Perceptions and Acceptability of Sensor-Based Activity Recognition

Systems Among Older Adults and Their Families

David D

ıaz-Jim

enez

, Jos

e L. L

opez

, Francisco J. Flores-Avil

es and Macarena Espinilla

University of Ja

en, Department of Computer Sciences, Spain

{ddjimene, llopez, fjfa0001, mestevez}@ujaen.es

Keywords:

Activity Recognition Systems, Sensor Devices, Multi-Occupancy Environments, User Perceptions, Perceived

Social Support.

Abstract:

This paper examines activity recognition systems that use sensor devices with speciﬁc activity models. It

presents a new system that combines motion, open/close, and ambient sensors with wristband devices and

location beacons. Alongside a detailed review of the system, the study also explores the views of two main

groups of users: older adults and their family members. Although related studies exist, this research introduces

the system and thoroughly analyzes user feedback. An important aspect is the acknowledgment of improve-

ments in sensor-based smart devices, especially in terms of size and subtlety compared to earlier bracelet

designs. This study included 40 anonymous participants who tested these system and key factors analyzed

include acceptability, safety, peace of mind, privacy, quality of life, autonomy, trust, perceived social support,

loneliness, and economic cost. This assessment offers useful insights into how users perceive and accept the

system, to understand the main concerns with commonly used devices like cameras and sensors helps identify

which devices older adults might be open to using in their routines. These insights are expected to guide the

development of future systems that better address user needs.

1 INTRODUCTION

The growing demographic shift towards an aging pop-

ulation presents a formidable challenge in society to-

day (United Nations, 2022), requiring the develop-

ment of innovative solutions to enhance the well-

being of older individuals. In response to this pressing

need, Human Activity Recognition Systems (HAR)

(Arshad et al., 2022; Bian et al., 2022) have emerged

as promising tools with the potential to address di-

verse health and autonomy-related needs among the

elderly.

These systems employ various approaches, with

the key determining factor being sensor selection. A

wide variety of sensors have been studied for activity

recognition (Liu et al., 2020; Fu et al., 2020; Dang

et al., 2020), particularly environmental and vision-

based sensors. The efﬁcacy and applicability of activ-

ity recognition systems hinge signiﬁcantly on these

two sensor types.

On the one hand, environmental sensors, designed

https://orcid.org/0000-0003-1791-4258

https://orcid.org/0000-0003-2583-8638

https://orcid.org/0000-0003-1118-7782

to capture data about the surrounding context, offer

insights into activities involving movement or envi-

ronmental changes (Dang et al., 2020; Yuan et al.,

2022). While helpful for recognizing activities with-

out direct visual observation, their limitations include

the inability to capture ﬁne details, potentially leading

to information loss that might be crucial for accurate

identiﬁcation (Ahad et al., 2020). External factors,

such as multi-occupancy, can also cause interference

and affect data reliability.

On the other hand, vision-based sensors use cam-

eras or optical devices to capture activities visually

(Beddiar et al., 2020), excelling in recognizing ac-

tions involving speciﬁc movements, gestures, or inter-

actions with objects (Franco et al., 2020; Dang et al.,

2020; Ramirez et al., 2021). Despite their ability

to provide detailed visual information, they also face

challenges, such as dependency on proper illumina-

tion, sensitivity to obstructions, and privacy concerns

arising from image and video capture (Langheinrich,

2002).

Alternative methods involve wearable devices

such as activity bracelets and smartwatches, which

extract biometric data from users (Fan and Gao, 2021;

Huang et al., 2021).

240

Díaz-Jiménez, D., López, J. L., Flores-Avilés, F. J. and Espinilla, M.

Perceptions and Acceptability of Sensor-Based Activity Recognition Systems Among Older Adults and Their Families.

DOI: 10.5220/0013278700003938

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 11th International Conference on Information and Communication Technologies for Ageing Well and e-Health (ICT4AWE 2025), pages 240-247

ISBN: 978-989-758-743-6; ISSN: 2184-4984

Fog nodes are placed near data sources to col-

lect and pre-process data, optimizing transmission to

the cloud and reducing network load. They may use

lightweight models for initial tasks. The cloud man-

ages then data storage, applying more complex activ-

ity recognition models. This division optimizes trans-

mission, ensures ﬂexibility, and enables comprehen-

sive data management.

