Applying IoT technologies to analyse sleeping pattern

WHITE PAPER :
SLEEP ANALYTICS
By
Communications Business Unit

Are you getting enough
sleep?
Your answer to this question has significant impacts on your
heart, weight, mind, and much more. During sleep, your mind is
super busy working to improve memory, lower stress and
anxiety, sharpen attention and other factors that could better
overall well-being and quality of life. A healthy sleep rejuvenates
your immune, nervous and muscular systems.
The fact that too much or too little sleep is associated with a
shorter lifespan makes it all the more imperative to maintain a
healthy sleeping pattern.
While we all agree that a good night sleep makes our day more
pleasant and successful, many of us are not getting enough
sleep due to varied reasons. The first step towards getting a
good night sleep is figuring out what is wrong with your current
sleep situation.
Sleep Monitoring
Sleep monitoring solution analyses the sleeping pattern of a
person (considering age, gender, health conditions, geographi-
cal position etc.) and evaluates the sleeping habit. For this
purpose, a force sensing resistor that monitors respiration rate
and posture movements during sleep is employed. The force is
converted to a voltage level and compared with a threshold
value, on the basis of which it is determined if the person is in
deep sleep, light sleep or not in bed. With the obtained data, the
proposed solution will calculate the sleeping efficiency and
sleep score of an individual and provide the results.
Schematic of Bed Sensor
The force exerted by the person lying on the bed is sensed by
FSR placed on the bed in a strip form along the shoulder of the
person. The force is converted to voltage levels by a transducer.
These analog voltage levels are converted to the digital form
and sampled. The sampled values are aggregated and conver-
ted to symbols with reference to a threshold value.
© SFO Technologies Private Limited, 2016-2017. Unauthorized use and/or duplication of this material without express and written permission from this paper’s author and/or
owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to SFO Technologies Private Limited with appropriate and specific
direction to the original content.
Sleep analysis
Sleep is a recurring, periodic and reversible state of mind and
body in which there is reduced or absent consciousness, dimini-
shed sensory and muscular activity and reduced responsive-
ness to outside stimuli.
Stages of sleep
Sleeping is classified into two types; NREM (Non- Rapid Eye
Movement) and REM (Rapid Eye Movement) sleep. NREM
consists of three separate stages; Stage 1, Stage 2 and Stage 3.
Stage 1 or NREM1 or N1 is the stage in which the muscles are
still quite active; eyes roll around slowly and may open and
close from time to time. During this stage, many people
experience sudden muscle contractions preceded by a sensa-
tion of falling. This is a short period of very light, easily disrup-
ted sleep, usually lasting less than 10 minutes, i.e., about 5% of
the total sleep time.
Stage 2 or N2 is the one in which eye movement stops, muscle
activity still decreases further and conscious awareness of the
external world begins to fade completely. There is also a drop in
body temperature and heart rate and breathing becomes
regular. This constitutes 45%-50% of total sleep time.
Stage 3 or N3 is a deeper stage of sleep with even lesser
responsiveness to stimuli and full relaxation of the muscle.
Blood pressure and breathing rate drops and muscles relax
completely. This is a transitional period between light sleep and
a very deep sleep. Sleep walking also happens during this
stage. It occupies 15%-20% of total adult sleeping time.
In REM or Stage 4 sleep, muscles become immobilized or
paralyzed and completely unresponsive. This stage is marked
with a rapid eye movement, increased brain activity, increased
heart rate and increased blood pressure. Most of the memo-
rable dreams occur during this stage. Body loses some of its
ability to regulate temperature. Infants spend almost 50% of
their time in REM sleep. Adults spend 20% to 25% of their
sleeping period in REM sleep. Older adults spend progressively
lesser time.[3]
FSR Transducer ADC Sampler Aggregation
Threshold level
Fig 1: Bed Sensor Schematic

