Soft Computing (2023) 27:10145–10154

https://doi.org/10.1007/s00500-023-08242-5 (0123456789().,-volV)(0123456789().
,- volV)

FOCUS

Intelligent system simulation and data accuracy of physical fitness

training for sports majors based on real-time status update
of wearable Internet of Things
Zhou Yong1

Accepted: 15 April 2023 / Published online: 2 May 2023

Ó The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2023

Abstract
With the development of modern science and technology, sports training has begun to develop in the direction of new
ideas, new technologies and new equipment. Wearable devices with wireless data transmission and real-time feedback
functions are more and more widely used in sports training, providing a simpler and more scientific way to explore the
biological significance of sports training. In this paper, the wearable Internet of Things is applied to the intelligent system
simulation of physical training for sports majors. With the support of the Internet of Things technology, wearable devices
can quickly capture the movement data of students’ physical training and quickly transmit this information to the intel-
ligent system of physical training for sports majors, so that the system can be updated in real time. Wearable devices are
high-tech products developed in the context of big data. The use of wearable devices to achieve sports training and data
collection is only an external manifestation of sports scientific monitoring. The collection and statistics of the original
sports data obtained through more scientific information processing technology is the focus of sports scientific monitoring.
At present, the use of wearable devices in sports training also has some measurement errors, which affect the validity and
credibility of data. Data processing is not systematic and perfect, and the compatibility and connectivity are low. Statistical
analysis is a common way to analyze sports training data at home and abroad. This paper discusses the physical training
data based on the status update of the wearable object network and uses the optimized data processing method to improve
the accuracy of the data.

Keywords Wearable  Internet of Things  Real-time monitoring and updating  Physical training system

1 Introduction feedback, especially football, basketball, hockey and other

sports with complex forms, huge changes in human body
With the rapid development of competitive sports, strong load and the inability to obtain quantitative information,
international competition, science and technology to help have become more widespread and in-depth, greatly pro-
sports become more and more urgent (Jones et al. 2017). moting the progress of scientific physical training level of
Sports training is further expanding toward new concepts, competitive sports (Hill and Leiszler 2011; Pirbhulal et al.
new technologies and new equipment. Internationally, as 2019). In China, wearable devices are also widely used in
the focus of research and development of cutting-edge competitive sports, especially in collective ball games,
sports monitoring methods, competitive sports that use while there are few relevant research reports. There are
mobile Internet and cloud platforms and have the ability of certain differences between the research and actual appli-
wireless data transmission and high-speed information cation of wearable Internet of Things equipment in China’s
sports training, so it is urgent to carry out research and
discussion and summarize the rules (Xiao et al. 2020). At
& Zhou Yong the same time, the application of wearable devices in sports
zhouyong88661@163.com training still faces some shortcomings. Its calculation
1 accuracy, communication mode and the stability and reli-
Chongqing Creation Vocational College, Chongqing 402160,
China ability of its operation have great prospects for research