Numerous proposals have emerged within this do-

main, aiming to provide effective and non-intrusive

solutions for older adults. Examples include the

Konekta2 system (Codina, 2022), Beprevent (Ger,

2020), the Noa smart pill dispenser (Inb, 2020), and

the ACTIVA system (Montoro Lend

ınez et al., 2023),

each addressing speciﬁc aspects of monitoring and

support for daily activities.

When comparing state-of-the-art activity recogni-

tion systems, several points can be noted. Some ap-

proaches require the installation of surveillance cam-

eras and image processing, like the HAR system in

Hussain et al. (Hussain et al., 2022), which combines

a pre-trained Vision Transformer and recurrent neural

networks (LSTM) to capture long-term temporal in-

formation. Su et al. (Su et al., 2023) propose a deep

learning-based framework for real-time non-contact

human activity detection using self-powered sensors.

In this work, a multi-layer bidirectional short- and

long-term memory (MBLSTM) network is used to

process Wi-Fi channel state information (CSI) and

recognize human activities. While the system shows

very promising results, multi-occupancy is a chal-

lenge when using this technique, since it cannot dis-

cern several subjects. This is also the case in (Bha-

vanasi et al., 2023), which uses compact radar sen-

sors for patient activity recognition in hospitals. Con-

versely, our approach avoids these concerns by using

sensors that do not compromise individuals’ privacy,

along with a system to deal with multi-occupancy.

The success of these HAR systems hinges on user

perceptions, particularly those of older adults, and the

acceptability of the sensors employed (Camp et al.,

2022). Older individuals often face challenges re-

lated to mobility, health, and safety (Mottram et al.,

2008; Niccoli and Partridge, 2012). To ensure the ef-

fectiveness of these solutions, it is crucial to under-

stand the acceptability of different sensor types in ac-

tivity recognition systems. Thus, the research ques-

tion guiding this study focuses on the acceptability of

sensor types in the activity recognition of older adults,

as follows:

Research Question: How acceptable are different

sensor types to older adults for activity recognition?

This work aims to address the research question

through a comprehensive analysis of the perceptions

held by older adults and their family members regard-

ing sensor-based activity recognition systems. The

speciﬁc objectives include:

1. Study Older Adults’ Perceptions: Conduct an

in-depth analysis of how older individuals per-

ceive activity recognition systems, with an em-

phasis on identifying both advantages and disad-

vantages that these systems may entail for them.

2. Assess Acceptability: Evaluate how acceptable

activity recognition systems are to older adults,

addressing crucial aspects such as trust, security,

privacy, and peace of mind.

3. Evaluate Family Members’ Opinions: Investi-

gate how family members view activity recogni-

tion systems, particularly focusing on aspects like

security, privacy, usability, and other relevant con-

siderations.

This study will analyze the role of sensors in activ-

ity recognition systems in the homes of older adults.

The aim is to gain a greater understanding of their

needs, since these systems can be used for a wide

range of purposes, such as detection of abnormal be-

haviors (K

onig et al., 2015; Umbricht et al., 2020) or

monitoring activities that encourage autonomy.

Identifying the primary concerns around com-

monly used devices, such as cameras and sensors, is

crucial to understand which devices people would be

willing to integrate into their daily lives. This under-

standing will enable a focused effort to develop new

systems that capitalize on these ﬁndings.

The structure of the paper is as follows: in Sec-

tion 2, we present a novel sensor-driven connected

health system designed to monitor people’s activities

within their homes. Section 3 presents the method-

ology used to conduct the analysis. In Section 4, the

results from the questionnaires are discussed. Finally,

Section 5 sets out the conclusions of the paper.

2 MATERIALS AND METHODS

In this section, we present the activity recognition sys-

tem used in our study and explain the method fol-

lowed for our acceptability analysis. This system

serves as a basic framework for subsequent evaluation

by direct users, older adults, and family members, en-

suring they are able to provide a comprehensive as-

sessment and usability feedback.

2.1 System

The architecture of the proposed activity recognition

system is illustrated in Figure 1. This ﬁgure pro-

Perceptions and Acceptability of Sensor-Based Activity Recognition Systems Among Older Adults and Their Families

241

vides a visual representation of the system’s design

and components, offering a better understanding of

its structure and functionality.

Figure 1: System Architecture.

