Information Technology Reference
In-Depth Information
Chapter 2
Background
2.1 Introduction
In this chapter, we explore four areas of study directly related with HAR which are
essential to develop a global perspective of our research problem. They explain the
context where our proposed HAR systems can be applied to and also the mechanisms
through which it is possible to implement them.
Firstly, we introduce AmI and AAL as the frameworks where we focus our
research on HAR (Sect. 2.2 ). In particular, we target the detection of physical activ-
ities to provide context information for daily living applications, predominantly, for
people with motor impairments or activity limitations. Following this, we concen-
trate on the requirements to implement HAR systems. From the hardware aspect, we
give an overview of the sensors used for the recognition of activities in Sect. 2.3 and
also present the smartphone as it is our selected device for HAR (Sect. 2.4 ). Then,
from the software and mathematical standpoint, we give an overview of ML with
emphasis on SVMs (Sect. 2.5 ). Finally, we summarize the chapter in Sect. 2.6 .
2.2 Ambient Intelligence and Assisted Living
AmI is the paradigm that explores environments that behave intelligently and adap-
tively based on the actions of the people that are part of it. These environments are
aimed to respond, in a cooperative manner, to their needs which are identified through
the retrieval of context information gathered by means of the available technology.
This process stimulates the generation of a dynamic human-machine interaction
with the purpose of providing a better QoL to people, expressed, for instance, via
more efficient ways of solving tasks, faster assistance, and greater comfort (Aarts and
Wichert 2009 ). In the last decade of the 20th century, the termUbiquitous Computing
(UbiComp) made its appearance (Issarny et al. 2005 ). It considered the introduction
of unobtrusive computing devices in the environment. Following this, AmI emerged
 
Search WWH ::




Custom Search