123
10146 Z. Yong

and promotion. Coaches and players must strengthen their training methods, as well as the feasibility of implementation
grasp and understanding of this technical field from both (Yang et al. 2020). In the literature, data mining technology
the principle and practice aspects to avoid that theory has was used to redesign the athlete training evaluation system,
always been ahead of practice (Kong et al. 2020). There- and appropriate modifications were made according to the
fore, this paper will study and summarize the relevant characteristics of different algorithms. The application of
application research of wearable devices in sports training, data mining technology in the athlete training mode decision
and summarize the current situation of their use, and apply support system was further studied (Marinov et al. 2011;
the Internet of Things information technology to wearable Sangaiah et al. 2023). The literature proposed that the
devices (Li et al. 2021). On the basis of updating the status Internet of Things is to effectively connect people and things
of intelligent systems in physical training for sports majors, through the mobile Internet. Through RFID, infrared sens-
it will help sports majors use wearable devices more rea- ing, GPS, virtual simulation and other technologies,
sonably and effectively for sports training. However, at according to the algorithm protocol, the real objects in the
present, the sports data management of sports training in objective world are transformed into digital instructions of
China is basically in a disordered stage. There is no virtual world informatization, so that people can communi-
information accumulated in the history of competitive cate with objects through information technology (Sund-
sports in developed countries for decades, nor can we find maeker et al. 2010). Build a new dimension of
relevant laws and methods (Metcalf et al. 2016; Sangaiah communication between people and things, things and
et al. 2023). According to several phenomena in current things, greatly improve people’s control, facilitate people’s
sports training, this paper discusses the characteristics of monitoring and management of things, improve work effi-
the use of the wearable Internet of Things in sports training ciency and open a new way of work and life through the
and uses advanced processing technology to discuss the Internet of Things. Through the analysis of big data, the
information of athletes’ physical training (Xiao et al. literature shows that the application of wearable devices has
2018). The experimental data also show that the experi- a positive impact on the physical training mentality and
mental data have a strong approximation function to the action mode of college students majoring in sports, and the
preside of sports training results and have a good gener- duration and frequency of the application of wearable devi-
alization, which means that the use of experimental data ces are also positively related to the students’ sports training
with higher accuracy has a large application space for attitude (Haghi et al. 2017). Therefore, students who use
sports training (Rajšp and Fister 2020). wearable devices are higher than those who do not use
wearable devices in sports duration, frequency and training
intensity. In combination with big data mining technology,
2 Related work the literature has developed a perfect sports training simu-
lation decision support evaluation system and conducted in-
With the full application and market promotion of artificial depth research on the possibility of using big data analysis
intelligence technology, 5G technology and Internet of technology in student physical training (Wang et al. 2020).
Things technology in the real economy, the literature pro- The literature has developed a training quality monitoring
vides a good technical environment and high-tech support and on-the-spot tactical statistics system through data mining
for the development of intelligent wearable sports goods technology, which can help coaches make and adjust training
industry (Huang and Qiu 2020). The literature shows that the plans. Literature has developed a set of student physical
traditional mechanical instruments that record human body training system, which can conduct data management and
data are gradually being replaced by various wearable big data analysis on training, so as to improve students’
intelligent products. Wearable intelligent products become sports training level.
the ‘‘darling’’ of the new era, inject people’s consumption
goals and requirements into the design of front-end wearable
intelligent products, and display the corresponding design 3 Real-time update technology and data
results on the front end after data analysis of the back-end accuracy analysis of the wearable Internet
cloud computing platform (Sohn and Kwon 2020). To some of Things
extent, wearable intelligent products provide a terminal
carrier for the ‘‘Internet of Things ? ’’ and determine the era 3.1 Wearable devices of the Internet of Things
development direction of ‘‘Internet of Things ? .’’ The lit-
erature analyzed the multi-attribute model of athletes’ In the sports industry, sports monitoring technical equip-
training methods, explored the simulation effects of different ment is mainly based on wearable technology. With the
algorithms on athletes’ training models and analyzed the development trend and technology improvement, wearable
mechanism of applying data mining technology to physical devices will also become the main development trend of

123
Intelligent system simulation and data accuracy of physical fitness training for sports majors 10147

the future sports goods industry. Wearable technology, as and near-infrared light, so as to establish the model of
one of the most important contents in China’s industry, has arterial wave oxygen saturation. To further improve the
been occupying a small share. Due to the shrinking market, accuracy of blood oxygen saturation measurement, through
the penetration rate has gradually decreased. In the wear- the spectrum characteristic absorption curve of oxygenated
able field, wearable devices with different functions and hemoglobin Hb and reduced hemoglobin RHb, the 660-nm
forms will become the main scientific and technological and 940-nm red light and near-infrared light source are
support for China’s sports in the future with the constant selected to irradiate the finger tip of human body, and the
updating and breakthrough of scientific and technological blood oxygen content can be determined by correcting
development. Relying on wearable technology and equip- Lambert Beer law.
ment, it is very important to form an industrial positioning With the change of capillary contraction and relaxation
and service system integrating healthy sports culture with caused by blood pulsation, the length of light path will also
Internet plus new thinking. change accordingly in the transmission measurement pro-
In the application conception of sports curriculum, cess. It is assumed that the transmission light path length of
wearable technology is taken as the research and devel- small artery blood flow increases from L to D L long, then:
opment cornerstone, wearable equipment is integrated with
I DC  I AC ¼ I DC eðeHbO2 CHbO2 þeHb CHb ÞDl ð1Þ
virtual reality equipment and Internet information tech-
nology, sports data are managed visually through real-time In½ðI DC k1  I AC k1 Þ=I DC k1  eHbO2 k1 C HbO2 þ eHb k1 CHb
¼ ð2Þ
collection of subject’s sports data indicator signals, and the In½ðI DC k2  I AC k2 Þ=I DC k2  eHbO2 k2 C HbO2 þ eHb k2 CHb
application model of role psychological simulation is used
in virtual scenes. It further shows the development idea of According to the formula of arterial oxygen saturation:
applying sports simulation technology to student sports C HbO2
SaO2 ¼ ð3Þ
courses and simulation training activities, and then carrying C HbO2 þ C Hb
out visual management of student sports. According to
The following can be obtained:
relevant research, different wearable devices can be used in
I AC k1
different aspects. For example, activity trackers are mostly I DC k1
used for students’ own fitness activities; sports intelligent SaO2 ¼ A  B ð4Þ
I AC k2
watches mostly record changes in students’ posture skills. I DC k2

