Device-free localisation for assisted living applications

Abstract

Determining the location of individuals within indoor locations can be useful in variousscenarios including security, gaming and ambient assisted living for the elderly.Healthcare services globally are seeking to allow people to stay in their familiarhome environments longer due to the multitude of benefits associated with living innon-clinical environments and technologies to determine an individual's movementsare key to ensuring home emergencies can be detected and responded to promptly.Popular localisation technologies are device-based which requires the user to activelytake part in the aggregation of context information via wearing a traceable device.Such systems are currently inefficient in terms of the quality of data received and the expense of wearable device units as the user often either mislays the device,accidentally breaks it or forgets to wear it on a daily basis.

This work presents a device-free localisation system in which the user is not required to actively take part in the localisation process and therefore can continue their daily routine without the need to wear a traceable device. The principle behind this device-free strategy is the absorption phenomenon of the Received Signal Strength(RSS) of transmitted wireless signals as the human body crosses a transmitter receiver path. By using transmitter-receiver pairs, the absorption capacity of a human can be shown to exhibit signal patterns which can be used in locating and tracking within an environment. To this end, a new methodology which observes the absorption phenomenon that a human body exhibits on RSS signals is presented. Findings show that detection abilities are possible using a single transmitter-receiver pair, though for localisation, additional nodes are needed. The proposed DFL system design facilitates the use of a minimum number of wireless nodes with the help of a principal component analysis (PCA) based intelligent signal processing technique. Upon presenting the optimal setup for DFL, a new method for localisation is outlined known as the Signature Graph (SG) technique. Results demonstrate that human detection and tracking are possible to within 0:6m resolution with a minimal hardware infrastructure.

Date of Award	Aug 2011
Original language	English
Sponsors	Department for Employment and Learning, Northern Ireland
Supervisor	Girijesh Prasad (Supervisor), Jim Harkin (Supervisor) & Kevin Curran (Supervisor)