2.1.1 Device Layer

The device layer comprises various devices that con-

tribute data to the system, including sensors and loca-

tion devices:

• Open and Close Sensors. These sensors use a

magnetic ﬁeld to detect separations between their

components. In our system, they are deployed at

the main entrance and on containers storing medi-

cations for the users. The speciﬁc sensor we have

deployed is the Aqara Door and Window Sensor,

which uses Zigbee technology and has an esti-

mated battery life of two years.

• Motion Sensors. Infrared-based motion sensors

detect infrared radiation emitted by objects within

their ﬁeld of view. Strategically positioned in ar-

eas like bathrooms, communal spaces, and beds,

these sensors track residents’ movements, provid-

ing insights into activities such as sleeping and

toileting. The Aqara P1 motion sensor features

a sensing angle of approximately 170° and a sens-

ing distance of approximately 7 meters. It works

with Zigbee 3.0 wireless connections and has a

battery life of up to 5 years. This device, unlike

others, allows us to modify the detection range

and time, which makes it adaptable to different

activities by deﬁning speciﬁc detection zones.

• Temperature and Humidity Sensors. These

sensors monitor ambient conditions like temper-

ature and humidity. They provide data regard-

ing environmental parameters in monitored ar-

eas. Within our system, they are deployed in

bathing areas to detect changes in related activi-

ties. The Aqara Temperature and Humidity Sen-

sor uses Zigbee technology and has a battery life

of two years.

• Location Detection. To address multi-occupancy

scenarios, we adopted the methodology outlined

in the ACTIVA system (Montoro Lend

ınez et al.,

2023; Espinilla et al., 2018a; Espinilla et al.,

2018b). This method relies on Received Sig-

nal Strength Indicator (RSSI) measurements be-

tween ﬁxed anchor devices positioned in various

rooms and a beacon device carried by the user.

Raspberry Pi 4 devices and external Bluetooth 4.0

modules serve as anchor devices, while Mi Band

3 activity bracelets function as beacons. These

leverage their Bluetooth connectivity to capture

RSSI between anchors, enabling location detec-

tion. The activity bracelets have an estimated bat-

tery life of 3 weeks.

These sensors have been selected speciﬁcally for

the activities under monitoring: physical activity,

sleep patterns, hygiene routines, and dietary habits.

2.1.2 Fog Layer

The fog layer is a pivotal component within the sys-

tem, featuring a Raspberry Pi equipped with a Blue-

tooth adapter and a Zigbee Conbee2 communica-

tions module. This central node interconnects vari-

ous devices distributed throughout the environment.

Leveraging the Zigbee protocol, it establishes con-

nections with different sensors positioned throughout

the living space. We adopted Zigbee because of its

widespread availability in commercial sensors and be-

cause it offers extensive communication capabilities

coupled with low power consumption, thus render-

ing it preferable over alternatives such as Wi-Fi (Lee

et al., 2007).

These sensors are managed through the Home As-

sistant platform

, which allows us to conﬁgure and

integrate devices operating on different technologies.

Upon receiving data within the platform, it is dissem-

inated across a network under the MQTT protocol

(through by Mosquitto Broker

) for further process-

ing and transmission to the database.

Regarding location detection, each anchor device

publishes its RSSI values relative to the beacon within

the MQTT network. When the central node receives

https://www.home-assistant.io/

https://mosquitto.org/

ICT4AWE 2025 - 11th International Conference on Information and Communication Technologies for Ageing Well and e-Health

242

these values, a location model is run to determine the

device’s location. The resulting locations are subse-

quently transmitted to the cloud layer for storage and

further processing.

2.1.3 Cloud Layer

The cloud layer is another core element of the system,

housing the majority of its components. Comprising

four key elements – a database, a reverse proxy, an

AI computing module and web platform – this layer

orchestrates crucial system functionalities.

• Database. MongoDB

, a NoSQL document-

based database, has been selected to accommo-

date the heterogeneous data obtained from vari-

ous sensors. Its ﬂexible schema aligns well with

the diverse nature of the sensor data, meeting the

system’s requirements effectively.

• Reverse Proxy. We chose Nginx

as a reverse

proxy because it meets stringent security demands

across all system elements, while allowing for en-

hanced scalability. The system’s security is forti-

ﬁed by TLS implementation for inbound and out-

bound communications, extensive logging for au-

dit purposes, and deployment of attack mitigation

techniques, such as DDoS attacks.

