提出了一种基于概率模型的预测性时空区域查询处理方法.该方法采用Filter-Refinement方式来处理查询.首先,从数据库中选择所有可能满足查询的候选移动对象;然后,根据概率模型中定义的方法来计算候选移动对象满足查询的概率;最后,根据查...提出了一种基于概率模型的预测性时空区域查询处理方法.该方法采用Filter-Refinement方式来处理查询.首先,从数据库中选择所有可能满足查询的候选移动对象;然后,根据概率模型中定义的方法来计算候选移动对象满足查询的概率;最后,根据查询中指定的最小概率阈值过滤候选移动对象并返回查询结果.该概率模型将移动对象未来可能出现的位置定义为一个随机变量,并给出了计算移动对象在两种不同的运动模式下满足查询的概率值的方法.还提出了一种通过对大量历史轨迹抽样来获得概率密度函数(probability density function,简称PDF)的轨迹分析算法,并设计了概率密度函数索引STP-Index(spatio-temporal PDF-index).该索引能够有效地提高轨迹分析算法和概率计算的效率.实验结果表明,该查询处理方法能够有效地支持预测性时空区域查询的处理,提高查询结果的正确性,特别适合于具有较小的空间区域和长时间范围的预测性时空区域查询.展开更多
With the advances in mobile computing and mobile communication technology, there comes a kind of novel applications in which the locations of moving objects are maintained and processed. In existing literatures, a dat...With the advances in mobile computing and mobile communication technology, there comes a kind of novel applications in which the locations of moving objects are maintained and processed. In existing literatures, a data model called moving objects sptio-temporal (MOST) is proposed and a new location record is generated when the distance between the actual location and the database location of a moving object exceeds a pre-defined distance threshold. In a mobile computing environment, a user can issue location-dependent continuous queries (LDCQs). To cater for the large number of moving objects in the system, this paper first gives a hierarchical distributed location database model to store the locations of moving objects. Based on the distribution of the location databases for different moving objects, this paper then proposes a method to determine the processing site for a location-dependent query. When a LDCQ is processed, a set of tuples <O, begin, end> Is provided indicating that object O satisfies the condition presented in the LDCQ from time begin to end. In the existing literatures, when there is a location update generation, the related LDCQ is re-processed and the answering tuples are re-transmitted via the wireless channel. This location-update-based LDCQ processing method has its disadvantages: it has much CPU calculation cost and imposes a high overhead in the wireless bandwidth which is very undesirable In a wirelss environment. Based on the maximal speed of a moving object, this paper presents a deferred LDCQ evaluation strategy.展开更多
文摘提出了一种基于概率模型的预测性时空区域查询处理方法.该方法采用Filter-Refinement方式来处理查询.首先,从数据库中选择所有可能满足查询的候选移动对象;然后,根据概率模型中定义的方法来计算候选移动对象满足查询的概率;最后,根据查询中指定的最小概率阈值过滤候选移动对象并返回查询结果.该概率模型将移动对象未来可能出现的位置定义为一个随机变量,并给出了计算移动对象在两种不同的运动模式下满足查询的概率值的方法.还提出了一种通过对大量历史轨迹抽样来获得概率密度函数(probability density function,简称PDF)的轨迹分析算法,并设计了概率密度函数索引STP-Index(spatio-temporal PDF-index).该索引能够有效地提高轨迹分析算法和概率计算的效率.实验结果表明,该查询处理方法能够有效地支持预测性时空区域查询的处理,提高查询结果的正确性,特别适合于具有较小的空间区域和长时间范围的预测性时空区域查询.
文摘With the advances in mobile computing and mobile communication technology, there comes a kind of novel applications in which the locations of moving objects are maintained and processed. In existing literatures, a data model called moving objects sptio-temporal (MOST) is proposed and a new location record is generated when the distance between the actual location and the database location of a moving object exceeds a pre-defined distance threshold. In a mobile computing environment, a user can issue location-dependent continuous queries (LDCQs). To cater for the large number of moving objects in the system, this paper first gives a hierarchical distributed location database model to store the locations of moving objects. Based on the distribution of the location databases for different moving objects, this paper then proposes a method to determine the processing site for a location-dependent query. When a LDCQ is processed, a set of tuples <O, begin, end> Is provided indicating that object O satisfies the condition presented in the LDCQ from time begin to end. In the existing literatures, when there is a location update generation, the related LDCQ is re-processed and the answering tuples are re-transmitted via the wireless channel. This location-update-based LDCQ processing method has its disadvantages: it has much CPU calculation cost and imposes a high overhead in the wireless bandwidth which is very undesirable In a wirelss environment. Based on the maximal speed of a moving object, this paper presents a deferred LDCQ evaluation strategy.
