Research Journal of Humanities and Cultural Studies Vol. 5 No.

1 2019 ISSN 2579-0528

www.iiardpub.org

New Technologies in Sport through the Internet of Things

Systems

Davide Di Palma, Pompilio Cusano, Carmine Russo, & Antonio Ascione

Department of Movement and Well-being Sciences,
University of Naples Parthenope,
Italy
[email protected], [email protected]

Abstract
In recent years, the Internet of Things (IoT) technology has revolutionized the global market
in the sports sector. It is interesting to learn about digital technological innovations and their
connection to athletic performances. The use of these technologies is largely developed in the
performance evaluation and monitoring; new cutting-edge equipment allows sports operator
identifying parameters and indexes of considerable interest for the sports performance. The
application of new technologies allows researchers investigating and increasing the focus of
the research activity on the new possibilities of objectively observing aspects concerning the
match-analysis, such as wearable technologies, motion capture systems, and other systems
allowing professionals implementing new training methods through the use of tools connected
to reality by applying IoT technology-based systems.

Keywords: Systems, Sports, Match-Analysis, Parameters, Technology, Application.

Introduction
The conception of the Internet of Things refers to the extension of the Internet to concrete
objects and places. This is the evolution of the network through which connected objects enter
into communication with other objects (such as mobile phones), to which they transfer their
data. The IoT makes it possible to manage remotely (by monitoring and controlling) commonly
used objects such as refrigerators and other household appliances, and makes it possible to
carry out daily activities completely automatically, such as managing the home temperature.
In short terms, the Internet of Things allows objects taking an active role by connecting to the
network.

Potentially, the Internet of Things can be applied in countless sectors. If for some of them, such
as home automation for the control of one's own home and the remote use of air
heating/conditioning systems, household appliances and window frames, this is a practical
reality, for some others this process is still under development, such as in the automotive
industry for which we are currently speaking of automated guided cars. For some years now,
the production of sports equipment has also shown great interest in the development of
technologies connected to the IoT; more specifically, interest in the in-depth analysis of the
performance monitoring systems, able to make its measurement more immediate, has
increased. This is possible by means of both wearable and performance observation
instruments.

Relationship between Performance and IoT systems

Leaders in the Internet of Things (IoT) applications progresses, concerning the analysis of the
athletic performance, have been able to understand something significant: more than ever,
today's technology can be useful to the mind and the human body, because athletes can

IIARD – International Institute of Academic Research and Development Page 16

 Research Journal of Humanities and Cultural Studies Vol. 5 No. 1 2019 ISSN 2579-0528
www.iiardpub.org

mentally correlate their perceived performance (how quickly they think they are moving, how
tired they think they are) with respect to their actual performance (based on real and instant
data). The use of innovative technologies, such as a wearable sensor, allows coaches and team
personnel monitoring the information on the player, such as his movement, heartbeat and other
parameters useful for his health state and performance. This information, combined with
player’s location and environmental data, is typically sent to a secure, cloud-based analysis
platform that provides critical information to team personnel via a dashboard. By informing
about the best player interventions and improving tactical game decisions, the IoT solution
helps provide significant competitive advantage to teams. In addition, the user interface can be
designed to allow for a better overall health management, recovery and elimination of injuries,
extending benefits far beyond a single game. The potential of this approach is so high that even
a market leader, such as IBM, has started several activities in this area. For example, it set up
a collaboration with the USA Cycling team through their IoT service, in the framework of a
project called "IBM jStart". Through their service, they provide information to cyclists
immediately after a training session and in real time during a competition, by means of smart
glasses. These real-time feedbacks and data improve the efficiency of the cyclists, drive them
on when to cycle faster, and help them avoid the much feared “bonk ", i.e. the moment when
the reservoir, intended as energy storage, has no more fuel. Wearable technologies are starting
to be a turning point for coaches, athletes and fans. The ability to measure and track
performance data and take advantage of the analysis will ultimately allow teams being more
sophisticated about their game plan, athletes' performance and injury prevention.