• AI Computing Module. The AI computing mod-

ule analyzes user activities and veriﬁes their ad-

herence to predeﬁned activity regimens estab-

lished by the researchers or healthcare staff, which

set out the objectives or healthy habits individu-

als must uphold (Montagut-Mart

ınez et al., 2022).

The module processes daily sensor and location

data for each user, leveraging predeﬁned rules to

categorize activities. Subsequently, it evaluates

whether these activities align with the stipulations

outlined in the established regimen.

• Web Platform. The web platform, developed un-

der the Django framework

, serves as a central-

ized hub for managing sensors and disseminating

information to healthcare personnel. Django’s ro-

bust toolset and comprehensive library of exten-

sions facilitate agile development.

It is noteworthy that all components operate

within containers, leveraging Docker technology to

establish a private network, which provides the advan-

tage of enhanced scalability and security. This con-

tainerized approach enables the creation of container

https://www.mongodb.com/

https://www.nginx.com/

https://www.djangoproject.com/

replicas to bolster system capacity and ensures isola-

tion from other host system components, improving

overall system robustness and resilience.

2.1.4 Application Layer

As previously discussed, the web platform provides

comprehensive management functionalities for all

system components. Speciﬁcally, it allows techni-

cal personnel to oversee sensors, anchors, activity

bracelets, and residential units within the system. At

the same time, healthcare professionals can tailor

health or activity regimens to meet the unique needs

of individual users. Staff are also able to monitor

compliance with these regimens across various time

intervals, ranging from days to weeks and months. In

addition, they can aggregate data by specifying start

and end dates, enabling an effective assessment of

users’ compliance with predeﬁned activities.

2.1.5 Test Environment

Prior to deploying the system in the homes of the peo-

ple to be monitored, the system was deployed in a

SmartLab. The SmartLab is a testing environment

where several technologies are deployed in evalua-

tion for research purposes. At the moment it has a

wide variety of sensors and systems, mainly focus on

activity recognition. Although the SmartLab presents

a testing environment, it is constrained in certain sit-

uations. An example of this is the unavailability of

running water in the facility and the lack of a shower,

having to simulate activities such as brushing teeth,

showering and so on, therefore results may vary in

real environments. As can be seen in Figure 2, the

different elements that conform the system are shown

deployed.

3 METHODOLOGY

To conduct this research, a sample of 40 anonymous

individuals was recruited on a voluntary basis, each

having provided informed consent. The participants

were categorized into two distinct groups: a cohort

of 20 adults aged between 65 and 82 years, with an

equal gender distribution, and another cohort of 20

caregivers or adult family members, aged between 23

and 70 years, predominantly female.

To address the understanding of the system by the

participants recruited for our sample group, it is im-

perative to note that we ensured that they received in-

depth explanations of the sensors and devices used

along with the methods of data collection and the

Perceptions and Acceptability of Sensor-Based Activity Recognition Systems Among Older Adults and Their Families

243

Figure 2: System elements deployed in the SmartLab.

recording of their activities. To ensure that partici-

pants understood the implications of the study before

giving their informed consent.

Two bespoke instruments were used for data col-

lection: a survey drawn up for older adults and an-

other designed for their family members. The ques-

tionnaire directed at older adults, comprising 19 items

and three open-ended questions, delved into critical

domains such as acceptability, safety, quality of life,

personal autonomy, intimacy, privacy, trust, loneli-

ness, perceived social support, and ﬁnancial invest-

ment. The questionnaire targeting relatives of the

older adults, encompassing 12 items and 10 short-

answer questions, explored similar themes, with the

aim of obtaining a complementary perspective from

the family members’ point of view. The possible an-

swers are Not at all, A little, Moderately, Quite a lot

and A lot, depending on how much the person agrees

with the question. The questionnaires as well as the

results of the questionnaires can be found at the fol-

lowing link: Perceptions study

The survey was designed based on previous re-

search that explored older adults’ perception and ac-

ceptance of sensor technology and home health mon-

itoring. The previous studies used as the basis for the

design were the following:

• “Elderly persons’ perception and acceptance of

using wireless sensor networks to assist health-

care” (Steele et al., 2009), which examined how

older people perceive and accept the use of wire-

less sensor networks for healthcare and provided

valuable insights into their attitudes and concerns

regarding home monitoring technology and their

willingness to adopt it;

