Location-based Systems

By: Keara Barrett

One of the objectives for computing in the future is to bind together the computer and the user so that the user’s interaction with the computing environment is as seamless and natural as possible. To make this happen a user’s location or position must be accessible.

Many location sensing and tracking systems have been developed in recent years, most notably the Global Positioning System (GPS), which we wrote about in an article not long ago.
This system has reliable coverage and by using a worldwide satellite system it can compute the location of a user to within a range of one to five metres. Emergency services and aircraft currently use the GPS.

However, if you want to get a more precise location of a user or if you need to locate a user indoors, GPS is not suitable.
Indoor location systems are not as mature or widely publicised as the GPS and whereas location to within a five metre range is considered acceptable from the GPS any indoor location system is required to provide much more precise information.
Broadly speaking this is how it happens. Firstly, sensor measurements are attained and then these measurements are collated to acquire the user’s position.

In most indoor systems the user carries either a transmitter or a receiver. If the user carries a transmitter that broadcasts signals then sensors fitted in the room are used to detect the emitted signal. These signals may be radio frequency signals, infrared signals or ultrasonic pulses. The measurements that are used to pinpoint the user may be either the signal strength of the signal detected by the sensor or the ‘time-of-flight’ of the signal from the transmitter to the sensor. For example the BAT system, which is one of AT&T’s indoor location based systems can find the 3D position of a user when given three or more ‘time-of-flight’ measurements.
All systems that exploit transmitters and receivers have their own pros and cons but the fact that the users must carry either a transmitter or receiver is a universal problem for such systems.

The solution to this problem is to use computer vision. With computer vision the room is fitted with cameras that gather colour and depth information and it is these colour and depth measurements that are interpreted to locate the user. Microsoft’s ‘Easy Living’ system is an example of a location system that utilizes computer vision.

Regardless of the technique used to locate the user the important thing is that this system is scalable. That is, it must be able to cover a relatively large area and in addition it must be able to locate a specified number of users within a particular distance in a specified time.

While significant advances have been made with indoor location systems, one important aspect remains an issue, user privacy. While information regarding the location of users is very beneficial for advances in computing this information also has the potential to threaten user security as the location of a user is known at all times.
A significant issue that many of the current systems do not address, this is one of the areas where much research will be carried out in the future.