Therefore, some key factors behind the competitive advantage of the IoT technology applied
to athletic performance have been recognized:
 Track power; it measures strength production during flight in real time, providing data to
maximize the athlete's live performance.
 Energy Tracking; this allows athletes better understanding the production and use of energy
in order to optimize training load and recovery, since there are metrics (such as muscular
oxygen) that are unique for the individual, and can be related to many physiological
conditions.
 Wide Eyes; by using an optical sensor to observe large sizes (such as a microchip in an
American football helmet), a high-performance essential product can encompass wearable
technologies and include innovative heads-up display and audio technologies.
 Performance Analysis Structure; computer interfaces first convert raw data into user-
friendly messages, then automatically send messages almost in real time to an IoT; for more
immediate feedback and integration with other data feeds (power, heart rate, etc.), in order
to highlight correlation and advanced analysis, it is possible to derive value from things.
Through the IoT analysis, team members can get a clear analytical framework of what they
have to do to win. The IoT now allows them reading both structured and unstructured data
to create a comprehensive overview.

Further insights from which to consider that the team can calculate the metrics in real time,
will allow the team coaches monitoring the performances not only after the training session,
but while it is still in progress. It can also collect and share IoT data in a virtual cloud database:
the cloud solution removes the complexity of mobile app development and management, and
allows teams focusing on the needs and demands of the user dynamic mobile; this means
having relevant information at the right time with the ability to act immediately. Since this new
training solution is mobile and cloud-based/IoT platform-based, teams can train anywhere and
benefit from real-time monitoring, analysis and communication. By bringing together not only

IIARD – International Institute of Academic Research and Development Page 17

 Research Journal of Humanities and Cultural Studies Vol. 5 No. 1 2019 ISSN 2579-0528
www.iiardpub.org

tools to connect wearable devices but also data and analysis services, providers can create
personalized experiences for end users, drive progress and reduce costs. While using their
mobile app, coaches can also access training sessions in real time, at any time and from
anywhere, allowing for a remote involvement when everyone cannot be in the same place at
the same time. This provides a lot of flexibility for the team of professionals who are constantly
on the move.

At this point we can understand that, in today's sports training and performance analysis, almost
all the performances are captured on a video or via IoT sensors. The captured videos and sensor
data are then displayed by expert instructors/analysts. It is in this field that platforms and
software applications play the most important role, in a range that goes from reducing or
eliminating time to manually taking note of and labeling important performance indicators to
evaluate performance, computer-aided self-training systems for sports exercise, and tactical
information extraction alongside periodic event detection, by also taking advantage of virtual
reality and acceleration sensors on the body to perform motion and force analysis.

Motion Capture System and Biomechanics

Motion Capture (MoCap) is another performance monitoring system, and is also defined as
"the practice of dynamically recording positions of various predefined points on the body or
object in 3D space". In addition to general studies on motion, biomechanics intends to measure
and relate it to the underlying musculoskeletal function. This provides information for the
performance of a single athlete, but it can also provide information on the health status of the
individual in relation to his musculoskeletal system. The whole model is known as the
biomechanical model. The complexity behind these models is not only linked to the
methodology underlying the analysis, but also to the applied technology: a very wide range of
technologies is available for the acquisition of motion, thus the result differs significantly from
each system. Optical tracking devices employ multiple infrared-sensitive cameras to track the
positions of the "markers" attached to the objective (to the body joints, for example). The
"passive" systems illuminate the scene with infrared light and detect the positions of retro-
reflecting markers. "Active" systems use infrared light-emitting diodes as markers, but they are
similar anyway. Provided that the positions of the cameras in space are known (i.e., if the
system is calibrated) the 3D position of each marker can be retrieved by analyzing its position
in the image of each camera (at least two are needed to recover the 3D position of a marker).
These systems often work in conjunction with the IoT sensors, but in this specific field of
analysis, they have some constraints; the non-functional requirements of the system are mainly
related to its wearability: the system must be portable and wearable, and easily usable as part
of the athlete's ordinary training program. This limits the range of usable sensors, since they
must be non-invasive and minimally influenced by physical and biological responses, such as
vibrations and sweating. The size and weight of the system must also be minimized, avoiding
any performance limitations or movement. It should also have a sufficient power supply, at
least for a full training session, while wearability also implies that information between the
athlete and the system's main control unit should be transmitted wirelessly. In addition to
technologies that capture human and object movement, eye and gaze tracking technology can
also be used in sports training: gaze behavior can differ significantly between experts and
beginners (for example, when keeping track of a ball), bad habits and unconscious actions
conditioning performance can also be identified by monitoring the athlete's eyes during the
game. Furthermore, technique and analysis are very different depending on the methodology
employed. For example, animation needs a continuous and constant signal, but it does not need
an individual’s physiological precision. The latter would be essential for clinical science, in
which motion capture must be integrated with other systems, such as force measurements and
IIARD – International Institute of Academic Research and Development Page 18
 Research Journal of Humanities and Cultural Studies Vol. 5 No. 1 2019 ISSN 2579-0528
www.iiardpub.org