In terms of selection criteria, consumers in different In the actual measurement, add a secondary item, and
countries have different consumer perceptions of wearable set:
devices. The consumer perceptions focus on health detec-
I AC k1
tion, health data and sleep quality. I DC k1
R¼ ð5Þ
Electrocardiogram mainly shows the electrical activity I AC k2
I DC k2
of myocardium when it is excited, which has great clinical
reference value for the basic function of myocardium and The measurement formula of arterial blood oxygen
disease research, and also reflects the significance of concentration is:
electrocardiogram. Electrocardiogram can not only analyze SaO2 ¼ A  R2 þ BR þ C ð6Þ
and differentiate various arrhythmias, but also reflect the
degree of myocardial damage and the functional structure Among them, A, B and C are undetermined coefficients,
of atria and ventricles. With the physical and physiological which can be obtained by experimental calibration. Pho-
movement of the body, the respiratory signal will produce toelectric volumetric pulse wave signal is used to measure
a large baseline deviation, which will affect the accuracy of human pulse at the same time. Compared with conven-
the respiratory volume measurement results. In order to tional piezoelectric sensor pulse wave measurement, it has
eliminate the influence of baseline drift, a three-axis better noise resistance and detection efficiency.
acceleration motion sensor is used in the control system to
synchronously collect human motion signals with small 3.2 Data real-time monitoring and updating
mean square detection and adaptive filter filtering. Arterial technology
blood oxygen content and pulse measurement based on the
theory of pulse spectral brightness, the optical capacitance The basic principle of PPG signal detection is the basic law
plethysmography (PPG) is used to determine the peak of light absorption, which indicates that the object selec-
characteristic points of pulse light wavelength by adaptive tively absorbs light of various wavebands. If a beam of
differential closure method according to the radiation parallel monochromatic light irradiates the medium verti-
absorption characteristics of oxygenated hemoglobin Hb cally, the absorbance of the solution is positively correlated
and reduced hemoglobin RHb in human tissues to red light with the content and thickness of the solvent. If A is the

123
10148 Z. Yong

1X m X
absorbance and T is the transmittance, then the incident p

light intensity is Io and the penetrating light intensity is It. E¼ ð obk  ok Þ2 ð13Þ
2 i¼1 k¼1
The correlation is:
I0 1 According to the conjugate gradient descent method,
A ¼ lg ¼ lg ¼ K  c  l ð7Þ calculate the weight coefficient between the nodes of the
It T
modified input layer and the hidden layer:
where K is the molar absorption coefficient, L is the
wh0 ðt þ 1Þ ¼ wh0 ðtÞ þ aðb o ð1  obÞxh
o  oÞb ð14Þ
strength of the absorbing material, and c is the content of
the light absorbing material, in mol/L or g/L. T refers to the The weight coefficient between the nodes of the modi-
difference between the incident light Io and the transmitted fied hidden layer and the output layer is:
light It on the light attractive medium. The formula is: X
n

