Abstract
Indoor localisation systems are concerned with locating people or devices within an indoor environment. Within the field, there are a large number of technologies available for this task. Depending on the technology and underlying method of location estimate the accuracy is variable.
This thesis examines ways to increase the accuracy of indoor localisation systems focusing on using ubiquitous technologies. Two popular methods for indoor localisation, fingerprinting and trilateration, were investigated.
The improvement to the fingerprint location accuracy is approximately 85% and was achieved by excluding poor quality measurements through the manipulation of the radio hardware transmission powers. The trilateration scheme location accuracy was improved by 32% accompanied by a decrease in computation time by 70% in general, and up to two orders of magnitude in some cases.
One problem with these popular indoor localisation methods is the need for a large amount of training data. To counter this a crowd-sourced approach was developed to decrease the amount of time taken to deploy a localisation system while improving the accuracy by 25% with state-of-the-art systems.