• “Perceptions of In-home Monitoring Technology

for Activities of Daily Living: Semistructured In-

terview Study With Community-Dwelling Older

Adults” (Camp et al., 2022), which used semi-

structured interviews to understand older adults’

perceptions of in-home monitoring technology for

activities of daily living, providing relevant infor-

mation on their opinions on the usefulness, ease

of use and concerns associated with monitoring

technology; and

• Statistical Study of User Perception of Smart

Homes during Vital Signal Monitoring with an

Energy- Saving Algorithm” (Del-Valle-Soto et al.,

2022), which analyzed user perception of smart

homes during vital sign monitoring with an

energy-saving algorithm and was fundamental to

understanding how users interact with smart home

technology and their expectations in terms of en-

ergy efﬁciency and functionality.

These studies provided a conceptual framework

for the design of the survey, allowing us to iden-

tify key factors related to the perception and ac-

ceptance of monitoring technology in the home.

Based on this previous work, the design of the

questionnaires focused on critical aspects such as

ease of use, perceived usefulness, privacy con-

cerns, and willingness to adopt home monitoring

technologies.

Although the hardware conﬁgurations may be dif-

ferent from previous studies, the underlying technol-

ogy remains very similar. As mentioned above, the

ICT4AWE 2025 - 11th International Conference on Information and Communication Technologies for Ageing Well and e-Health

244

sensors and their functionality have been explained in

detail so as to ensure that they are thoroughly under-

stood by all participants. This helps to minimize the

potential effects that variations in hardware or soft-

ware could have on subjective opinion, even if envi-

ronmental factors have not been fully controlled for.

Data collection began with reaching out to par-

ticipants by telephone, followed by distributing the

survey links via the Google Forms platform. Ba-

sic demographic information, including age, gender,

and marital status, was gathered. This standardized

procedure was uniformly applied to both participant

groups, ensuring methodological consistency and the

acquisition of high-quality data.

4 DISCUSSION

4.1 Older Adult Results

The results of the surveys for older adults provide

valuable insights into their perceptions of sensor-

based activity recognition systems. The primary aim

was to analyze their perspectives on acceptability,

safety, quality of life, personal autonomy, privacy,

trust, perceived social support, and their views re-

garding the advantages and disadvantages of such sys-

tems.

Regarding acceptability , 25% of participants ex-

pressed reluctance to use cameras providing clear

footage, while 10% were very willing, and 20% were

quite willing. Regarding the installation of ther-

mal cameras, 45% had no objection. With regards

to the use of sensors on everyday objects, 75% ex-

pressed no reservations. The results revealed a gen-

eral acceptance of these systems among older adults,

with motion and open/closed sensors emerging as the

most widely accepted components, which is consis-

tent with previous research (Camp et al., 2022).

Concerning security and quality of life, 45% of

the sample would signiﬁcantly feel safer with the im-

plementation of sensor-based systems. In terms of

quality of life and personal autonomy, 35% believed

their personal autonomy would substantially increase,

while 25% thought it would increase to a considerable

extent. Regarding overall quality of life, only 20% an-

ticipated a signiﬁcant increase. Overall, our survey

suggests that sensor-based activity recognition sys-

tems enhance personal autonomy and support inde-

pendent living for older adults, especially when used

for medical purposes or emergencies. These ﬁndings

align with the Steele et al. study (Steele et al., 2009),

where participants prioritized timely assistance over

data privacy concerns.

On the matter of privacy, 55% of participants felt

that sensor-based systems would moderately respect

their privacy. Regarding trust, 75% believed these

systems could enhance the speed of response in the

event of an emergency.

When it came to perceived social support and

loneliness, 40% of participants believed they would

experience increased social support with the im-

plementation of these systems. Additionally, 50%

thought that these systems would moderately help al-

leviate feelings of loneliness. In terms of peace of

mind, our study found that sensor systems provide

an increased sense of safety and calmness, consistent

with ﬁndings by Del Valle et al. (Del-Valle-Soto et al.,

2022) on the perceptions of smart home users during

monitoring.

Finally, regarding advantages and disadvantages,

it is noteworthy that the majority of participants be-

lieved these systems would offer peace of mind, secu-

rity, assistance, and independence. As for the draw-

backs, price, lack of privacy, and complexity of use

emerged as the most prevalent concerns. Concerns

about the cost and usability of these systems remain

prevalent among older adults, as seen in our study and

supported by previous research by Claes et al. (2015)

(Claes et al., 2015).

