A Distributed IoT System for Real-Time Sports Performance Analysis in

Physical Education

Nelson Bilber Rodrigues

, Rui Jorge Ramos

, Mafalda Castro

, Nuno Jesus

, Pedro Guedes

Miguel Soares Ferreira

, Rafael Silva

and Lino Oliveira

INESC TEC, Rua Dr. Roberto Frias s/n, 4200-465, Porto, Portugal

Keywords:

Internet of Things, Educational Technology, Teacher Support Technology, Information Technologies.

Abstract:

Integrating Internet of Things (IoT) technologies into physical education (PE) presents opportunities for im-

proving the methodologies for collecting, analysing, and managing student performance data. However, it also

introduces technical challenges, particularly related to the real-time handling and protection of sensitive data

in dynamic training environments. This paper presents a comprehensive solution outline based on a private

local network architecture that supports scalable sensor data processing, real-time database integration, and

mobile application interfaces. The proposed distributed system ensures data integrity, low-latency communi-

cation, and secure access while enabling educators to monitor student performance in real-time and review

historical data. The system supports more personalised, data-driven training strategies by providing actionable

insights for sports education.

1 INTRODUCTION

Integrating software-driven Internet-of-Things (IoT)

solutions into sports and education has gained signif-

icant attention, providing new ways to process, anal-

yse, and manage student performance data in physical

activities (Rajaa et al., 2024). Advances in data sys-

tem architectures and processing frameworks have en-

abled the development of scalable, high-performance

systems that support educators in delivering more ef-

fective training programs. Teachers and coaches can

access structured performance data using software in-

frastructure, offering valuable insights into student

training progress, engagement levels, and general

well-being.

Despite this potential, effectively incorporating

IoT in educational sports settings presents several

challenges from the software development perspec-

https://orcid.org/0000-0002-0519-7151

https://orcid.org/0000-0002-9635-6815

https://orcid.org/0009-0007-8635-3147

https://orcid.org/0009-0004-1026-3186

https://orcid.org/0009-0008-5506-434X

https://orcid.org/0009-0002-4075-8070

https://orcid.org/0009-0002-0936-206X

https://orcid.org/0000-0003-1036-1072

tive, such as the demands of real-time data process-

ing, ensuring the security and privacy of sensitive in-

formation, and providing scalable infrastructure ca-

pable of handling high-frequency sensor inputs in dy-

namic environments. The following research question

emerges:

”How can a software architecture baseline be

outlined to process real-time data processing efﬁ-

ciency in IoT-based educational sports monitoring

systems?”

This paper presents a software architecture for an

IoT-enabled student training system designed to col-

lect, process, and visualise sports performance data

efﬁciently. The system consists of a private network

infrastructure for handling incoming sensor data and

a database for structured storage and retrieval of in-

formation in real-time with back-end and front-end

mobile applications that provide educators with ac-

tionable insights.

The proposed solution aims to optimise data in-

tegrity, improve system scalability, ensure privacy,

and enhance the accessibility of training analytics and

performance metrics, allowing teachers and coaches

to make informed decisions based on real-time and

historical data analysis.

In experimental trials, the system captured and

processed real-time data from low latency IoT sen-

230

Rodrigues, N. B., Ramos, R. J., Castro, M., Jesus, N., Guedes, P., Ferreira, M. S., Silva, R. and Oliveira, L.

A Distributed IoT System for Real-Time Sports Performance Analysis in Physical Education.

DOI: 10.5220/0013746400003988

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

In Proceedings of the 13th International Conference on Sport Sciences Research and Technology Support (icSPORTS 2025), pages 230-237

ISBN: 978-989-758-771-9; ISSN: 2184-3201

sors, demonstrating its ability to handle sensitive data

in practical scenarios.

2 RELATED WORK

2.1 IoT in Activity Monitoring

IoT technologies have transformed activity moni-

toring across diverse domains, including healthcare,

sports, ﬁtness, and industrial safety. IoT systems

enable continuous real-time tracking of physiologi-

cal and movement-related data by utilising wearable

sensors, smart devices, and interconnected networks.

These capabilities support a wide range of applica-

tions, such as motion analysis, fatigue detection, in-

jury prevention, and performance optimization. IoT

enhances education by collecting real-time data from

small wearable devices, enabling personalized and

adaptive learning. In addition, it supports educators in

monitoring student engagement, stress, and cognitive

states, leading to more effective and responsive teach-

ing strategies (Hern

andez-Mustieles et al., 2024).