electromyography. This is necessary to not only measure motion, but to analyze and understand
the reason of the movement. This is why the general motion analysis is usually applied in the
field, while the clinical motion analysis is more restricted to specialized laboratories that
consist of complex technologies and specialized professionals to manage them. MoCap
systems have been developed over the past two decades to track and record human movements
at high spatial and temporal resolutions. These applications require efficient methods and tools
for the analysis, synthesis and automatic classification of motion capture data.

The main challenge in articulating the movement of the articulated body is the large number of
degrees of freedom (about 30) to be recovered. Research has generally focused on two
opposing approaches: one consists in introducing hypothesis constraints on motion trajectories
or visualization restrictions, both by labeling and using markers or color code. The other
approach consists in relaxing the constraints deriving from articulation, and following the limbs
as if their movements were independent. In both cases, the analysis is challenging due to the
technological and practical difficulties associated with the resolution and accuracy of 3D video
analysis across large volumes. Since even a hundredth of a second improvement between two
different movements could be significant for the outcome of the gesture, improving
performance for elite athletes could result in technical adjustments that go beyond the
capabilities of video-based systems, and have to rely on the coach-athlete intuition instead.

To solve this complexity, an innovative system called motion capture fusion (MCF) has been
developed. It is able to capture3D kinematics and kinetics by overcoming the technological
difficulties associated with the monitoring of the athlete’s performance in an elite environment.
MCF is a term that describes the motion capture when different streams of different data are
merged to measure the athlete's movement: inertial measurement unit (IMU), global
positioning system (GPS), pressure sensitive insoles and video measurements have been
combined. The core of the MCF is the fusion of IMU and GPS data: the IMU typically contains
accelerometers, gyroscopes, magnetometers and a thermometer, and tracks the local orientation
and acceleration of each segment of interest. GPS data are merged with local acceleration data
to track the athlete's overall trajectory. In this way, the software is able to draw a complete
picture of the athlete's gesture, highlighting also the different vectorial forces and the
momentum of each part of the body during the activity. In addition, the most advanced software
also includes an eye-mark recorder, a device to measure a subject's focus point and his pupil
diameter, which is able to draw a precise single-eye or double-eye movement of vision lines.
These data can then be synchronized with the motion capture systems, in order to display them
both in real time and in post-processing.

Notational Analysis and Tracking

Both during training sessions and during the game, efforts on tracking the player's movement
in the field have traditionally involved a series of data collection techniques, from live
observation to post-event video analysis, where the players' patterns of movement are
registered manually and classified to determine the effectiveness of the performances. This
process is called Notational Analysis and involves the subjective quantification of single
players' movements by an investigator, and the frequency and timing of particular movements
are therefore linked to their relative success. Obtaining accurate positional information about
sports players is interesting for coaches and high performance support teams, because of the
ability to correlate performance with tactics and help planning better training programs. The
complex nature of the movement inherent in many physical activities is also a significant
obstacle to overcome: athletes tend to produce rapid and agile movements, with many
unpredictable changes in direction and frequent collisions with other players. Each of these
IIARD – International Institute of Academic Research and Development Page 19
 Research Journal of Humanities and Cultural Studies Vol. 5 No. 1 2019 ISSN 2579-0528
www.iiardpub.org

