Chinese Journal of Computers   Full Text
Title	A Multi-Object Tracking Indoor Positioning System for Smart Space
Authors	GU Hong-Liang1) SHI Yuan-Chun2) SHEN Rui-Min1) CHEN Yu2)
Address	1)(Network Education College, Shanghai Jiaotong University, Shanghai 200030) 2)(Department of Computer Science and Technology, Tsinghua University, Beijing 100084)
Year	2007
Issue	No.9(1603—1611)
Abstract &   Background	Abstract The location sensing system (also called positioning system) is an underlying component for location-aware computing in Smart Space, which not only demands the location sensing system to locate multiple objects with desirable precision, but also superimposes the requirements, such as directional characteristic and portability etc, on it. This paper presents an indoor location sensing system, Cicada. This system is based on the TDoA(Time Difference of Arrival) between radiofrequency and ultrasound to estimate distance, and employs a technology integrating Slide Window Filter and Extended Kalman Filter to calculate location. As the experiments prove, Cicada can provide the average 5cm location precision both for static objects and for mobile objects, owns a nearly omni-directional working area ranging from -90degree to +90degree. Besides, it is able to run independently, mini and light so that it is convenient to be portable and even embedded into people’s paraphernalia. Further, it is also flexible in scalability, for neither quantity range nor constraint of structure form is demanded on its infrastructure. Last, it is fairly facile to be employed, attributing to the wireless connectivity in its infrastructure. Those advantages illuminate that it is able to be competent for all-round location sensing requirements of Smart Space. keywords pervasive/ubiquitous computing; location-aware computing; positioning system; extended Kalman filter; slide window filter background This research is supported by the National Natural Science Foundation of China under grants (60672066), the National High Technology Research and Development Program(863 Program) of China under grant No.2006AA01Z131. Pervasive computing (also called ubiquitous computing) is a hot research field recently. Context-aware computing is a paramount characteristic of pervasive computing. Location (along with ID) is the most important kind of context, as well as most robustly acquired by actual means, so that it turns out to be an actual hot research field. By far the comprehension on location-aware computing popularly accepted by the most scholars is that the applications or services can accord to the location and spatial relationship of objects to modify their own behaviors unobtrusively or non-intrusively to adapt to user’s purpose. Smart Space is a test-bed intensively embodying pervasive computing. The projects, Smart Classroom, SNC (Standard Natural Classroom) are the Smart Spaces applying in the fields of tele-education. In those projects, one of important objectives is to study how to realize location-aware computing in Smart Space. Location sensing systems (also called positioning system) is the underlying components for location-aware computing in Smart Space, which are to provide objects’ location for location-awareness. Until now, overseas there have emerged quite a few positioning systems, such as GPS, Active badge, RADAR, LANDMARC, 3D-iD, Cricket and Active BAT etc, some of which have even come into market in the international commercial fields. However, Smart Space, especially the one on tele-education, is a special indoor environment, which owns its appropriative requirements distinct from the general purpose environments. Those requirements span a variety of factors, including location precision, functionality of tracking multi-objects, rigorous directional characteristic, convenience of portability, flexibility of scalability, and facility of employment etc. Though those already existent positioning systems are prominent in this or that performance, so as to succeed in dominating some commercial fields, they are difficult in meeting the all-round demands of Smart Space satisfyingly. Perhaps, what Smart Space needs is not a system excelling at the general purpose, but owning a synthetic performance covering various factors, which might as well be a tradeoff between various performance. For designing a tailored system for Smart Space, we have struggled for a long time. This paper is a skeleton narration on our system, Cicada. A prototype as it is, it is bound be to own a promising perspective. Now we are managing to apply a patent for it.