Nevertheless, signiﬁcant challenges remain in in-

tegrating these technologies into existing infrastruc-

tures, mainly concerning economic accessibility, pri-

vacy, and data security (Rahmani et al., 2022). IoT

improves physical activity by providing real-time

feedback, enabling self-monitoring, promoting goal

setting, and supporting data-driven improvements in

performance. Yang et al. (Yang et al., 2024) grouped

the application of IoT in sports into the following sec-

tions: activity recognition and motion tracking, injury

prevention via fatigue/stress monitoring, performance

analytics and physiological variable prediction. How-

ever, the study executed by Raj

sp and Fister (Raj

and Fister, 2020) point out that the use cases missing

real-world validations and the scarcity of open, pub-

licly available datasets limit reproducibility and cross-

validation. Fresta et al. (Fresta et al., 2024) developed

a low-cost, end-to-end system architecture for human

activity data collection using an edge-cloud model.

The system captures data from sensors via Bluetooth

and utilizes a cloud-based, open-source framework to

collect, process, and distribute the information to end

users through stand-alone applications. Also, sup-

ports dynamic adjustment of key parameters like sen-

sor sensitivity and sampling rate, enabling adaptabil-

ity for various activity-tracking use cases.

2.2 IoT in Physical Education

In educational contexts, IoT technologies have been

shown to enhance student engagement, improve

learning outcomes, and enable instructors to design

more effective and personalized training programs

based on objective performance metrics (Verma and

and, 2018). Also, the study performed by Kassab

et al. (Kassab et al., 2019) highlights that the use

of IoT technologies enhances collaboration among

students, instructors, and staff. IoT supports vari-

ous learning principles and can provide diverse de-

livery modes, including face-to-face, online, and hy-

brid education. The common devices include smart-

phones, sensors, RFID tags, wearables, and remote

lab equipment. These are applied to attendance track-

ing, remote experimentation, personalized feedback,

and support for students with special needs. Despite

these beneﬁts, signiﬁcant challenges include security

risks, data scalability, and concerns over the dehu-

manisation of education.

In the context of sports-related education, Xu et

al. (Xu et al., 2024) report that the implementation

of IoT technologies contributes positively to the en-

hancement of physical education (PE) performance

among college students. However, their effectiveness

depends on students’ acceptance of the technology.

The study highlights that students are more inclined

to adopt and engage with IoT-enabled systems when

they perceive the devices as practical, user-friendly,

and conducive to an interactive learning experience.

Software solutions such as the IoT-IPSF frame-

work (Yang et al., 2021) integrate sensor-based mon-

itoring, web interfaces, and mathematical analysis -

it offers a scalable and accurate solution for modern-

izing PE through IoT. Li et al. (Li et al., 2022) used

artiﬁcial intelligence allied with IoT to enhance PE by

analysing real-time data from wearable devices. The

system monitors, classiﬁes, and predicts students’

physical activities, enabling personalized training and

performance optimization. Wu et al. (Wu et al., 2024)

integrated gesture recognition using wearable sensors

and multi-source data fusion supported by machine

learning algorithms. Basketball was the chosen activ-

ity for the trials, and the students strongly preferred

video-based learning content. The study from Tier-

ney et al. (Tierney et al., 2024) applies wearable de-

vices to football training to collect sensor data about

speed, walking distance, and heart rate. Also, the

authors notice the lack of intuitive, educationally fo-

cused software limits how wearable tech is applied

in learning, reclaiming better software tools to sup-

port students’ performance monitoring and evaluate

the learning outcomes. Wang (Wang, 2023) combines

IoT with mobile edge computing to help to improve

physical education by processing data locally at the

network edge, the system reduces latency, improves

responsiveness, and allows for real-time monitoring

A Distributed IoT System for Real-Time Sports Performance Analysis in Physical Education

231

of students’ physical activities. The developed so-

lution supports scientiﬁc management through accu-

rate monitoring, performance evaluation, and deﬁning

training plans. Also, it helps overcome infrastructure

and resource limitations in schools by decentralizing

computing and optimizing task distribution.

The integration of artiﬁcial intelligence with data

from wearable IoT devices and institutional learning

management systems has enabled the development of

systems capable of automatically generating person-

alized sports training recommendations for individual

students (Rajaa et al., 2024). The system was built

to protect sensitive personal health data. Privacy is

an important topic when sharing students’ health and

ﬁtness data collected via wearable biosensors during

public sports activities.