4.2 Family Survey Results

The results of the survey for family members or care-

givers provide additional insights into their percep-

tions of sensor-based activity recognition systems.

Regarding acceptability, 28.57% of participants

expressed a strong willingness to use these systems,

and an additional 28.57% are moderately willing.

Concerning security and peace of mind, 38.10% of

respondents believe that these systems would signiﬁ-

cantly contribute to providing peace of mind regard-

ing their elderly relatives.

Concerning quality of life, 28.5% of participants

believe that use of these systems could substantially

enhance autonomy and safety, contributing signiﬁ-

cantly to an improved quality of life. Regarding pri-

vacy, 42.86% of family members express concerns

that these systems may not adequately respect their

relative’s privacy.

In terms of utility, 52.38% of participants be-

lieve that these systems would provide a considerable

amount of information about their family member’s

well-being, while 42.86% think it would be highly

beneﬁcial if their family member had this type of sys-

tem. Our study suggests that sensor systems offer

peace of mind and utility, although direct support-

ing research is lacking. However, Camp et al.(Camp

Perceptions and Acceptability of Sensor-Based Activity Recognition Systems Among Older Adults and Their Families

245

et al., 2022) found that younger participants were

more knowledgeable and accepting of activity mon-

itoring systems than older adults.

Opinions regarding home monitoring systems

among family members encompass a diverse spec-

trum. Paramount among concerns is the safeguarding

of privacy, with a notable reluctance to install cameras

in speciﬁc living spaces. At the same time, consider-

ations around convenience, user-friendliness, and the

upkeep of these systems are pivotal.

In their answers to the open-ended questions, par-

ticipants placed signiﬁcant value on the efﬁcacy, pre-

cision, and incorporation of virtual assistants within

these systems. In particular, speed of connectivity

was considered critical for an effective response in

case of emergencies.

In terms of technological preferences, there is a

clear preference for sensor-based solutions, followed

by cameras and wearable devices.

Financial commitment towards these systems

varies, with respondents indicating a willingness to

invest between 200 to 4000 euros, alongside an open-

ness to monthly subscription models. However, un-

certainty prevails around the matter of pricing.

Regarding information alerts, unanimous prefer-

ence is observed for urgent notiﬁcations via telephone

calls and real-time updates through mobile applica-

tions. Conversely, periodic reports on the well-being

or activities of family members are deemed non-

essential by certain respondents.

Connectivity to emergency services is unani-

mously seen as imperative among participants.

In discussions pertaining to camera-based sys-

tems, opinions diverge. While a portion of respon-

dents expresses indifference, the remaining half un-

derscores the necessity for stringent privacy protocols

and data management practices.

5 CONCLUSIONS

In this paper, we propose an innovative activity recog-

nition system utilizing a suite of sensors—such as

motion detectors, wrist-worn devices, open/close sen-

sors, and location beacons to cover multi occupancy

environments. The research explores system’s accep-

tance, with particular emphasis on reliability and efﬁ-

ciency, offering a nuanced understanding of its prac-

tical applications in everyday contexts.

The paper examines perceptions among older

adults and their families regarding this technological

solution, considering factors like perceived accept-

ability, usability, and the impact on autonomy, qual-

ity of life, and social support. This dual approach

bridges the technical capabilities of the system with

the subjective experiences of potential users, provid-

ing a well-rounded assessment of the system’s bene-

ﬁts and possible areas for reﬁnement.

The main key ﬁndings underscore the necessity

of ensuring future designs are guide by user pri-

vacy, accessible in terms of ease of use, and ﬁ-

nancially sustainable. By identifying prevalent con-

cerns—especially around commonly deployed de-

vices like cameras and sensors—this study suggests

a pathway toward developing supportive technologies

that older adults can trust and rely on. Prioritizing

simplicity, privacy, and reliability can lead to solu-

tions that not only offer technological support but also

enhance daily living and respect individual prefer-

ences, fostering a genuine integration of technology

in their routines.

ACKNOWLEDGMENTS

This result has been partially supported by

grant PID2021-127275OB-I00 funded by MI-

CIU/AEI/10.13039/501100011033 and by “ERDF A

way of making Europe”, grant PDC2023-145863-I00

funded by MICIU/AEI/10.13039/501100011033 and

by “European Union NextGenerationEU/PRTR”,

and grant M.2 PDC 000756 funded by Consejer

ıa

de Universidad, Investigaci

on e Innovaci

on and by

ERDF Andalusia Program 2021-2027.