characteristics of the player's behavior violates the hypotheses of fluid motion, on which
computer tracking algorithms are generally based. However, the increasing ability of digital
technology to collect, manage and organize video images has allowed improving the specific
current analytical procedures for sport: these are the localization systems based on the
automated vision, which consist of a variety of methods used to analyze the athletes' movement
during sports in which the movements vary in duration, field position and surface, speed,
technique of direction and tactics. Unlike manual visual tracking systems, automatic motion
tracking does not require human operators to manually locate and continuously record the
position of the tracked object. There are many potential applications of an automatic motion
detection system, such as: tactical planning and strategies, measurement team organization,
provision of significant kinematic feedback, and objective measures of the effectiveness of the
intervention in the sports team, which could provide benefits to coaches, players and sports
scientists.

Object tracking is widely used in sports analysis. Balls and players are the most frequently
monitored subjects, since significant events are mainly caused by player-player interactions.
The most common tracking techniques include the trajectory-based analysis and tracking of
the small balls, the reconstruction of the physical model-based 3D trajectory, and the estimation
of the 3D position with multiple cameras.

The visual signals used for the highlights detection are the movement of the ball, the game field
zone, the players' positions and the colors of the uniforms: for example, a model can use the
dominant chromatic ratio and the intensity of the movement to map the structure of a specific
sports video, based on the syntax and characteristics of the video content, since in a sports
game the positions of the cameras are generally fixed and the rules for showing the game
progress are similar in different channels. Other algorithms are able to exploit the markers lines
on the ground to determine the calibration parameters: they are based on a specialized court
line detector followed by a combinatorial optimization phase to locate the court within the set
of line segments detected, and an iterative phase of court model tracking. By taking advantage
of these properties, object tracking methods are able to perform a classification based on mid-
level representations, including a vector field model in motion, physics-based algorithm, color
tracking model, and shooting rhythm model. With the statistical chart for ball movements, the
coach is able to observe the game actions distribution at a glance, and quickly understand where
players can be more effective. As for object tracking strategies, many shot classification
methods are proposed on the basis of camera motion, color information, weave information or
face detection. Most of the existing shot classification approaches follow the same procedure
too, focusing on grouping key photograms or shots with similar low-level features. In addition,
the bullet classification scheme can employ a supervised learning process to perform a video
shot classification from top to bottom. This scheme is made up of three main phases:
identifying the video shot classes for each sport; developing a common set of representations
of the movement, color, dots related to length, and supervised learning of sports video data.
Thanks to space analysis for mapping low-level functions to the semantic attributes of mid-
level video recording (such as player movement, camera motion patterns or field shape, etc.),
a match analyst can combine these mid-level shot features and classify the videos into a small
number of predefined shots classes, which are able to cover a 90% -95% of the broadcasted
sports videos. A recently introduced system has been named 3DRSBA: it stands for 3D Remote
Sports Biomechanics Analysis, and its purpose is to establish a remote biomechanical service
for athletes' health screening. "Remote" means that, by combining existing high-end 3D
acquisition systems with modern communication tools, such as high-speed Internet, cloud
storage and the new 4G mobile network, biomechanics studies can be carried out outside the
IIARD – International Institute of Academic Research and Development Page 20
 Research Journal of Humanities and Cultural Studies Vol. 5 No. 1 2019 ISSN 2579-0528
www.iiardpub.org

training sphere, without losing the medical support that a motion laboratory can provide. This
is a response to one of the main obstacles to laboratory studies: the limited availability of
biomechanically-trained technical personnel and trainers. Currently, the professional training
of these professionals is extensive and requires many years of experience; therefore, the
availability of such experts is limited. In addition, a complete analysis session in the laboratory
is an interruptive event in the already short program of an elite athlete with a professional
training plan and a life. On the other hand, the laboratory approach provides high technical
quality outcomes, with carefully controlled parameters. Therefore, the ability to combine the
power of 3D motion capture in biomechanical laboratories with the needs of sports analysis to
provide an environment close to actual measurement, is the real key factor that is driving the
success of 3DRSBA systems.

