不确定环境下轨迹k-匿名隐私保护

综合考虑不确定环境下移动对象的影响因素,提出了对船舶自动导航系统所记录的轨迹进行k-匿名的隐私保护方法。该方法首先建立一个不确定空间索引,并将不确定空间用四叉树结构存储,然后通过使用连续近邻查询方法找出与当前轨迹相似区域的轨迹,并将这些轨迹加入匿名候选集。由于考虑到路网规模影响匿名信息的有效性以及攻击者对轨迹的攻击概率,采用启发式算法生成匿名轨迹最佳利用链,从而加大对轨迹隐私保护的力度。最后实验结果表明,所提出的方法比传统方法信息丢失率降低了20%~50%,信息扭曲度随着查询范围的增大能保持在50%以下,代价损失相比传统方法降低了10%~30%。该方法可以有效防止恶意者对轨迹进行攻击获取信息,应用于公务船艇海上执勤执法。

K-anonymity privacy-preserving for trajectory in uncertain environment

To comprehensively consider the factors influencing the moving objects in uncertain environment, a k-anonymity privacy-preserving method for the trajectory recorded by automatic identification system was presented. Firstly, an uncertain spatial index model was established which was stored in grid quadtree. Then the continuous k-Nearest Neighbor (KNN) query method was used to find the trajectory which had the similar area to the current trajectory, and the trajectory was added to the anonymous candidate set. By considering the network scale influence on the effectiveness of the anonymous information and the probability of attacker's attack on trajectory, the optimal exploit chain of trajectory was generated by using the heuristic algorithm to strengthen the trajectory privacy-preserving. Finally, the experimental results show that, compared with the traditional method, the proposed method can decrease the information loss by 20% to 50%,while the information distortion can maintain below 50% with the enlarge of query range and the cost loss is cut down by 10% to 30%.The proposed method can effectively prevent malicious attackers from the information access of trajectory,and can be applied for the official boat to law enforcement at sea.