It wih ðt þ 1Þ ¼ wih ðtÞ þ a ððb o ð1  obÞwih ðtÞÞxi

o  oÞb
T¼ ð8Þ j¼1
I0
ð15Þ
When scattering and other effects are ignored, after the
incident light passes through the medium, the light beam is The acceleration and gyroscope sensors are combined to
mainly divided into three parts, as shown in: obtain the acceleration vector. The acquisition device is
placed in a unified place in advance for attitude correction.
I0 ¼ Ix þ Ir þ It ð9Þ
At this time, the initial angle of the gyroscope is:
The dynamic ECG information and motion information  
hx ; hy ; hz ¼ ð90; 90; 0Þ ð16Þ
are processed by format conversion and normalization to
obtain the original information of corresponding form and It is not difficult to find out that the instantaneous x-axis
scale. In the experiment, the data sampling rate is 0.25 Hz, angle (y, z) is similar when the following gyroscope detects
and the AD conversion bit is 24 bit. The sampling rate is the three-axis angular velocity
reduced to 200 Hz by using the method of down sampling, hx ðaÞ ¼ hx ða  1Þ þ xX ða  1Þ  Dt ð17Þ
and then the digital compression algorithm is used to
convert the 24 bit number to 16 bit. The numerical value If the triaxial accelerations measured by the acceleration
with small capacity is obtained, only to meet the require- sensor are A1, A2, A3 (free coordinates, non-x-, y-, z-axis
ments of the neural network training system. The normal- components), then the size of A is the square root of the
ization algorithm usually uses the linear transformation three. The recurrence formula of the x-axis component
algorithm, whose expression is: (similar to the Y-axis) of the horizontal acceleration vector
Ah = (Ahx, Ahy) is:
2  ðx  minÞ
f ð xÞ ¼ 1 ð10Þ Ahx ðaÞ ¼ hx ðaÞ  jAðaÞj ð18Þ
ðmax  minÞ
The initial neural network mode is constructed through The core concept of a Person’s movement process is the
the method of artificial neural network: The neural network coordinate system and rotation matrix. I have discussed the
mode is composed of three levels of structure: input layer, switching method between the right-handed coordinate
hidden layer and control layer, and the connection between system and the virtual left-handed coordinate system in the
the input layer and hidden layer is as follows: real world before. Here, the drive between the upper and
lower levels of the skeleton is generally used for quaternion
X
3
rotation. Suppose the pose of the left upper arm bone A at t,
ai ¼ gð whij xj þ bhi Þ ð11Þ
j¼1
the quaternion is:
qA ðtÞ ¼ qAcol qA ðtÞþ qD A ð19Þ
The input of the input layer is the working data sequence
of the x-, y-, z-axes of the three-axis acceleration sensor. The spatial orientation of the left lower arm bone B in
The activation functions of the hidden layer and the output system a is:
layer are shown in the formula:
qBloc ¼ qA ðtÞþ qA ðtÞ1 qABloc ð20Þ
X
10
Ni ¼ w0ij xj þ b0i ð12Þ qABloc ¼ qAloc þ qBloc ð21Þ
j¼1
The attitude quaternion of left lower arm B at time t is:
Calculate the cumulative error E between the output
qB ðtÞ ¼ qBcol ðtÞþ þ qD B ð22Þ
layer output of all training samples and the expected result
of training samples: qB0 ðtÞ ¼ qB ðtÞþ qA ðtÞ ð23Þ