These privacy considerations underscore the need

for responsible data handling in IoT-enabled educa-

tional environments. Data-driven analysis of stu-

dents’ exercise habits and health metrics (Xu and Liu,

2023) enables tailored, personalized training to boost

teaching effectiveness, though ensuring data security,

managing equipment costs, and protecting user pri-

vacy is essential for ethical, scalable implementation.

Liu et al. (Liu et al., 2024) propose an edge-cloud

computing for IoT wearable devices that scramble the

sensitivity of data to securely embed private informa-

tion within visual data formats (e.g. ECG graphs or

movement visualizations) before sharing them over

networks.

Collecting data for statistical analysis in PE

presents several technical challenges. Capturing lim-

ited data may be insufﬁcient to accurately reﬂect stu-

dent performance, while delays in real-time data col-

lection can interfere with teachers’ decision-making.

In response to these limitations, Ding et al. (Ding

et al., 2023) propose a publish/subscribe model using

the Message Queuing Telemetry Protocol (MQTT)

in a client-server architecture. This architecture de-

sign overcomes data collection and transmission chal-

lenges by sending data to a cloud platform in JSON

format. The results demonstrate controlled perfor-

mance with low latency (averaging 65 milliseconds),

enabling teachers to monitor student performance

during exercises and make better real-time decisions.

Future directions recommended by the study per-

formed by Deng et al.(Deng et al., 2023), indicate

the IoT integration with AI for improving the person-

alised learning and ensuring data privacy and acces-

sibility. Also, emphasis focuses on developing cost-

effective, scalable solutions for massive adoption in

schools.

3 METHODS

A fundamental requirement of the proposed solution

is the capability to transmit substantial volumes of

data in near real-time, resulting from the simultane-

ous participation of multiple students wearing sensor-

equipped devices during physical activities. This dy-

namic environment introduces challenges that differ

from those encountered in static data collection sce-

narios. In addition to performance-related demands,

the designed solution needs to preserve the security

of sensitive data, adding another layer of complexity

to the system, which makes the task of specifying the

distributed components and their integration crucial

to the solution’s successful performance.

3.1 System Architecture

As an architectural inspiration for the solution, we use

the Lambda model (Kiran et al., 2015), a software de-

sign pattern for big data systems that uniﬁes batch

processing and real-time (stream) processing to pro-

vide both comprehensive historical insights and low

latency updates.

As illustrated in Figure 1, the system architecture

is structured into three key layers:

• The ﬁrst layer is composed of IoT devices, sported

by garments equipped with sensors, which collect

data on physical activity in real-time.

• The second layer is responsible for processing in-

formation, acting as the core infrastructure, host-

ing and executing various software components,

including the central database, WebSocket-based

data communication, post-session metrics pro-

cessing services, and the management Applica-

tion Programming Interface (API).

• The third layer, the presentation layer, two end-

user applications are designed to manage athletes

and groups, plan and monitor sports sessions,

and facilitate other functions related to managing

sports groups.

The self-contained system runs on a private net-

work, with no internet access to external servers,

thus maintaining the conﬁdentiality of the data and

preventing its transfer to external storage locations

within the institution.

3.2 Data Collection

The process begins by collecting data directly from

the source: athletes wearing the embedded sensors

in the textiles. The sensors transmit the collected in-

formation via WebSocket, integrated with RabbitMQ

icSPORTS 2025 - 13th International Conference on Sport Sciences Research and Technology Support

232

Figure 1: Software architecture diagram.

(RabbitMQ, 2025) queues, to a local low-end com-

puter (LLEC) responsible for communication and

managing data input. Embedded in sports apparel,

these hardware components continuously capture sen-

sor data from athletes in real time, such as ECG (Elec-

trocardiogram) voltage, acceleration, and orientation.

The data transmission process operates locally and

independently of the Internet, utilizing a private net-

work to ensure optimal performance and enhance data

security. Communication is facilitated through a lo-

cal router that is conﬁned to the pre-conﬁgured net-

work environment. All messages exchanged within

the system are formatted using JavaScript Object No-

tation (JSON), enabling structured, lightweight, and

platform-independent data representation.

3.3 Data Processing and Availability

Once the data is received by the dedicated service, it

is sent to the server, the component responsible for all

the data processing tasks. The server handles process-

ing and service hosting, including receiving sensor

data, storing it, processing it, and distributing the data

to mobile applications. A low-end computer was used

in the system setup. All services operate continuously

and are managed using Docker (Docker, 2025), which

provides containerization for consistent deployment,

scalability, and isolation across the system environ-

ment.

