|
||||||
|
||||||
| Undergraduate Honours Projects | ||||||
|
Carleton University - School of Computer Science Undergraduate Honours Project Winter 2012 Augmented Reality Using Geolocation Data and Dynamic Caching Dean Pearce 
ABSTRACT As the power of mobile devices increases, the ability to visually present up-to-date visual to users has become a reality. Many mobile devices now have the sensors available such as GPS, magnetometer and accelerometer which make geolocation based augmented reality possible. Key limitations in modern mobile network architecture include the limited speed at which devices can transfer data as well as the download bandwidth allowed to users based on their carrier plans. This paper describes a predictive caching algorithm which is capable of selectively retrieving augmented reality data for users based on their current geographic location. It investigates the success of such a predictive caching scheme in relation to the success rate for cache hits and the amount of data it caches unnecessarily. Results show that once a user is on a determined path, the rate of false positives significantly decreases in comparison to when the next location of a user is determined solely by heading. |
||||||
|