Sleep cycle
Sleep does not progress through all of these stages in sequen-
ce. Sleep begins with Stage 1 and progresses through stages 2
and 3. After stage 3 sleep, stage 2 sleep is repeated before
entering REM sleep. Once REM sleep is over, the body usually
returns to Stage 2 sleep. Sleep cycles during these stages
approximately four or five times throughout the night.
The first cycle of REM often lasts only a short amount of time,
but each cycle becomes longer. This is why we need long
periods of sleep each night. If we get short periods of sleep, we
can’t really get through the stages we need to heal and stay
healthy. [4]
Recommended hours of sleep
During the busy schedule of our present life sacrificing an hour
or two of rest in order to get more work done may sound like a
reasonable trade-off. But the truth is that even minimal sleep
loss takes a toll on our mood, energy, mental sharpness, and
ability to handle stress. And over the long-term, chronic sleep
may affect our health conditions. The quality of one’s sleep
directly affects the quality of waking life, including their
productivity, emotional balance, creativity, physical vitality, and
even body weight.
The recommended hours of sleep for different age groups are
as follows:
(a) According to age group
(b) According to gender
Generally, adult men and women require 7 to 8 hours of sleep at
night. Women often sleep more than men and experience a
lighter sleep that is more easily disrupted. Their sleep is disrup-
ted due to problems like depression, pregnancy, hormonal
changes related to menopause, sleep disorders and medical
problems like arthritis, back pain etc.
(c) According to geography
The recommended hours of sleep for an average individual at
any part of the world are 7 to 8 hours. However the actual hours
of sleep are different for each of the listed places.
Age group
Newborn (0-3 months)
Infant (4-7 months)
Toddler (1-2 years)
Preschool (3-5 years)
School age (6-13 years)
Teen (14-17 years)
Young adult (18-25 years)
Adults (26-64 years)
Older adult (65+)
11-13
10-11
9-10
8-9
7-8
7
6
6
5-6
14-17
12-15
11-14
10-13
9-11
8-10
7-9
7-9
7-8
18-19
16-18
15-16
14
12
11
10-11
10
9
Appropriate lower limit
of sleeping hours
Recommended
sleeping hours
Appropriate upper limit
of sleeping hours
Country
India
Russia
South Africa
USA
UK
France
UAE
Australia
18-19
16-18
15-16
14
12
11
10-11
10
Avg.Sleep Hours
In extreme regions like polar areas, the sleep pattern and sleep
hours vary from person to person and the average hours of
sleep is not estimated properly. Due to the cycle of 6 months of
winter followed by 6 months of summer, most of the people
suffer from Insomnia. Period of stages 1 and 2 of sleep decrea-
ses whereas delta sleep increases. [7]
Table 1: Sleep Data according to age group [5]
Table 2 : Sleep Hours on country basis [6], [8], [9], [10]
Stage 2 NREM
45%Stage 3 NREM
25%
Stage 1 NREM
5%REM
25%
Duration of Sleep Stages
Fig 2 : Pie chart for sleep stages

