室内移动对象的数据管理

发布时间：2018-08-15 12:24

【摘要】： 现代室内空间可以容纳大量的移动对象。例如,人们在日常生活中通常会花费大量的时间活动在诸如办公楼、购物中心、会展中心、机场及地铁等交通基础设施在内的各种室内空间。随着各种室内定位技术的发展,这些室内移动对象的位置可以被确定和记录下来。对室内移动对象的数据管理,可以作为一系列室内位置服务的基础,例如室内导航,员工安全,室内空间规划,商铺促销,广告竞价等。因此,有效的管理室内移动对象,具有重要的应用价值。虽然当前对室外移动对象的数据管理问题已经有了较充分的研究,然而,这些技术并不能直接应用于室内移动对象的管理,主要原因有如下两点：首先,复杂的室内拓扑结构使得广泛应用于室外空间的距离模型,轨迹表达等都不再适用于室内空间。其次,室内的定位技术通常不能像广泛应用于室外空间的GPS定位技术那样,连续不断的报告室内移动对象的位置,从而带来较大程度的位置不确定性。本文首先对室外移动对象的管理技术进行了综述,指出其直接应用在室内环境下的不足。针对室内空间的特点,提出了基于图的室内空间建模方法、室内移动对象的跟踪方法,以及室内时空范围查询、连续范围查询和概率阈值k近邻查询等多类查询处理的新算法及相关索引结构。本文的主要贡献如下： 1.提出了基于图模型的室内空间建模方法。基础图用于对室内空间拓扑信息建模,而部署图则可以有效的管理符号化定位设备,并做为室内移动对象数据管理的基础。根据部署图,对室内移动对象的状态进行了划分。 2.基于部署图模型,分别提出了室内移动对象的离线跟踪方法和在线跟踪方法。 3.提出了一种基于符号空间的室内移动对象历史轨迹表示方法,并提出了新型索引结构RTR-tree和TP2R-tree对该类型历史轨迹进行索引,以支持室内时空范围查询和室内逻辑查询。 4.根据室内移动对象所处的状态,提出了基于哈希的索引结构用于索引室内移动对象的当前位置。并在此基础上提出一种查询感知的、增量的连续范围查找算法。 5.形式化的分析了室内移动对象的不确定性。提出了一种有效的概率阈值k最近邻查询处理算法。本文对室内移动对象的数据管理进行了系统研究。针对室内空间和符号化定位技术的特点,提出了基于图模型的室内移动对象管理基础;并且基于该图模型,对室内时空范围查询,连续范围查询和概率阈值k最近邻查询的处理方法做了充分研究。这些技术可以作为今后室内移动对象管理研究的基础,并且可以为实际应用中的室内位置服务提供技术保障。
[Abstract]:Modern indoor space can accommodate a large number of moving objects. For example, people usually spend a lot of time in indoor space such as office building, shopping center, convention and exhibition center, airport and subway and so on. With the development of various indoor positioning technologies, the location of these indoor moving objects can be determined and recorded. The data management of indoor moving objects can be used as the basis of a series of indoor location services, such as indoor navigation, employee safety, indoor space planning, shop promotion, advertising bidding and so on. Therefore, the effective management of indoor moving objects, has an important application value. Although the current data management of outdoor moving objects has been fully studied, these technologies can not be directly applied to the management of indoor moving objects. The main reasons are as follows: first of all, Because of the complex indoor topology, the distance model, trajectory representation and so on, which are widely used in outdoor space, are no longer suitable for indoor space. Secondly, indoor positioning technology usually can not report the location of moving objects continuously as the GPS positioning technology is widely used in outdoor space, which brings a large degree of location uncertainty. In this paper, the management technology of outdoor moving objects is reviewed, and the shortcomings of its direct application in indoor environment are pointed out. According to the characteristics of indoor space, the paper puts forward the modeling method of indoor space based on graph, the tracking method of indoor moving object, and the query of indoor space-time range. A new algorithm for multi-class query processing, such as continuous range query and probabilistic threshold k-nearest neighbor query, and related index structure. The main contributions of this paper are as follows: 1. A method of indoor space modeling based on graph model is proposed. The basic map is used to model the topological information of indoor space, and the deployment plan can effectively manage the symbolic positioning equipment and serve as the basis for the data management of indoor moving objects. According to the deployment diagram, the state of indoor moving objects is divided. 2. 2. Based on the deployment diagram model, an off-line tracking method and an on-line tracking method for indoor moving objects are proposed, respectively. In this paper, a method of representing the historical track of indoor moving object based on symbol space is proposed, and a new index structure, RTR-tree and TP2R-tree, is proposed to index the historical track of this type. To support indoor space-time range query and indoor logic query. 4. According to the state of indoor moving objects, a hash-based index structure is proposed to index the current position of indoor moving objects. On this basis, a query aware, incremental continuous range lookup algorithm is proposed. 5. The uncertainty of indoor moving object is analyzed formally. An effective probability threshold k nearest neighbor query processing algorithm is proposed. In this paper, the data management of indoor moving objects is studied systematically. According to the characteristics of indoor space and symbolic positioning technology, the paper puts forward the management foundation of indoor moving objects based on graph model, and queries the scope of indoor space and time based on the graph model. The processing methods of continuous range query and probability threshold k nearest neighbor query are studied. These technologies can be used as the basis of the research on indoor moving object management in the future, and can provide technical support for indoor location service in practical application.
【学位授予单位】：复旦大学
【学位级别】：博士
【学位授予年份】：2010
【分类号】：TP274

【引证文献】