Conclusions
Ultimately, we believe that new coaching and training concepts can become interactive through
the use of new technologies. This concept offers the trainer the opportunity to concentrate on
new specific areas within the training, in spatial and temporal terms, able to provide users with
different points of view with respect to performance; moreover, it is important both for
evaluation and purely didactic purposes that sportsmen must learn, study and know how to use
these new forms of sharing performance, and focus more on the development of increasingly
innovative solutions to transmit more information to athletes and improve sports practice.

References
Alam, S., Chowdhury, M. M., & Noll, J. (2010, November). Senaas: An event-driven sensor
virtualization approach for internet of things cloud. In Networked Embedded Systems
for Enterprise Applications (NESEA), 2010 IEEE International Conference on (pp. 1-
6). IEEE.
Anzaldo, D. (2015, November). Wearable sports technology-Market landscape and compute
SoC trends. In SoC Design Conference (ISOCC), 2015 International (pp. 217-218).
IEEE.
Ashton, K. (2009). That ‘internet of things’ thing. RFID journal, 22(7), 97-114.
Atzori, L., Iera, A., & Morabito, G. (2010). The internet of things: A survey. Computer
networks, 54(15), 2787-2805.
Borgia, E. (2014). The Internet of Things vision: Key features, applications and open
issues. Computer Communications, 54, 1-31.
Castillejo, P., Martínez, J. F., López, L., & Rubio, G. (2013). An internet of things approach
for managing smart services provided by wearable devices. International Journal of
Distributed Sensor Networks, 9(2), 190813.
Davidson, N. (2002). Internet gambling: Should fantasy sports leagues be prohibited. San
Diego L. Rev., 39, 201.
Fernandez, F., & Pallis, G. C. (2014, November). Opportunities and challenges of the Internet
of Things for healthcare: Systems engineering perspective. In Wireless Mobile
Communication and Healthcare (Mobihealth), 2014 EAI 4th International Conference
on (pp. 263-266). IEEE.
Hutchins, B., & Rowe, D. (2012). Sport beyond television: The internet, digital media and the
rise of networked media sport. Routledge.
Mainetti, L., Patrono, L., & Stefanizzi, M. L. (2016, July). An Internet of sport architecture
based on emerging enabling technologies. In Computer and Energy Science (SpliTech),
International Multidisciplinary Conference on (pp. 1-6). IEEE.
Paschou, M., Sakkopoulos, E., Sourla, E., & Tsakalidis, A. (2013). Health Internet of Things:

IIARD – International Institute of Academic Research and Development Page 21

 Research Journal of Humanities and Cultural Studies Vol. 5 No. 1 2019 ISSN 2579-0528
www.iiardpub.org

Metrics and methods for efficient data transfer. Simulation Modelling Practice and
Theory, 34, 186-199.
Ray, P. P. (2015, December). Generic Internet of Things architecture for smart sports.
In Control, Instrumentation, Communication and Computational Technologies
(ICCICCT), 2015 International Conference on (pp. 405-410). IEEE.
Ray, P. P. (2015). Internet of Things for Sports (IoTSport): An architectural framework for
sports and recreational activity. Proceeding of IEEE EESCO, Vizag, 79-83.
Rodríguez-Molina, J., Martínez, J. F., Castillejo, P., & López, L. (2013). Combining wireless
sensor networks and semantic middleware for an internet of things-based
sportsman/woman monitoring application. Sensors, 13(2), 1787-1835.
Wei, J. (2014). How Wearables Intersect with the Cloud and the Internet of Things:
Considerations for the developers of wearables. IEEE Consumer Electronics
Magazine, 3(3), 53-56.
Wilson, B., & Hayhurst, L. (2009). Digital activism: Neoliberalism, the Internet, and sport for
youth development. Sociology of sport journal, 26(1), 155-181.
Ye, Q., Wang, Z., Qian, J., Gao, S., & Sun, Y. (2012). Smart sport-emergence trend of sport
information. International Journal of Digital Content Technology and its
Applications, 6(11).
ZHAO, H. Q., SUN, J., HUA, Y. M., & JIN, J. C. (2008). A Study of Data Mining Technology
Applied on Technique and Tactic Analysis of Sport Competitions. Journal of Beijing
Sport University, 5, 046.

IIARD – International Institute of Academic Research and Development Page 22