REFERENCES

(2020). Beprevent un asiente personal que facilita la vida

independiente de los mayores. Available online:

https://www.geriatricarea.com/2017/08/24/beprevent-

un-asiente-personal-que-facilita-la-vida-

independiente-de-los-mayores/ (accessed on 14

May 2024).

(2020). Noa smart pill dispense. Available online:

https://inbizi.es/noa/ (accessed on 14 May 2024).

Ahad, M. A. R., Antar, A. D., and Ahmed, M. (2020).

Sensor-based human activity recognition: Challenges

ahead. pages 175–189.

Arshad, M. H., Bilal, M., and Gani, A. (2022). Human ac-

tivity recognition: Review, taxonomy and open chal-

lenges. Sensors (Basel, Switzerland), 22.

Beddiar, D., Nini, B., Sabokrou, M., and Hadid, A.

(2020). Vision-based human activity recognition: a

survey. Multimedia Tools and Applications, 79:30509

– 30555.

Bhavanasi, G., Werthen-Brabants, L., Dhaene, T., and

Couckuyt, I. (2023). Open-set patient activity recogni-

tion with radar sensors and deep learning. IEEE Geo-

science and Remote Sensing Letters, 20:1–5.

ICT4AWE 2025 - 11th International Conference on Information and Communication Technologies for Ageing Well and e-Health

246

Bian, S., Liu, M., Zhou, B., and Lukowicz, P. (2022). The

state-of-the-art sensing techniques in human activity

recognition: A survey. Sensors, 22(12):4596.

Camp, N., Johnston, J., Lewis, M. G., Zecca, M., Di Nuovo,

A., Hunter, K., and Magistro, D. (2022). Percep-

tions of in-home monitoring technology for activi-

ties of daily living: Semistructured interview study

with community-dwelling older adults. JMIR aging,

5(2):e33714.

Claes, V., Devriendt, E., Tournoy, J., and Milisen, K.

(2015). Attitudes and perceptions of adults of 60 years

and older towards in-home monitoring of the activi-

ties of daily living with contactless sensors: an explo-

rative study. International journal of nursing studies,

52(1):134–148.

Codina, T. (2022). Konekta2, sensorizaci

on inteligente

de viviendas de personas mayores - fundaci

isocial. innovaci

o en l’acci

o social. Available

online: https://isocial.cat/es/konekta2-sensoritzacio-

intel-ligent-dhabitatges-de-persones-grans/ (accessed

on 14 May 2024).

Dang, L. M., Min, K., Wang, H., Piran, M. J., Lee, C. H.,

and Moon, H. (2020). Sensor-based and vision-based

human activity recognition: A comprehensive survey.

Pattern Recognition, 108:107561.

Del-Valle-Soto, C., Nolazco-Flores, J. A., Del Puerto-

Flores, J. A., Vel

azquez, R., Valdivia, L. J., Rosas-

Caro, J., and Visconti, P. (2022). Statistical study

of user perception of smart homes during vital signal

monitoring with an energy-saving algorithm. Interna-

tional Journal of Environmental Research and Public

Health, 19(16):9966.

Espinilla, M., Mart

ınez, L., Medina, J., and Nugent, C.

(2018a). The experience of developing the ujami

smart lab. Ieee Access, 6:34631–34642.

Espinilla, M., Medina, J., Hallberg, J., and Nugent, C.

(2018b). A new approach based on temporal sub-

windows for online sensor-based activity recognition.

Journal of Ambient Intelligence and Humanized Com-

puting, pages 1–13.

Fan, C. and Gao, F. (2021). Enhanced human activity recog-

nition using wearable sensors via a hybrid feature se-

lection method. Sensors (Basel, Switzerland), 21.

Franco, A., Magnani, A., and Maio, D. (2020). A multi-

modal approach for human activity recognition based

on skeleton and rgb data. Pattern Recognit. Lett.,

131:293–299.

Fu, B., Damer, N., Kirchbuchner, F., and Kuijper, A. (2020).

Sensing technology for human activity recognition: A

comprehensive survey. Ieee Access, 8:83791–83820.

