Using Colocation to Support Human Memory
Caitlin Lustig
ODCSSS Program
University College Dublin
Dublin, Ireland
celustig@cs.washington.edu
Hristo Novatchkov
ODCSSS Program
Dublin City University
Dublin, Ireland
ico@novatchkov.org
Lucy E. Dunne
University College Dublin
Dublin, Ireland
lucy.dunne@ucd.ie
Mike McHugh
Dublin City University
Dublin, Ireland
mmchugh@computing.dcu.ie
Lorcan Coyle
University College Dublin
Dublin, Ireland
lorcan.coyle@ucd.ie
ABSTRACT
The progress of health care in the western world has been
marked by an increase in life expectancy. Advances in life
expectancy have meant that more people are living with
acute health problems, many of which are related to impair-
ment of memory. This paper describes a pair of scenarios
that use RFID to assist people who may suffer from memory
defects to extend their capability for independent living. We
present our implementation of an RFID glove, describe its
operation, and show how it enables the application scenar-
ios.
1. INTRODUCTION
Increasing global life expectancies has opened up new issues
for society. Many people, faced with the reality of living
longer, wish to maintain their accustomed level of indepen-
dence as they age. The problem is that the health care that
allows them to live longer has not managed to conquer many
age-related ailments. As a result, there is a need to balance
the desire to live independently with the facts of deteriorat-
ing health. This paper describes an approach that have the
potential to help achieve that balance.
The impetus behind this work are the demographic changes
that are a result of the improved health care. On a local
level, the 2002 Irish census showed 11.13% of the popula-
tion are over 65; figures published by the Office for National
Statistics in the UK indicate that the equivalent figure was
15.96% of the population in 2003. Many of these people
live alone: 28.5% of the over-65s in the 2002 Irish figures.
This aging is not an isolated phenomenon, and is reflected
throughout the developed world [15].
From a technological point of view, the computing power
present in the average home is staggering compared with
50 years ago. How to harness this technological power to
assist independent living is a challenge which has been ad-
dressed via approaches ranging from the full instrumenta-
tion of smart homes [17], to the more individually targeted
life-logging and review [8, 5].
Much research on aides-me´moire for sufferers of memory
loss has been conducted. Oriani et al. demonstrate the ad-
vantages that electronic memory aids (EMAs) have, as well
as the motivation for their use by Alzheimer’s patients [13].
Hart et al. explore the use of EMAs for a different pop-
ulation, that of people recovering from traumatic head in-
juries [7]. These studies both comment on the need to re-
duce the barrier of use of electronic devices. If a prospective
user needs to remember complicated procedures or receive
specialised training to use the device, they are less likely to
want one.
RFID is a commercially available technology, and has proven
to be useful in real life applications. Our investigation into
the application of RFID to this area indicates that it signif-
icantly reduces the complexity of operation exposed to the
user since the system makes its observations based on nor-
mal every-day activities of the user. Additionally, the user
retains ownership of the gathered information – there is no
need for it to leave the home.
2. RFID AND COLOCATION
Radio Frequency Identification (RFID) technology has been
around since 1948 and has become an integral part of many
business processes [11]. On a personal scale, RFID badges
are prevalent in many environments (for example, [6]). These
types of badges lead to a common scenario, in which the
RFID tag is mobile and uniquely identifies an individual,
while the associated reader(s) are stationary (for example,
[6, 1]). Our application proposes to reverse this structure,
and use a mobile reader – which is associated with a unique
user – with multiple tags which are relatively immobile.
Our environment is populated with RFID-tagged artifacts.
These passive RFID tags uniquely identify each artifact; as
the user comes into contact with them the system records
the interaction. We have developed an RFID glove to gather
this information and transmit it to a database. The glove is
Copyright is held by the author/owner(s).
Supporting Human Memory with Interactve Systems, workshop at the
HCI 2007 (British HCI conference 2007), September 4th, 2007,
Lancaster, UK
Workshop "Supporting Human Memory with Interactive Systems", HCI Conference, September 4th, 2007, Lancaster, UK
- 41 -

Figure 1: The RFID glove. Clockwise from top left: the circuitry connecting the RFID reader to the Gumstix;
the portable RFID reader; the Gumstix motherboard; three types of RFID tags; the front view of the glove
– the circuitry and batteries are visible and the RFID reader is in a pouch under the palm; the rear view of
the glove – the Gumstix, serial connection, and WiFi antenna are visible.
made up of a portable RFID reader1, a full function minia-
ture computer system2, power, and connecting circuitry.
The RFID glove was assembled as follows3: We used a sports
glove as a scaffold; A pouch was added on the palm to house
the RFID reader. A wrist-strap was made with pouches for
the batteries that power the glove and a strap for the Gum-
stix. A circuit board was sewed into the wrist strap; this
houses a Voltage regulator, an LED for diagnostics, and a
simple circuit that converts the output of the RFID reader
into a form that can be consumed by a serial interface. The
total cost of components for the RFID glove is approxi-
mately $350 and individual tags cost in the region of $1
each. Figure 1 shows some images of the RFID glove in
action and some of its constituent parts.
In our system, passive RFID tags that are placed on various
artifacts transmit their tag IDs when activated by the RFID
reader in the glove. The RFID reader activates the tags,
which return their IDs to the reader. The tag IDs are then
sent to the Gumstix computer via a serial connection. The
tag IDs are processed and sent to a remote database across
a WiFi connection. In the database, tag IDs are associated
with descriptions of the objects to which they are attached.
1From Parallax — http://www.parallax.com/
2From Gumstix, our computer consists of a connex400
XMBT motherboard, console-st expansion, and wifistix CF
(EU) expansion — http://www.gumstix.com
3More details on the glove assembly are available on
the project’s wiki — https://secure.ucd.ie/twiki/bin/
view/GumstixRFID/WebHome
In order to ensure privacy, tag reads are only recorded if the
tag ID is already in the database. The database provides the
framework on which our scenario applications are developed.
RFID was an attractive technology to use in our system
because of its low cost and small size. The small size is par-
ticularly important because it allows us to easily attach tags
onto objects in our kitchen model in an unobtrusive man-
ner. Our RFID reader has a range of about 3 cm and so the
user must be picking up the tagged object or hovering their
hand over it in order for the tag to be read. However, RFID
comes with many problems: tags cannot be read through
the human body since it is mostly composed of water. For
this reason, we placed the reader onto the palm of the glove
rather than the back, even though this made it more dif-
ficult to grasp onto objects. We are currently considering
alternative mountings for the reader.
3. APPLICATION SCENARIOS
We chose to implement applications to target two scenarios
to assist sufferers of short-term memory loss in the home.
The first attempts to assist them to complete routine (but
complex) at-home tasks, the second tries to ensure that in-
structions for taking prescription medication are followed.
Both applications use RFID and colocation to attempt to
solve the problems at hand.
Workshop "Supporting Human Memory with Interactive Systems", HCI Conference, September 4th, 2007, Lancaster, UK
- 42 -