All incoming data is ﬁrst stored in the database

and subsequently transmitted to the front-end services

in real time, enabling educators to access and moni-

tor relevant information as activities unfold. Due to

the large amount of data received in real time, it is

crucial that the database fulﬁls criteria that guaran-

tee robustness, reliability, and low streaming latency.

For this reason, we chose RethinkDB (RethinkDB,

2025) as the database, an open-source solution that

can handle massive data ingestion in near real-time

while maintaining system integrity and performance.

Following its storage in the database, the information

is transmitted to the end-user services using the Web-

Socket communication protocol, in a similar way as

the data collection process from the athlete garment’s

sensors. This is achieved through a dedicated Web-

Socket server and the use of the websocket-ts library,

enabling real-time delivery to the User Interface (UI).

The participant identities are dynamically pre-

sented in real-time for educators, while the stored

dataset is encoded using anonymous numeric labels.

The mobile application, responsible for visualis-

ing the data, listens for incoming messages whenever

there is an active sports activity. In this live scenario,

the loss of data in real time is not problematic because

the metrics calculated are used by the teacher to mon-

itor the course of the session. The crucial aspect is

storage in the database, because more complex met-

rics are calculated after the session, with all the data

collected during it, resulting in post-session reports

per student.

3.4 End-User Services

The end-user interface consists of two distinct mobile

applications, developed using the IONIC Framework

(Ionic, 2025), which enable real-time activity moni-

toring and post-session data analysis. The two appli-

cations share a similar tabs layout, designed to eas-

ily navigate and group the main sections of content,

which can also be collapsed to increase the size of the

content in mobile devices.

The back-ofﬁce application, AURORA Studio, as-

sists the coach in managing athletes, groups, such as

classes or teams, sports sessions, as well as handling

other contextual information within the system. Once

a set of athletes are assigned to a group, the user can

schedule a session for that group in a given date, con-

sisting of a sequence of sports activities, and a set

A Distributed IoT System for Real-Time Sports Performance Analysis in Physical Education

233

Figure 2: AURORA Studio mobile application.

of metrics to be calculated for each sport, as shown

in Figure 2. After a session, the user may also use

the app to trigger the processing of the data generated

during its time frame. Once the processing of the ses-

sion data is completed, the user can generate reports

and perform analytics over this data.

The front-ofﬁce application, AURORA (Figure 3),

processes data in real time, helping to track and mon-

itor training sessions previously planned in the back-

ofﬁce application. In this application, a teacher or

coach can check the state of the wearable data sources

and assign them to the athletes of a session. After be-

ginning a session with the selected group and devices,

the user is able to start, pause, or stop the activities

that were planned, as the session is ongoing. Dur-

ing the activity, the teacher/coach monitor the perfor-

mance of multiple athletes simultaneously, through a

set of simple metrics displayed on the screen, both

numerical and categorical: Heartbeat Rate, Exercise

Intensity, Cadence and Activity State.

By selecting a speciﬁc athlete, the user can also

visualize this data in different representations, such

as charts, to monitor the variation of these values

through time. These charts are updated as the data is

received through repeated calls of functions that per-

form the required calculations. In this way, the user

can monitor some athletes’ physical indices during

their sporting activity.

A Representational State Transfer (REST) API

was designed to control all system-related conﬁgura-

tions and data, including managing athletes, groups,

activities, and sessions. It serves as the primary access

point for retrieving detailed information across the

data ecosystem, and its implementation is based on

the Flask framework (Flask, 2025) for minimal web

applications. Regarding sports metrics, they are pro-

cessed in two stages, accounting for the metric type

calculated. Real-time metrics, such as the heartbeat

rate or velocity of the athlete, are processed as the

data is collected, at the front-ofﬁce application. On

the other hand, post-session metrics are calculated af-

ter the sports session due to the complexity of these

processing algorithms and the data size involved, on

the server.

4 RESULTS

The proposed IoT-based monitoring system was eval-

uated in a controlled physical training environment

using a sensor-embedded garment conﬁgured to trans-

mit data over a secure private network.

The main objective of the evaluation was to assess

the system’s ability to reliably capture and process

real-time performance data with minimal latency.

Preliminary ﬁndings emerged during the trials, in-

volving a small group of students from a sports uni-

versity. The results demonstrate the system’s effec-

tiveness and reliability in capturing and analyzing

real-time student performance data. Figure 4 presents

quantitative performance metrics focusing on mes-

sage transmission latency between the IoT device and

the mobile application, over a span of 5 minutes.

