IoT based Circadian Rhythm Monitoring using Fuzzy Logic
K. Sornalakshmi, Revathi Venkataraman, N. Parthiban and V. Kavitha
Department of Computer Science and Engineering, SRM Institute of Science and Technology,
Kattankulathur, Tamilnadu, India
Keywords: Internet of Things, Circadian Rhythm, Sleep, Wake up, Light, Fuzzy.
Abstract: A healthy body and happy mind is essential to lead a successful life. For the betterment of health, it is
important to develop positive health habits. The human body has a 24-hour internal clock called circadian
rhythm that can be affected by our lifestyle. It is a natural intelligence of the human body to perform certain
tasks including hormone secretion, memory functions, immune system functions, etc. during certain periods
of the day. This rhythm is synchronized with the light and dark cycle of the environment. However, when
there are variations in light, sleeping at unusual times, exposure to bright lights at night and traveling across
time zones, certain functions of the body may get activated and deactivated at inappropriate times. With smart
devices, the Internet of Things (IoT) has a great impact on our everyday lives. Healthcare IoT systems use the
data provided by the IoT devices to make automated decisions or to provide recommendations to users. This
work is concerned with a health IoT system consisting of different IoT devices used by users who want to
know about their circadian rhythms. In addition to this, fuzzy logic has been used to evaluate the circadian
rhythm based on the effects of the time at which an individual starts to sleep, wake up time and mobile light
exposure time. It classifies the circadian rhythm as aligned, intermediate and disrupted.
1 INTRODUCTION
In modern working culture, with the opportunity to
work from anywhere at any time, many of us forgot
to keep our internal circadian rhythm regularised.
Circadian rhythms are 24-hour cycles in our body’s
internal clock to perform certain tasks during certain
times of the day using an intelligence that nature has
designed for us. It plays a key role in regulating our
metabolism, hormone production, cell growth and
many activities. Living with the lifestyle against
circadian rhythm has a high impact on health in such
a way that increases risk of obesity, metabolic
syndrome, and cardiovascular diseases (Kessler and
Pivovarova, 2019). It is vital to understand the impact
of circadian disruption on human health and align
them to improve the quality of our health and life on
earth (Walker et al., 2020). Almost all functions of the
body depend upon the circadian clock which operates
with the light and darkness cycle. While the other
cues like social activity, feeding and fasting cycle,
nutritional factors also influence the circadian
rhythm. From this perspective, an understanding
about the circadian clock and how the accurate timing
of every activity in our lifestyle will help to shed light
on our health (Alemdar, 2018). Lifestyle aligning
with the laws of nature regulates the circadian rhythm
to prevent health illness. There are a number of health
habits that can support us to live in synchronize with
the circadian rhythm. Lifestyle changes that honour
our body’s natural schedule will help to improve
health, disease prevention and more.
The light/dark cycle of nature is the best known
regulator of the circadian clock (Fukuda and Morita,
2017). Other than natural light, humans are exposed
to artificial lighting from electronic screens on
televisions, computers, tablets, and cell phones. The
impact of light on the circadian rhythm depend upon
the intensity and timing of exposure to light (Blume
et al., 2019). Environment temperature seems to play
a big part in the circadian system. The clock neurons
get the temperature information from external
thermoreceptors which are specialized neurons used
by the skin to detect changes in temperature
(Yadlapalli et al., 2018). Environmental factors
disorganise the circadian system, and thereby
adversely affect the health. Since there is an
interaction between the circadian rhythm and
metabolism, timing of food intake is a crucial factor
to regulate metabolism (Jennifer et al., 2019).
Sornalakshmi, K., Venkataraman, R., Parthiban, N. and Kavitha, V.
IoT based Circadian Rhythm Monitoring using Fuzzy Logic.
DOI: 10.5220/0010451502230228
In Proceedings of the 6th International Conference on Internet of Things, Big Data and Security (IoTBDS 2021), pages 223-228
ISBN: 978-989-758-504-3
Copyright
c
 2021 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
223