123
Intelligent system simulation and data accuracy of physical fitness training for sports majors 10149

correlation coefficient and Kendall correlation coefficient

are monotone trend and classification variable of relevant
3.3 Model simulation and data accuracy analysis
variables, respectively. The Pearson correlation coefficient
analysis method can also be used as the relevant index of
The test data are shown in Fig. 1, which is the original
this experiment according to the correlation between the
ECG data with motion noise interference.
expected heart rate index and the actual heart rate index to
Using the above noise simulation, the ECG waveform
be analyzed.
with motion noise removed can be finally obtained. As
shown in Fig. 2, since the restored waveform has clear CovðBPM est ; BPM true Þ
R¼ ð26Þ
ECG signal characteristics, the denoising effect is rBPM est  rBPM true
significant.
The average absolute error of the calculated heart rate
Average absolute error, correlation analysis, Bland–
results of the actually collected data is 11.06 bpm, and the
Altman chart and other three types of objective indicators
average absolute error percentage is 10.68%. For data set
are used to analyze and evaluate the test results to measure
S-1, one group of heart rate prediction results is shown in
the deviation between the predicted average heart rate and
Fig. 3.
the actual average heart rate, and the robustness of the real-
For data set S-1, by comparing the heart rate change
time average heart rate algorithm provided in this paper.
results obtained by LMS and NLMS denoising algorithm,
The average absolute error can be expressed as the average
as shown in Fig. 4, it can be seen that compared with LSM
of the sum of the absolute errors between the calculated
and NLSM denoising, the heart rate change calculation and
actual heart rate and the predicted heart rate:
denoising algorithm presented in this paper can track the
1X M
heart rate change trend during running well, while LSM
AAE ¼ jBPM est ðiÞ  BPM true ðiÞj ð24Þ algorithm is not good at processing nonsmooth informa-
M i¼1
tion. However, NLSM algorithm is better than RLS algo-
Taking into account the different physical factors and rithm in computing real time. However, because the
the varying degrees of average heart rate of different test methods presented in this paper use a low-order RLS
individuals, this study uses the average absolute error algorithm framework, the overall computing capabilities on
percentage pair to calculate the absolute error between the embedded wearables are not very different. Combined with
expected result of the whole test and the actual average the different methods investigated previously and the
heart rate value, and the average of the proportion between information collected in the test equipment in this paper,
the expected result and the actual average heart rate dif- the RLS adaptive signal filtering calculation efficiency of
ference as the method to evaluate the robustness of the the multi-sensor technology presented in this paper is the
results: best. In the wearable real-time human heart rate detection
platform applied in this paper, the twelve data and analysis
1X M
jBPM est ðiÞ  BPM true ðiÞj
AAEP ¼ ð25Þ results of the data set s-1 obtained in the experiment are
M i¼1 BPM true ðiÞ
shown in Fig. 4. Under the condition that the maximum
In correlation analysis, the degree of correlation spectral peak is directly selected for heart rate calculation
between two or more variables is generally considered. when comparing the spectral regions without the applica-
Common indicators include Pearson’s maximum correla- tion of the decision tree, the average error of the maximum
tion coefficient analysis, Spearman’s correlation coefficient heart rate calculation is still not applied when the spectral
analysis and Kendall’s correlation coefficient analysis. regions of the decision tree are selected.
Pearson correlation coefficient generally includes ordered
discrete or coherent variables, taking into account the
degree of linear correlation between variables; Spearman