The IoT device transmitted messages at ﬁxed in-

tervals of 250 milliseconds. Latency measurements

were conducted across the system architecture, de-

icSPORTS 2025 - 13th International Conference on Sport Sciences Research and Technology Support

234

Figure 3: AURORA mobile application.

Figure 4: Message latency from IoT to Mobile Application.

ﬁned as the time elapsed between data acquisition by

the sensor and its reception by the mobile application.

The system exhibited stable performance, with mes-

sage transmission latency consistently ranging from

1.8 to 2.2 seconds. With some spikes observed, reach-

ing up to 3.2 seconds. These ﬁndings indicate that

the system is capable of supporting near real-time

data handling requirements, providing end users with

timely access to critical performance metrics.

5 DISCUSSION

The use of message brokers to manage the queues

with massive data from sensors, produce results that

are in line with ﬁndings from Ding et al. (Ding

et al., 2023), whose MQTT-based system reported

sub-second latencies under ideal conditions.

The Lambda Architecture adopted in this work en-

abled the uniﬁcation of batch and stream processing

layers, which proved beneﬁcial for maintaining both

real-time monitoring and comprehensive post-session

analysis. This dual capability is particularly important

in educational settings, where teachers beneﬁt from

A Distributed IoT System for Real-Time Sports Performance Analysis in Physical Education

235

both immediate feedback and detailed retrospective

insights.

Compared to prior IoT-based physical education

systems, e.g., the work from Fresta et al. (Fresta

et al., 2024), which relied on cloud processing, the

proposed architecture uses a local, private network

coupled with Docker-based containerisation, provid-

ing strong performance isolation and reduced depen-

dency on internet connectivity. This setup also con-

tributes to improved data security — an increasingly

critical concern noted by Xu and Liu (Xu and Liu,

2023) and Liu et al. (Liu et al., 2024), who empha-

sis the importance of safeguarding students’ personal

health information in real-world deployments.

Nevertheless, some latency spikes were observed.

While they did not impact overall functionality, they

suggest potential areas for improvement in communi-

cation protocol optimization or network load balanc-

ing.

6 CONCLUSIONS AND FUTURE

WORK

This paper presented a distributed IoT-based software

architecture designed to enable real-time monitoring

and analysis of student performance data in physi-

cal education settings. By leveraging a private local

network, Docker-based service deployment, and the

Lambda Architecture model, the system successfully

integrated real-time data collection, processing, and

visualization within a secure environment.

Experimental trials demonstrated that the system

achieved consistent message transmission latencies

between 1.8 and 2.2 seconds, with occasional spikes

up to 3.2 seconds. These preliminary results demon-

strated that the architecture is robust and reliable,

managing considerable amounts of sensor-generated

data with low latency. The system provided near-

real-time data visualization, using a private network,

ensuring security and privacy, and addressing critical

ethical and data protection concerns in the context of

student information handling.

The proposed solution, also empowers educators

through intuitive mobile applications that facilitate

session planning, live monitoring, and post-session

analytics. Offering an educator-focused platform that

aligns with pedagogical goals and privacy standards.

Future research will focus on advancing privacy-

preserving mechanisms within edge computing

frameworks. In particular, the integration of encryp-

tion and anonymization data processing techniques to

ensures security while maintaining system scalability

and usability.

ACKNOWLEDGEMENTS

This work is co-ﬁnanced by Component 5 - Capital-

ization and Business Innovation, integrated in the Re-

silience Dimension of the Recovery and Resilience

Plan within the scope of the Recovery and Resilience

Mechanism (MRR) of the European Union (EU),

framed in the Next Generation EU, for the period

2021 - 2026, within project TEXPACT, with reference

61.

REFERENCES

Deng, C., Feng, L., and Ye, Q. (2023). Smart physical ed-

ucation: Governance of school physical education in

the era of new generation of information technology

and knowledge. Journal of the Knowledge Economy,

15(3):13857–13889.

Ding, Z., Mei, J., and Zheng, K. (2023). An mqtt-based stu-

dent condition monitoring system for physical educa-

tion. In 2023 IEEE International Conference on Ad-

vanced Learning Technologies (ICALT), pages 353–

355. IEEE.

Docker (2025). Docker - accelerated container application

development. https://www.docker.com/. Accessed:

2025-03-12.

Flask (2025). Flask - a lightweight wsgi web application

framework. https://ﬂask.palletsprojects.com/en/stable

/. Accessed: 2025-03-12.