Changes in the core body temperature and skin
temperature reflect the circadian disruption.
Nutritional compounds have robust effects on the
circadian system. To maintain the circadian rhythm,
the choice of the nutrition should be clean and
healthy. The misaligned circadian cycle leads to
emotional instability and also results in mood
disorders such as depression, anxiety and bipolar
disorder (McClung, 2013). Light and dark cycles play
a central role in the regulation of the body's internal
clock that affects the secretion of melatonin which is
an essential hormone for promoting sleep. The
melatonin suppression and circadian disruption are
modulated by the amount of light seen during the day.
Also, the effect of light on the circadian rhythm
mainly depends on the duration, timing and intensity
of light exposure (Papatsimpa et al., 2020). Today,
because of the technological addiction and the
overuse of computers, laptops, and smartphones,
people are increasingly exposed to artificial light
which develops sleep deprivation (Lee and Kim,
2019). The circadian system is most sensitive to light-
induced delays particularly in the evening hours. In
our proposed work, the circadian rhythm of an
individual is classified using the fuzzy model with the
input parameters sleep, wake up and light exposure
time.
The rest of the paper is organized as follows.
Section 2 discusses the role of the internet of things
in circadian rhythm. The proposed fuzzy model is
presented in Section 3. Section 4 gives the analysis of
the proposed model with its results. The conclusion
and the future work is presented in Section 5.
2 INTERNET OF THINGS IN
CIRCADIAN RHYTHM
The study conducted in (Roomkham et al., 2019) to
investigate the possibility of using an Apple watch for
sleep monitoring recommends that Apple watch
could be best in detecting sleep and wakefulness as it
has high accuracy (97%) and sensitivity (99%). Apple
Watch consists of a triaxial accelerometer and heart
rate sensors to measure sleep/wake up cycle. Hence,
we used that Smart watch to measure input
parameters 𝑋

and 𝑋

. Exposure to artificial light is
mentioned in suppressing Melatonin, which is a
hormone inducing sleep, causing sleep disorders by
disrupting our natural circadian rhythms. To measure
the third input parameter, mobile light exposure time,
Apple iPhone was used. This smartphone will send
the overall mobile usage time of the particular user to
the cloud. We recorded 30 days of sleep/wake up and
mobile light exposure data from 15 healthy adults (10
female and 5 male) using their smart devices such as
Apple Watch and Apple iPhone. The participants
involved in this study were morning type. They used
to have normal hours of work from 9.00 A.M. to 5.00
P.M.
3 FUZZY MODEL FOR
CIRCADIAN RHYTHM
CLASSIFICATION
For the present study, wearable technology has been
used to calculate sleep/wake up schedule and light
exposure. To measure the output Circadian Rhythm
(𝑍), Sleep Time (𝑋

), Wake up Time (𝑋

), Light
Exposure Time ( 𝑋

) are considered as input
parameters.
Mamdani fuzzy rules have been used for building
a fuzzy model. It is of the form in Equation 1.
𝐼𝐹 𝑋 𝑖𝑠 𝐼 𝑇𝐻𝐸𝑁 𝑌 𝑖𝑠
𝐽
(1)
The general form of the rule is shown in Equation
2.
𝐼𝐹 𝑋

