Making Location-Aware Computing Working Accurately in Smart Spaces

Making Location-Aware Computing Working Accurately in Smart Spaces

Teddy Mantoro (University of Technology Malaysia, Malaysia), Media Ayu (International Islamic University Malaysia, Malaysia) and Maarten Weyn (Artesis University College of Antwerpen, Belgium)
DOI: 10.4018/978-1-60960-042-6.ch035
OnDemand PDF Download:
List Price: $37.50


In smart environment, making a location-aware personal computing working accurately is a way of getting close to the pervasive computing vision. The best candidate to determine a user location in indoor environment is by using IEEE 802.11 (Wi-Fi) signals, since it is more and more widely available and installed on most mobile devices used by users. Unfortunately, the signal strength, signals quality and noise of Wi-Fi, in worst scenario, it fluctuates up to 33% because of the reflection, refraction, temperature, humidity, the dynamic environment, etc. We present our current development on a light-weight algorithm, which is easy, simple but robust in producing the determination of user location using WiFi signals. The algorithm is based on “multiple observers” on ?k-Nearest Neighbour. We extend our approach in the estimation indoor-user location by using combination of different technologies, i.e. WiFi, GPS, GSM and Accelerometer. The algorithm is based on opportunistic localization algorithm and fuse different sensor data in order to be able to use the data which is available at the user position and processable in a mobile device.
Chapter Preview

The State Of The Art Of Location Aware Computing

Currently, Location-Aware Computing becomes a rapidly growing field in the area of Context-Aware Computing. User and equipment location are the two main focuses of developing location-aware applications. Unfortunately a range of mobile devices (Laptop, PDA, Smart Phone) in the market are still lacking of a satisfactory location technology, which enables them to estimate their own location.

Location-Aware Computing which promises accuracy, economy and ease of deployment, is currently still seen to be under construction. Numerous location models have been proposed in different domains and can be categorised into two classes, i.e. symbolic or descriptive (hierarchical, topological) location such as a city or a room, and coordinate (Cartesian, metric or geometric) location such as (x,y,z) coordinate (latitude, longitude, altitude) in GPS or active bat.

User location is a main concern of Location-Aware Computing, symbolic location is preferred over coordinate location in the user’s daily activities. The use of coordinate location for human-serving can be converted into symbolic location, which is a more natural human location description, which, except in special cases, makes daily communication easier.

Our previous work proposes the ηk-Nearest Neighbour to estimate symbolic user location (Mantoro & Johnson, 2005), instead of the used of neural network approach, which required a heavy computation effort during learning process (Mantoro, 2003) and we also proposed the use of multivariate regression estimation in estimating a coordinate of user location in indoor environment (Mantoro et al, 2008) both using IEEE 802.11 (Wi-Fi) signals.

Our Opportunistic Localisation (Weyn, et al 2009) describes the concept of using all available information which can be grasped by the mobile device in order to infer a location instead of using one fixed technology, together with a dynamic motion model which models the possible behaviour of a user.

In our current development, we are targeting to improve the accuracy of the user location determination by developing a new user location model which based on “multiple observers” of IEEE 802.11 (Wi-Fi) signals from small devices such as smart PDA.

Another constraint for localisation is the necessary offline training phase when the algorithm uses an algorithm based on fingerprinting. Our current research is developing a novel algorithm to offer an automatic fingerprinting, where the fingerprint is continually updated when more useful information is coming to the server.

Key Terms in this Chapter

WiFi: Wireless Fidelity, the Alliance to certify interoperability of IEEE 802.11

Coordinate Location: location representation based on a coordinate location (x,y,z), such as GPS (latitude, longitude and altitude). It is also known as Cartesian, Metric or Geometric location.

Ekahau: Commercial software which has capability to locate location in wireless (IEEE 802.11) local area network environment.

Active Bat System: The system that use an ultrasound time-of-flight measurement technique to provide location information.

Symbolic Location: location representation based on the descriptive of the location itself, such as a city or a named room. It is also known as descriptive, hierarchical or topological location.

Location-Aware Personal Computing: is concerned with the acquisition of coordinates or symbolic names which showing location of certain points with applied or based on information provided by small/personal devices, such as smart PDA, iPhone, etc.

Smart Environment: is a physical space which is smart in mature. The smartness of this environment is a product of interaction of different devices and computing systems.

Light-Weight Algorithm: The algorithm which is based on “multiple observers” of the Wi-Fi’s signals, and it offers an easy and simple calculation but robust in producing the determination of user location.

Radar: A radio-frequency (RF) based system for locating and tracking users inside buildings.

Complete Chapter List

Search this Book: