terms of privacy disclosure risk and information distortion are compared.The experimental results show that the(k,l)-anonymity model based on the trajectory fitting rate and its implementation algorithm KLTPP can better resist the trajectory spatial distribution similarity attack,and achieve a good balance between privacy protection and data availability.