𝑖𝑠 𝐼

𝐴
𝑁𝐷 𝑋

𝑖𝑠 𝐼

…𝑋

is 𝐼

THEN … 𝑌

i
s
𝐴
𝑁𝐷 𝑌

𝑖𝑠
𝐽

…𝑌

is
𝐽

(2)
We assume a Gaussian membership function that
is defined by a mean m and a standard deviation 𝜎 >
0. The fuzzy membership value is computed using the
following Equation 3.
𝑦 = 𝑔𝑎𝑢𝑠𝑠𝑚𝑓(𝑥, 𝑝𝑎𝑟𝑎𝑚𝑠)
(3)
Where 𝑥 represents the input values for which
membership values to be computed and 𝑝𝑎𝑟𝑎𝑚𝑠
represents membership function parameters, which
specified as the vector (𝜎, 𝑐) , where 𝜎 represents
standard deviation and c represents the mean.
𝑔𝑎𝑢𝑠𝑠𝑚𝑓() computes fuzzy membership value 𝑦
using the Gaussian membership function shown in
Equation 4.
𝑓
(
𝑥; 𝜎, 𝑐
)
=𝑒
()



(4)
The membership values are calculated for each
input value in 𝑥. The general representation of the
fuzzy system is depicted in Figure 1.
IoTBDS 2021 - 6th International Conference on Internet of Things, Big Data and Security
224

Figure 1: General representation of the fuzzy system.
The first input parameter (𝑋

) describes the time
at which individuals start to sleep and it is mapped to
a scale varying from 0 to 10. Highest score implies
regular and consistent sleep time. The qualitative
descriptors of 𝑋
1
shown in Figure 2 are categorized as
‘Correct’, ’Moderate’ and ‘Late’.
Figure 2: Qualitative descriptors of Sleep Time (𝑋

).
The second input parameter (𝑋

) describes the
time at which an individual wakes up and it is also
mapped to a scale varying from 0 to 10 as shown in
Figure 3. Highest score indicates regular wake up
time and the lowest score indicates delayed or early
wake up time.
Figure 3: Qualitative descriptors of Wake up Time (𝑋

).
The input parameter (𝑋

) defines the artificial
blue light exposure duration of an individual. Scale
between 0 and 10 has been used to classify this input.
Light exposure in the late evening and early night i.e.,
the hours surrounding the individual’s typical
bedtime produces shifts in the circadian system and
also light exposure in the late night and early morning
produces shifts in the circadian system (Zisapel,
2001). From this inference, the highest score has been
assigned to limited exposure to artificial blue light
and the lowest score has been assigned to over
exposure to light. The qualitative descriptors of input
𝑋

shown in Figure 4 are categorized as ‘Limited’,
’Moderate’ and ‘Over’.
Figure 4: Qualitative descriptors of Light Exposure Time.
The qualitative descriptors of the circadian
rhythm (𝑍) presented in Figure 5 are categorized as
‘Aligned (A)’, Intermediate (I)’, and ‘Disrupted (D)’.
The inference from (Guo et al., 2020) and the
inference from the dataset (Rossi et al., 2020) was
used to classify the circadian rhythm. In their work,
the authors have presented an open dataset of Psycho-
Physiological responses of younger adults. They have
considered many parameters which include
sleep/wakeup data and small/large screen usage time.
In our work, we have considered this dataset as a
reference to classify the circadian rhythm.
Figure 5: Qualitative descriptors of Circadian Rhythm (𝑍).
The circadian rhythm score is specified in a way
that 0 indicates the lowest score, whereas 1 indicates
the highest score. Gaussian membership functions
have been used to describe the Circadian Rhythm
score. Since, three input parameters each with three
membership functions are used to determine the
circadian rhythm score, 27 if-then rules have been
framed for this study. For example, the first rule has
been created as shown in Equation 5.
𝐼𝐹 𝑋

=10𝑎𝑛𝑑𝑋

=10 𝑎𝑛𝑑 𝑋

=10,
𝑡ℎ𝑒𝑛 𝑍 𝑖𝑠
𝐴
𝑙𝑖𝑔𝑛𝑒𝑑
(5)
Correspondingly, remaining rules are framed by
taking into account all other possible combinations.
IoT based Circadian Rhythm Monitoring using Fuzzy Logic
225

