III: Small: Indoor Spatial Query Evaluation and Trajectory Tracking with Bayesian Filtering Techniques

今天，大多数人每天花相当大的一部分时间在室内空间，如办公楼，购物中心，会议中心，地铁系统和许多其他结构。此外，室内空间变得越来越大和复杂。例如，纽约市地铁有469个车站，包含233英里的路线。2014年，地铁系统提供了超过17.5亿次乘坐，平均每天约560万次乘坐。因此，用户将有越来越多的需求发起基于位置的（空间）查询，以寻找朋友，对象，或在室内空间的兴趣点。然而，现有的用于室外环境的空间查询评估技术不能应用于室内空间，因为这些技术假设可以从GPS信号或蜂窝定位获取用户位置，但是该假设在覆盖的室内空间中不成立。此外，室内空间通常与室外空间不同地建模。在室内环境中，用户移动由实体和拓扑（如门、墙和走廊）启用或约束。射频识别（RFID）是一种非常流行的电子标签技术，它允许使用电磁挑战和响应数据交换在一定距离处自动识别物体。基于RFID的系统由大量附着在物体和阅读器上的低成本标签组成，可以通过RF通信在没有直接视线的情况下识别标签。RFID技术在过去十年中变得越来越受欢迎，应用于供应链管理，医疗保健和运输等领域。在这个项目中，研究人员考虑设置一个室内环境，在室内空间部署了一些RFID阅读器。每个用户与RFID标签相关联，当用户在读取器的检测范围内时，读取器可以识别RFID标签。考虑到所有阅读器的RFID原始读数的历史，研究团队能够设计一个系统，可以有效地回答室内空间查询并跟踪物体的轨迹。该项目的研究成果将提高众多高附加值室内应用的性能，从而使我国的经济受益。此外，能够在室内空间定位人员的能力将改善应急反应。该项目将通过让本科生和研究生接触空间数据管理领域的数学和技术基础来促进教学、学习和培训。在该项目中，研究团队将开发一系列技术，从错误的RFID原始读数中获得准确的对象位置，以支持室内空间查询评估和轨迹跟踪。利用精确的空间查询结果和轨迹信息，许多高级应用（例如，室内布局规划和基于室内位置的服务）。该项目将通过试验新颖的室内数据管理技术为研究界做出贡献，这些技术将实现以下目标：（1）开发和比较了许多用于评估室内环境中的空间查询的基于贝叶斯过滤的位置推断方法，（2）针对各种空间查询类型（例如范围查询和k最近邻查询）设计新颖的室内查询评估算法，（3）提出了一种基于隐马尔可夫模型的室内目标轨迹跟踪方法;（4）实现了一个仿真工具包和一个原型系统，其中所有组件将被集成用于性能评估。所有研究成果和出版物都将在项目网站上公布（http：//www.eng.auburn.edu/indoor.htm）。

Today most people spend a significant portion of their time daily in indoor spaces such as office buildings, shopping malls, convention centers, subway systems, and many other structures. In addition, indoor spaces are becoming increasingly large and complex. For instance, the New York City Subway has 469 stations and contains 233 miles of routes. In 2014, the subway system delivered over 1.75 billion rides, averaging approximately 5.6 million daily rides on weekdays. Therefore, users will have more and more demand for launching location-based (spatial) queries for finding friends, objects, or points of interest in indoor spaces. However, existing spatial query evaluation techniques for outdoor environments cannot be applied in indoor spaces because these techniques assume that user locations can be acquired from GPS signals or cellular positioning, but the assumption does not hold in covered indoor spaces. Furthermore, indoor spaces are usually modeled differently from outdoor spaces. In indoor environments, user movements are enabled or constrained by entities and topologies such as doors, walls, and hallways. Radio Frequency Identification (RFID) is a very popular electronic tagging technology that allows objects to be automatically identified at a distance using an electromagnetic challenge-and-response exchange of data. An RFID-based system consists of a large number of low-cost tags that are attached to objects and readers, which can identify tags without a direct line-of-sight through RF communications. RFID technologies have become increasingly popular over the last decade with applications in areas such as supply chain management, health care, and transportation. In this project, the researchers consider the setting of an indoor environment where a number of RFID readers are deployed in the indoor space. Each user is associated with an RFID tag, which can be identified by a reader when the user is within the detection range of the reader. Given the history of RFID raw readings from all the readers, the research team is in the position to design a system that can efficiently answer indoor spatial queries and track trajectories of objects. The research results of this project will improve the performance of numerous high value-added indoor applications and hence benefit the economy of our country. In addition, the ability to be able to locate people in indoor spaces will improve emergency response. The project will promote teaching, learning, and training by exposing both undergraduate and graduate students to mathematical and technological underpinnings in the field of spatial data management.In this project, the research team will develop an array of techniques to derive accurate object locations from erroneous RFID raw readings for supporting indoor spatial query evaluation and trajectory tracking. With accurate spatial query results and trajectory information, many high level applications (e.g., indoor layout planning and indoor location-based services) can be supported. This project will contribute to the research community by piloting novel indoor data management techniques that will accomplish the following goals: (1) develop and compare a number of Bayesian filtering-based location inference methods for evaluating spatial queries in indoor environments, (2) design novel indoor query evaluation algorithms for various spatial query types such as range query and k nearest neighbor query, (3) invent a hidden Markov model-based approach for indoor object trajectory tracking, and (4) implement a simulation toolkit and a prototype system, where all the components will be integrated for performance evaluation. All the research results and publications will be available on the project web site (http://www.eng.auburn.edu/~xqin/Indoor.htm).