Instrumentation & Measurement Magazine 24-6 - 46

Human Activity Recognition with
Device-Free Sensors for Well-Being
Assessment in Smart Homes
Hossein Raeis, Mohammad Kazemi, and Shervin Shirmohammadi
H
uman Activity Recognition (HAR) has attracted
much attention in the last two decades with applications
such as remote health monitoring, security
and surveillance, and smart environments. Specifically, for
well-being assessment, HAR systems give us the possibility
of recognizing important physical activities in the patient's
daily living. For instance, using motion sensors to monitor
and record the physical situations and postures of patients
with chronic conditions such as arthritis and cardiovascular
disease which cause limitations in mobility can be useful in behavior
assessment [1]. These physical records, especially for
people with disabilities or elderly people, provide caregivers
with useful information for treatment. Another example
is fall detection that could notify caregivers instantly when
a person falls. Today, many types of sensors are used for human
activity recognition, including vision-based, wearable,
object-tagged and device-free. In this article, we focus on device-free
sensors and give and overview of their applications
in HAR for well-being assessment in smart homes, including
examples from the existing literature and our own test results
for simple activity recognition with some of these sensors. We
will see that device-free sensors are used predominantly for
fall detection, although other well-being applications such as
cognitive assessment, respiration monitoring, and dementia
detection have emerged as well. Let us begin by looking at various
types of HAR sensors and identifying the characteristics
of each of them.
Existing methods for HAR can be categorized into vision-based
and sensor-based. Sensor-based itself can be
divided into wearable, object-tagged and device-free sensors
[2]. Vision-based methods, also known as Vision-Based
Measurement [3], are the most common, because analyzing
recorded videos with appropriate computer/machine
vision techniques can provide accurate results. However,
cameras can face problems such as occlusion, viewpoint restriction,
lighting dependency and frame resolution. Privacy
is another important concern in such systems, although it
has been shown that older adults are willing to trade some
privacy for autonomy [4]. In wearable systems, sensors
46
such as gyroscopes, accelerometers and heartrate monitors
are attached to or carried by the users, and are hence considered
obtrusive. But the diversity in wearable sensors
and their low price have led to their wide usage. Objecttagged
sensors use tags attached to objects and appliances of
daily use, instead of being attached to the users. Interaction
between users and these tagged objects can then lead to activity
recognition. Although object-tagged sensors avoid the
obtrusive nature of wearable sensors, they are limited to recognizing
activities only when users interact with the tagged
objects. On the other hand, device-free sensors do not need
to be carried by users or attached to objects. Examples are
Wi-Fi and radar-based sensors, which can be deployed in the
smart home and capture physical activity data. Such sensors
also solve the privacy issue and have better performance in
complex indoor environments that include many obstacles
with moving objects which may cause several signal multipaths
[5]. Table 1 shows a summary of HAR methods and
their pros and cons.
In the past few years, device-free sensors have experienced
remarkable popularity due to their above-mentioned benefits
and their ability to easily identify basic activities such as
sitting, walking, running, and more. Health-related activities
such as exercising are complex activities composed of these
basic activities [2], so their recognition is important for wellbeing
assessment.
Also, the simple activities themselves have a relationship
with health status such as arthritis, cardiovascular, or neurodegenerative
diseases [1]. For example, it has been shown
that walking speed could predict dementia [18]. Device-free
sensors can therefore play an important role for well-being
assessment in smart homes. Such sensors are available as
piezoelectric, pyroelectric infrared (PIR), infrared array, Wi-Fi,
and microwave doppler radar, summarized in Table 2 and described
in detail next.
Device-Free Sensors
In this section, the sensors used for healthcare applications are
explained with sufficient details.
IEEE Instrumentation & Measurement Magazine
1094-6969/21/$25.00©2021IEEE
September 2021
Instrumentation & Measurement Magazine 24-6

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 24-6