Figure 6: Fuzzy Rule Base.
The total number of rules that completely define
the fuzzy set is 27 and it is depicted in Figure 6.
4 ANALYSIS
From Figure 7.A, it can be inferred that the circadian
rhythm of an individual falls low when the time to
start sleep is late and when they are over exposed to
light. For example, the circadian rhythm falls below
0.2 (i.e. Disrupted) when the sleep time is around 3
(i.e. Late) and the light exposure is around 3 (i.e. Over
exposure). From Figure 7 .B, it can be inferred that
the circadian rhythm of an individual falls low when
the time to wake up is late and when they are over
exposed to light. For example, the circadian rhythm
falls below 0.2 (i.e. Disrupted) when the wake up time
is around 3 (i.e. Late) and the light exposure is around
3 (i.e. Over Exposure).
(a) (b) (c)
Figure 7: Relationship between inputs and outputs for disrupted circadian rhythm. Output: Circadian Rhythm and inputs: A:
Sleep Time and Light Exposure, B: Wake up Time and Light Exposure and C: Sleep Time and Wake up Time.
(a) (b) (c)
Figure 8: Relationship between inputs and outputs for Intermediate circadian rhythm. Output: Circadian Rhythm and inputs:
A: Sleep Time and Light Exposure, B: Wake up Time and Light Exposure and C: Sleep Time and Wake up Time.
(a) (b) (c)
Figure 9: Relationship between inputs and outputs for aligned circadian rhythm. Output: Circadian Rhythm and inputs: A:
Sleep Time and Light Exposure, B: Wake up Time and Light Exposure and C: Sleep Time and Wake up Time.
IoTBDS 2021 - 6th International Conference on Internet of Things, Big Data and Security
226