Huang, W., Zhang, L., Gao, W., Min, F., and He, J.

(2021). Shallow convolutional neural networks for

human activity recognition using wearable sensors.

IEEE Transactions on Instrumentation and Measure-

ment, 70:1–11.

Hussain, A., Hussain, T., Ullah, W., and Baik, S. (2022).

Vision transformer and deep sequence learning for hu-

man activity recognition in surveillance videos. Com-

putational Intelligence and Neuroscience, 2022.

onig, A., Junior, C. C., Derreumaux, A., Bensadoun, G.,

Petit, P.-D., Br

emond, F., David, R., Verhey, F., Aal-

ten, P., and Robert, P. (2015). Validation of an au-

tomatic video monitoring system for the detection of

instrumental activities of daily living in dementia pa-

tients. Journal of Alzheimer’s disease : JAD, 44

2:675–85.

Langheinrich, M. (2002). A privacy awareness system

for ubiquitous computing environments. In UbiComp

2002: Ubiquitous Computing: 4th International Con-

ference G

oteborg, Sweden, September 29–October 1,

2002 Proceedings 4, pages 237–245. Springer.

Lee, J.-S., Su, Y.-W., and Shen, C.-C. (2007). A compara-

tive study of wireless protocols: Bluetooth, uwb, zig-

bee, and wi-ﬁ. In IECON 2007 - 33rd Annual Confer-

ence of the IEEE Industrial Electronics Society, pages

46–51.

Liu, J., Liu, H., Chen, Y., Wang, Y., and Wang, C. (2020).

Wireless sensing for human activity: A survey. IEEE

Communications Surveys & Tutorials, 22:1629–1645.

Montagut-Mart

ınez, P., Garc

ıa-Arenas, J. J., Romero-

opez, M., Rodr

ıguez-Rodr

ıguez, N., P

erez-Cruzado,

D., and Gonz

alez-Lama, J. (2022). Feasibility of an

activity control system in patients with diabetes: A

study protocol of a randomised controlled trial. Dia-

betes, Metabolic Syndrome and Obesity: Targets and

Therapy, pages 2683–2691.

Montoro Lend

ınez, A., L

opez Ruiz, J. L., Nugent, C., and

Espinilla Est

evez, M. (2023). Activa: Innovation in

quality of care for nursing homes through activity

recognition. IEEE Access, 11:123335–123349.

Mottram, S., Peat, G., Thomas, E., Wilkie, R., and Croft,

P. (2008). Patterns of pain and mobility limitation in

older people: cross-sectional ﬁndings from a popula-

tion survey of 18,497 adults aged 50 years and over.

Quality of Life Research, 17:529–539.

Niccoli, T. and Partridge, L. (2012). Ageing as a risk factor

for disease. Current Biology, 22:R741–R752.

Ramirez, H., Velast

ın, S., Meza, I., Fabregas, E., Makris,

D., and Farias, G. (2021). Fall detection and activ-

ity recognition using human skeleton features. IEEE

Access, 9:33532–33542.

Steele, R., Lo, A., Secombe, C., and Wong, Y. K. (2009).

Elderly persons’ perception and acceptance of using

wireless sensor networks to assist healthcare. Inter-

national journal of medical informatics, 78(12):788–

801.

Su, J., Liao, Z., Sheng, Z., Liu, A. X., Singh, D., and

Lee, H.-N. (2023). Human activity recognition us-

ing self-powered sensors based on multilayer bidirec-

tional long short-term memory networks. IEEE Sen-

sors Journal, 23(18):20633–20641.

Umbricht, D., Cheng, W.-Y., Lipsmeier, F., Bamdadian,

A., and Lindemann, M. (2020). Deep learning-

based human activity recognition for continuous ac-

tivity and gesture monitoring for schizophrenia pa-

tients with negative symptoms. Frontiers in Psychi-

atry, 11:574375.

United Nations (2022). World Population Prospects. Avail-

able online: https://population.un.org/wpp/ (accessed

on 14 May 2024).

Yuan, L., Andrews, J., Mu, H., Vakil, A., Ewing, R., Blasch,

E., and Li, J. (2022). Interpretable passive multi-

modal sensor fusion for human identiﬁcation and ac-

tivity recognition. Sensors (Basel, Switzerland), 22.

Perceptions and Acceptability of Sensor-Based Activity Recognition Systems Among Older Adults and Their Families

247