Fresta, M., Dabbous, A., Bellotti, F., Capello, A., Lazza-

roni, L., Pighetti, A., and Berta, R. (2024). Low-

Cost, Edge-Cloud, End-to-End System Architecture

for Human Activity Data Collection, pages 444–449.

Springer Nature Switzerland.

Hern

andez-Mustieles, M. A., Lima-Carmona, Y. E.,

Pacheco-Ram

ırez, M. A., Mendoza-Armenta, A. A.,

Romero-G

omez, J. E., Cruz-G

omez, C. F., Rodr

ıguez-

Alvarado, D. C., Arceo, A., Cruz-Garza, J. G.,

Ram

ırez-Moreno, M. A., and Lozoya-Santos, J. d. J.

(2024). Wearable biosensor technology in education:

A systematic review. Sensors, 24(8).

Ionic (2025). Ionic - a mobile sdk for the web. https://ioni

cframework.com/. Accessed: 2025-03-12.

Kassab, M., DeFranco, J., and Laplante, P. (2019). A sys-

tematic literature review on internet of things in edu-

cation: Beneﬁts and challenges. Journal of Computer

Assisted Learning, 36(2):115–127.

Kiran, M., Murphy, P., Monga, I., Dugan, J., and Baveja,

S. S. (2015). Lambda architecture for cost-effective

batch and speed big data processing. In 2015 IEEE In-

ternational Conference on Big Data (Big Data), pages

2785–2792.

Li, Q., Kumar, P., and Alazab, M. (2022). Iot-assisted

physical education training network virtualization and

resource management using a deep reinforcement

learning system. Complex & Intelligent Systems,

8(2):1229–1242.

icSPORTS 2025 - 13th International Conference on Sport Sciences Research and Technology Support

236

Liu, P., Shi, D., Zang, B., and Liu, X. (2024). Students

health physique information sharing in publicly col-

laborative services over edge-cloud networks. Journal

of Cloud Computing, 13(1):98.

RabbitMQ (2025). Rabbitmq - one broker to queue them

all. https://www.rabbitmq.com/. Accessed: 2025-05-

25.

Rahmani, A. M., Szu-Han, W., Yu-Hsuan, K., and Hagh-

parast, M. (2022). The internet of things for ap-

plications in wearable technology. IEEE Access,

10:123579–123594.

Rajaa, J. M., Manee, M. J., Mohammed, M. A., Mo-

hammed, M. A., and Sharif, S. F. A. (2024). Iot-driven

physical training and ﬁtness management in colleges.

In 2024 International Conference on Emerging Re-

search in Computational Science (ICERCS), pages 1–

Raj

sp, A. and Fister, I. (2020). A systematic literature re-

view of intelligent data analysis methods for smart

sport training. Applied Sciences, 10(9).

RethinkDB (2025). Rethinkdb - an open-source database

for the realtime web. https://rethinkdb.com/. Ac-

cessed: 2025-03-12.

Tierney, P., Clarke, N., and and, S. R. (2024). Use and

application of wearable technology in football further

education settings in the uk. Sport, Education and

Society, 0(0):1–14.

Verma, P. and and, S. K. S. (2018). Internet of things-

based student performance evaluation framework. Be-

haviour & Information Technology, 37(2):102–119.

Wang, Y. (2023). Optimal application of intelligent iot in

school sports teaching management based on resource

coordination and mobile edge computing. Interna-

tional Journal of System Assurance Engineering and

Management.

Wu, D., Guo, Z., Wang, Y., and Li, Z. (2024). Auxiliary

analysis of digital platform using internet of things

technology in physical education teaching. Education

and Information Technologies, 29(12):15855–15874.

Xu, Y. and Liu, M. (2023). Data collection and analysis

in physical education practical teaching based on in-

ternet of things. International Journal of Information

Technology and Web Engineering, 18(1):1–15.

Xu, Y., Peng, J., Jing, F., and Ren, H. (2024). From wear-

ables to performance: how acceptance of iot devices

inﬂuences physical education results in college stu-

dents. Scientiﬁc Reports, 14(1):23776.

Yang, L., Amin, O., and Shihada, B. (2024). Intelligent

wearable systems: Opportunities and challenges in

health and sports. ACM Comput. Surv., 56(7).

Yang, L., D

ıaz, V. G., and Kumar, P. M. (2021). Internet of

things-based intelligent physical support framework

using future internet of things. Technology and Health

Care, 29(6):1187–1199. PMID: 34092670.

A Distributed IoT System for Real-Time Sports Performance Analysis in Physical Education

237