From Figure 7.C, it can be inferred that the
circadian rhythm of an individual falls low when the
time to start sleep is late and when the time to wake
up is late. For example, the circadian rhythm falls
below 0.2 (i.e. Disrupted) when the sleep time is
around 3 (i.e. Late) and the wake up time is around 3
(i.e. Late).
From Figure 8.A, it can be concluded that when
the time to start sleep is moderate (in between correct
and late) and also when they are moderately exposed
to light, an individual's circadian rhythm falls
intermediately. For example, the circadian rhythm
lies at 0.5 (i.e. Intermediate) when the sleep time is
around 5 (i.e. Moderate) and the light exposure is
around 6 (i.e. Moderate). From Figure 8.B, it can be
inferred that the circadian rhythm of an individual
falls intermediately when the time to wake up is
moderate and when they are moderately exposed to
light. For example, the circadian rhythm lies at 0.5
(i.e. Intermediate) when the wake up time is around 5
(i.e. Moderate) and the light exposure is around 6 (i.e.
Moderate). From Figure 8.C, it can be inferred that
the circadian rhythm of an individual falls
intermediately when both the time to start sleep and
wake up is moderate. For example, the circadian
rhythm falls lies at 0.5 (i.e. Intermediate) when the
sleep time is around 5 (i.e. Moderate) and the wake
up time is around 6 (i.e. Moderate).
From Figure 9.A, it can be inferred that the
circadian rhythm of an individual becomes high when
the time to start sleep is correct and when they have
limited light exposure. For example, the circadian
rhythm becomes 0.8 (i.e. Aligned) when the sleep
time is around 9 (i.e. Correct) and the light exposure
is around 9 (i.e. Limited exposure). From Figure 9.B,
it can be inferred that the circadian rhythm of an
individual becomes high when the time to wake up is
correct and when they have limited light exposure.
For example, the circadian rhythm becomes 0.8 (i.e.
Aligned) when the wake up time is around 9 (i.e.
Correct) and the light exposure is around 9 (i.e.
Limited exposure). From Figure 9.C, it can be
inferred that the circadian rhythm of an individual
becomes high when both the time to start sleep and
wake up is correct. For example, sleep and wake up
is correct. For example, the circadian rhythm
becomes greater than 0.8 (i.e. Aligned) when the
sleep time is around 9 (i.e. Correct) and the wake up
time is around 9 (i.e. Correct).
5 CONCLUSION
The contributions of this work are the ability to
evaluate and classify the circadian rhythm of an
individual to improve their health. This classification
has been done using the input parameters, the time at
which individuals start to sleep, wake up time and
light exposure time. A fuzzy logic system for the
evaluation of an individual's circadian rhythm has
been developed. This evaluation will be useful for the
early diagnosis of circadian rhythm disruption. In
future work, the accuracy of the fuzzy model will be
validated with the results classified by the medical
experts. Future work includes other factors that
influence the circadian rhythm such as eating,
exercise for assessment and also inputs will be
gathered from individuals by using wearable devices
with capacities of measurement of physiological
parameters. The trust value of the source will be taken
into account in future.
REFERENCES
Kessler, K., & Pivovarova-Ramich, O. (2019). Meal timing,
aging, and metabolic health. International journal of
molecular sciences, 20(8), 1911.
Walker, W. H., Walton, J. C., DeVries, A. C., & Nelson, R.
J. (2020). Circadian rhythm disruption and mental
health. Translational psychiatry, 10(1), 1-13.
Alemdar, E. (2018). External factors that direct the
biological clock and role in treatment of disease. Arch
Med Vol, (10).
Fukuda, Y., & Morita, T. (2017). Effects of the light–dark
cycle on diurnal rhythms of diet-induced thermogenesis
in humans. Chronobiology international, 34(10), 1465-
1472.
Blume, C., Garbazza, C., & Spitschan, M. (2019). Effects
of light on human circadian rhythms, sleep and mood.
Somnologie, 23(3), 147-156.
Yadlapalli, S., Jiang, C., Bahle, A., Reddy, P., Meyhofer,
E., & Shafer, O. T. (2018). Circadian clock neurons
constantly monitor environmental temperature to set
sleep timing. Nature, 555(7694), 98-102.
Jennifer, L.K., Sharon V.T., & Hannah, D.H. (2017).
Complex interactions of circadian rhythms, eating
behaviors, and the gastrointestinal microbiota and their
potential impact on health, Nutrition Reviews, 75(9),
673-682.
McClung, C. A. (2013). How might circadian rhythms
control mood? Let me count the ways... Biological
psychiatry, 74(4), 242-249.
Papatsimpa, C., Bonarius, J. H., & Linnartz, J. P. M. G.
(2020, June). Human Centric IoT Lighting Control
based on Personalized Biological Clock Estimations. In
2020 IEEE 6th World Forum on Internet of Things
(WF-IoT) (pp. 1-6). IEEE.
IoT based Circadian Rhythm Monitoring using Fuzzy Logic
227

Lee, E., & Kim, M. (2019). Light and life at night as
circadian rhythm disruptors. Chronobiology in
Medicine, 1(3), 95-102.
Roomkham, S., Hittle, M., Cheung, J., Lovell, D., Mignot,
E., & Perrin, D. (2019). Sleep monitoring with the
Apple Watch: comparison to a clinically validated
actigraph. F1000Research, 8(754), 754.
Zisapel, N. (2001). Circadian rhythm sleep disorders. CNS
drugs, 15(4), 311-328.
Guo, J. H., Ma, X. H., Ma, H., Zhang, Y., Tian, Z. Q.,
Wang, X., & Shao, Y. C. (2020). Circadian
misalignment on submarines and other non-24-h
environments–from research to application. Military
Medical Research, 7(1), 1-12.
Rossi, A., Da Pozzo, E., Menicagli, D., Tremolanti, C.,
Priami, C., Sîrbu, A.... & Morelli, D. (2020). A Public
Dataset of 24-h Multi-Levels Psycho-Physiological
Responses in Young Healthy Adults. Data, 5(4), 91.
IoTBDS 2021 - 6th International Conference on Internet of Things, Big Data and Security
228