Fig. 1 Original ECG waveform

123
10150 Z. Yong

Fig. 2 ECG waveform after

removing the interference of
motion noise

Fig. 3 Comparison chart of

predicted heart rate and actual
heart rate

Fig. 4 Comparison of real-time

heart rate calculation results of
data set S-1

123
Intelligent system simulation and data accuracy of physical fitness training for sports majors 10151

4 Design of intelligent system for physical services. Sports majors submit their personal physiological
training of physical education students data to the network service system, which transmits it to users
through SMS, mobile phone APP and other channels, so as to
4.1 System requirement analysis achieve the purpose of sports training for users.
(3) Uniqueness
Physical exercise monitoring system refers to the comprehen-
sive monitoring and evaluation of the physical load, activity The Internet plus ? wearable sports health monitor-
level and physical condition of the exercisers and the evaluation ing platform means that the human body data of sports
of physical exercise after exercise in the normal sports envi- professionals can be uploaded online (Internet online
ronment. On the basis of traditional wearable research and website), and the data can be analyzed, judged and sci-
development, this research uses the real-time monitoring ser- entifically evaluated. It can also be monitored offline
vice in the early, middle and late stages of sports to achieve the (wearing wearable devices). The mode of online and
service functions of physical training and guidance, scientific offline integrated operation shows its unique business
prevention of sudden death in sports and so on. Through the mode.
establishment of Internet plus ? wearable sports monitoring (4) Differentiated services
service, to meet the needs of modern science and technology
development, so that the majority of sports students feel the Compared with the popular health club user services, the
great power of modern science and technology, so as to effec- network ? wearable sports monitoring service built in this
tively master the most scientific and advanced sports methods, paper takes sports as the main means and wearable devices
thus realizing the sports students’ scientific understanding and as the core. The health intervention service for social groups
mastery of their own body state in sports, which is of great has obvious advantages, while compared with the service
significance for improving physical and mental health and items of traditional health clubs, the network ? wearable
developing scientific sports cause. Through the establishment sports monitoring service has very prominent advantages in
of Internet plus ? wearable sports monitoring service, from the the acceptance of social groups and the scientific, timely, all-
perspective of macromanagement, joint research, development round and operable service items.
and promotion with hospitals, government departments, com-
panies, individuals, etc., can realize the practical significance of 4.2 System architecture and function design
its application and bring huge benefits to all sectors of society.
We can cooperate with the intelligent management system of The main architecture of the Internet plus ? wearable
physical education students’ training, join the social sports motion monitoring service system is the Internet network
guidance, expand the influence and provide favorable condi- platform, wearable technology and application foundation.
tions for promoting the development of the intelligent man- Both of them promote each other and are indispensable.
agement system of physical fitness training of sports students. They become the basic elements of the Internet
The biggest feature of Internet plus ? wearable health plus ? wearable motion monitoring service system and lay
exercise and monitoring service system is the service an important foundation for the formation of a more
advantage shown by the unique technology of ‘‘Internet complete business model.
plus ? wearable.’’ It has the advantages of convenience, (1) Wearable motion monitoring
interaction, uniqueness and differentiation.
Wearable device development is based on life health
(1) Fastness monitoring instrument. It can monitor various indicators of
The wearable device detects more physiological parame- the body in real time, for example, pulse, respiration, blood
ters in real time and completes real-time physiological oxygen, body temperature, ECG, blood pressure and other
parameter analysis and data processing, timely and accurately human technical parameters. Using wearable devices,
reflects to the user group, timely and effectively solves sci- through the health detection network in the human body
entific health care and exercise methods through the feedback area, under the premise of not affecting the work, select
of exercise prescription and timely and effectively solves the appropriate measurement points and conduct health moni-
unnecessary health tragedy, which also reflects the fast toring through miniaturization, integration and intelligent
characteristics of the system’s service mode. design ideas. The research uses portable and comfort-
able wearable instruments and wireless platforms to embed
(2) Interactivity ECG detection modules, pulse current detection modules,
The combination of network platform and wearable devi- respiration modules, data storage units and communication
ces shows the advantage of interactivity in network public

123
10152 Z. Yong

modules in wearable devices, and uses the microcontrollers

of each detection module to connect with smart phones or
computers as application terminals to complete the network
transmission of detection data through the point-to-point
wireless network connected with bluetooth. Schematic
diagram of system detection indicators is shown in Fig. 5.
Wearable monitoring mode: This service mode combined
with the new thinking of Internet plus technology will bring
more efficient service mode to wearable devices. The
wearable monitoring method is to use the microcontrol
device in the tester’s body to directly input the local infor-
mation into the computer on the bluetooth media platform
and then transfer the information to the Internet client. With
the help of the Internet plus network service platform,
experts and scholars in the field of sports medicine make
scientific evaluation on the collected data processing through
analytical means and techniques. It has achieved the real Fig. 6 Structure diagram of wearable monitoring mode
monitoring effect for sports majors. During exercise, wear-
able devices can be used to monitor life data such as res- important condition to complete the formation of Inter-
piratory mode, blood oxygen, ECG and pulse. In the net ? wearable sports monitoring service system mode. At
monitoring of ECG data, effective early warning operations the same time, wearable mode can be used as an important
can be made for the conditions such as irregular heart rate, basis for life health monitoring and protection in a more
arrhythmia, low blood oxygen. The server will send the data convenient, scientific and efficient way. Therefore, it is
results back to personal computers and mobile phones and necessary to build an Internet service platform integrating
timely feed back to sports students, which can effectively health and monitoring, which will promote the growth of
control the real-time changes of physiological parameters in intelligent system for physical training of sports majors.
the early, middle and late stages of physical exercise. Its Therefore, it is very important to carry out active health
structure and function are shown in Fig. 6. intervention with Internet service system as the carrier. The
operation mode of Internet services is just like the biological
(2) Application of Internet plus network service platform
chain and has certain circularity. Simply speaking, the
The application of Internet plus network platform is Internet service system is a transit station for data analysis
based on the technical basis of classifying and evaluating the and processing, life health, user feedback, etc. From the
data information obtained by wearable devices. It is to meet perspective of network service technology, it is the basis for
the needs of public service informatization function. It is an consulting service, tracking service and realizing its public
welfare role. Only when they are interdependent can the
Internet service network operate normally. In the design of
the network service system, the setting of each business
function is particularly important. In the network service
system, wearable motion monitoring equipment is generally
the main part, and the functions of information collection
and data analysis management, sports information module,
interactive communication area, sports personal manage-
ment consulting function, network feedback function,
tracking service and other functions become the basic
functions provided by the business system.
Interactive communication area. In the interactive
communication area, communication and interaction with
sports majors and expert teams can be carried out. Learn
about the basic knowledge of life health care and common
sense of science and sports guidance online. It can also
recognize the service focus of sports professional students
Fig. 5 Schematic diagram of system measurement index

123
Intelligent system simulation and data accuracy of physical fitness training for sports majors 10153

Table 1 System data

Data update rate 60 Hz 120 Hz 180 Hz
transmission update rate
parameters Total number of connected acquisition units 18 18 14
Total number of acquisition units actually working 18 18 14
Data throughput 191,320 bits/s 389,481 bits/s 4,591,294 bits/s

Table 2 Test data

Test case Test index Value (%)

PTC-01-0001 base load test Packet loss rate 0

Data integrity 98
CPU load of clearing service 3
Database disk I/O queue 4
PTC-01-0002 general load test Packet loss rate 1
Data integrity 100
CPU load of clearing service 25
Database disk I/O queue 14
PTC-01-0003 high load test Packet loss rate 0
Data integrity 97
CPU load of clearing service 37
Database disk I/O queue 29

Table 3 Comparison between this system analysis method and traditional methods
Comparison items Traditional analysis Analysis of this system

Physical performance in historical Handwritten record Shift at any time after collection
competitions
Physical characteristics of athletes Inefficient, time-consuming and error prone Automatic collection and calculation, easy to use analysis
results
Athlete Comparison Through historical account Provide data comparison support
books ? perceptual knowledge
Role positioning Comparison by experience Complete role indicators are available for reference
Personal theory Experience summary Highly customizable effect indicators are available for
definition and selection
Competitive level Cannot count Massive athlete database
Talent development Evaluation with sensibility Provide talent tracking function

from the side, so as to provide basis for timely and effec- power supply of the collection module is started to keep the
tive adjustment of service direction and focus. collection system in standby mode. Then, the information
Online guidance and feedback module. Online guidance collection process is started with the networking command
service is mainly composed of health questionnaire test, sent by the collection system through the PC terminal.
expert health promotion guidance program design and Using the cathode ray oscilloscope, the pin information of
effective feedback monitoring indicators. The function of the RS485 transceiver of the collection node can be
data extraction can be realized in the health questionnaire detected, so that the information update rate can be directly
test, and the personalized design concept of service can be observed. This technology can achieve 60 Hz and 120 Hz
reflected from the design of expert health guidance scheme, data refresh speed according to relevant indicators set.
so as to obtain real-time effective data feedback. Through inspection and analysis, the system update rate
data can be obtained as shown in Table 1:
4.3 System test results Based on the test cases in the previous section, the test
data are obtained as shown in Table 2:
In this paper, several sensor modules are connected with Based on the above test conclusions, this paper indicates
the collection unit through the RS485 bus, and then the that the system can support 20 DCTU ports under the

123
10154 Z. Yong

pressurization environment conditions provided by data References

information and the pressurization requirements of ten
messages per second for each port. The transmission dis- Haghi M, Thurow K, Stoll R (2017) Wearable devices in medical
tance in an obstacle-free environment is 200 M, and the internet of things: scientific research and commercially available
devices. Healthc Inform Res 23(1):4–15
transmission distance in an obstacle-free environment is Hill J, Leiszler M (2011) Review and role of plyometrics and core
50 M, with strong fault tolerance. The test result is quali- rehabilitation in competitive sport. Curr Sports Med Rep
fied and can meet the established training requirements. 10(6):345–351
The introduction of this information system has greatly Huang H, Qiu J (2020) Strategic choices for high-quality develop-
ment of intelligent wearable sporting goods industry in the new
improved the training management methods that coaches era. Wearable Tech-Nology 1(1):41–55
used to look, hear and inquire about, as shown in Table 3. Jones BA, Arcelus J, Bouman WP, Haycraft E (2017) Sport and
transgender people: a systematic review of the literature relating
to sport participation and competitive sport policies. Sports Med
47(4):701–716
5 Conclusion Kong X, Tong S, Gao H et al (2020) Mobile edge cooperation
optimization for wearable internet of things: a network represen-
Sports kinematics is a comprehensive specialty, covering tation-based framework. IEEE Trans Ind Inf 17(7):5050–5058
humanities, sports science, sports sociology, etc. The devel- Li X, Lu Y, Fu X, Qi Y (2021) Building the Internet of Things platform
for smart maternal healthcare services with wearable devices and
opment of computer science and information technology, cloud computing. Futur Gener Comput Syst 118:282–296
especially the Internet of Things and data mining technology, Marinov M, Mosa ASM, Yoo I, Boren SA (2011) Data-mining
has laid a solid ideological foundation for scientific research technologies for diabetes: a systematic review. J Diabetes Sci
training and the application of excellent methods. Traditional Technol 5(6):1549–1556
Metcalf D, Milliard ST, Gomez M, Schwartz M (2016) Wearables
sports wearable devices have a positive impact on the fre- and the internet of things for health: wearable, interconnected
quency and duration of physical training for sports majors, with devices promise more efficient and comprehensive health care.
a high degree of satisfaction, but there are problems such as IEEE Pulse 7(5):35–39
slow system response, delayed status update, inaccurate cal- Pirbhulal S, Samuel OW, Wu W, Sangaiah AK, Li G (2019) A joint
resource-aware and medical data security framework for wear-
culation, etc. It is necessary to strengthen the functional and able healthcare systems. Future Gener Comput Syst 95:382–391
safety research and design of wearable devices. In this paper, Rajšp A, Fister I Jr (2020) A systematic literature review of intelligent
the wearable Internet of Things device is used to improve the data analysis methods for smart sport training. Appl Sci 10(9):3013
intelligent system of physical fitness training for sports majors, Sangaiah AK, Rezaei S, Javadpour A, Zhang W (2023) Explainable
AI in big data intelligence of community detection for digital-
improve the speed of data updating and provide a better method ization e-healthcare services. Appl Soft Comput 136:110119
for physical fitness training of sports majors. In addition, due to Sohn K, Kwon O (2020) Technology acceptance theories and factors
the accumulation of a lot of training data in the process of sports influencing artificial Intelligence-based intelligent products.
training, it is difficult for conventional big data analysis tech- Telematics Inform 47:101324
Sundmaeker H, Guillemin P, Friess P, Woelfflé S (2010) Vision and
nology to find a reasonable model to explain the correlation challenges for realising the Internet of Things. Clust Eur Res
between these data. The discovery of the Internet of Things puts Proj Internet Things Eur Commis 3(3):34–36
forward optimization methods for finding scientific rules and Wang X, Huang D, Zhao X (2020) Design of the sports training
correlations in a large number of complex training data and decision support system based on improved association rule, the
apriori algorithm. Intell Autom Soft Comput 26(4):755–763
provides a basis for improving the training quality of sports Xiao F, Miao Q, Xie X et al (2018) Indoor anti-collision alarm system
training. based on wearable Internet of Things for smart healthcare. IEEE
Commun Mag 56(4):53–59
Xiao N, Yu W, Han X (2020) Wearable heart rate monitoring
Funding No funding is provided for the preparation of the intelligent sports bracelet based on Internet of things. Measure-
manuscript. ment 164:108102
Yang J, Li Y, Liu Q et al (2020) Brief introduction of medical
Data availability Data will be made available on request. database and data mining technology in big data era. J Evid
Based Med 13(1):57–69

Declarations Publisher’s Note Springer Nature remains neutral with regard to

jurisdictional claims in published maps and institutional affiliations.
Conflict of interest The authors declare that they have no conflict of
interests. Springer Nature or its licensor (e.g. a society or other partner) holds
exclusive rights to this article under a publishing agreement with the
Ethical approval This article does not contain any studies with human author(s) or other rightsholder(s); author self-archiving of the
participants performed by any of the authors. accepted manuscript version of this article is solely governed by the
terms of such publishing agreement and applicable law.

123