Sleep Disorders
A Sleep disorder is a medical disorder of the sleep patterns of a
person. The normal sleeping problems recognized are:
(a) Insomnia
The common symptoms include difficulty falling asleep or
getting back to sleep after waking during the night, waking up
frequently during the night, sleep seems to be light, fragmented,
or non refreshing, inability to sleep without taking sleeping pills,
nightcap, supplements etc, sleepiness and low energy during
the day.[11]
(b) Sleep Apnea
A breathing sleep disorder [16] in which our breathing tempora-
rily stops (10 seconds) during sleep due to the blockage of the
upper airways. Apnea episodes may occur from 5 to 100 times
an hour. These pauses in breathing interrupt one’s sleep,
leading to many awakenings each hour. Loud and chronic
snoring, Gasping, snorting, or choking during sleep, Waking up
with shortness of breath, chest pains, headaches, nasal conge-
stion, or a dry throat.
(c) Movement disorders
(i) Periodic limb movements
These are movements as brief muscle twitches, jerking
movements or an upward flexing of the feet. They cluster
into episodes lasting anywhere from a few minutes to
several hours.
(ii)Restless leg syndrome
It is a disorder that causes a strong urge to move one's legs.
There is often an unpleasant feeling in the legs that improves
somewhat by moving them. Occasionally the arms may also
be affected.
(iii)Teeth Grinding
Sleep Bruxism also known as nocturnal tooth grinding, is the
medical term for clenching or grinding teeth during sleep.
People who have sleep bruxism can also suffer headaches,
earaches, jaw pain, jaw joint disorders and damaged teeth.
[12]
Factors affecting sleep
(a) Age
Sleeping hours and sleeping pattern vary with age as noted
earlier.
(b) Dietary habits
Studies say that Dietary amino acid acts to promote sleep. A
small protein containing snack before bedtime used to be
recommended for patient with insomnia.
(i) Alcohol Intake
Alcohol beverages, when used in moderation, appear to
induce sleeping in some people. However, large quantities
have been found to limit REM and Delta (stage 2) sleep. This
effect may partially explain the phenomena of hangover
after excessive alcohol consumption.
(ii) Caffeine containing Beverages
Caffeine is a central nervous system stimulant. For many
people, beverages containing caffeine interfere with the
ability to fall asleep. Beverages containing caffeine include
coffee, tea and most cola- drinks and chocolates.
(iii) Smoking
Nicotine has a stimulating effect on the body and smokers
often have more difficulty falling asleep than non- smoker
do. Smokers are usually easily aroused and often describe
themselves as light-sleepers.[13]
(c) Shift works
It is estimated that almost 10% of shift workers experience Shift
Work Sleep Disorders (SWSD).This causes conditions of Insom-
nia, Excessive daytime sleepiness and non-refreshing sleep.
The hazards of shift works include restlessness, fatigue, decre-
ased attention and disruption of the body’s metabolic process.
(d) Pain, anxiety and depression
A wide range of medical conditions including chronic pain from
arthritis and other medical conditions, discomfort caused by
gastro esophageal reflux disease, pre-menstrual syndrome, and
many others tend to limit the depth of sleep and allow only brief
episodes of sleep between awakenings.
Individuals of all ages who experience stress, anxiety, and
depression tend to find it more difficult to fall asleep, and when
they do, sleep tends to be light and includes more REM sleep
and less deep sleep. This is likely because our bodies are
programmed to respond to stressful and potentially dangerous
situations by waking up.
Fig 3 : Sleep Apnea Diagram [16]
Apnea Sleep Graph - One Minute Shown

(e) Use of drugs
Dozens of prescribed drugs that are used to control common
disease symptoms may have varying effects on sleep. Beta
blockers, which are used to treat high blood pressure, congesti-
ve heart failure, glaucoma, and migraines, often cause decrea-
ses in the amount of REM and slow-wave sleep(stage 3), and
are also associated with increased daytime sleepiness. Alpha
blockers, which are also used to treat high blood pressure and
prostate conditions, are linked to decreased REM and increased
daytime sleepiness. Finally, antidepressants, which can decrea-
se the duration of periods of REM sleep, have unknown
long-term effects on sleep as a whole. Some antidepressants,
from the class of drugs known as SSRIs, have been found to
promote insomnia in some individuals.
(f) Exercise
Regular exercise is known to be good for helping people sleep
properly, but vigorous exercise during the evening can prevent
sleep.[14]
Breath pattern for normal sleep
Ventilation is significantly lower during all stages of sleep than
in wakefulness.
Breathing is remarkably regular in steady NREM sleep. Minute
ventilation decreases by 13% in steady stage 2 sleep and by
15% in steady slow wave sleep (Stage 3). There is an overall
decreased tidal volume. Diaphragm activity is little increased
or unchanged and abdominal muscle activity is slightly
increased during these sleep stages. [15]
Irregular breathing with sudden changes at times interrupted
by central apneas lasting 10–30 seconds are noted in REM
sleep. These breathing irregularities are not random, but
correspond to bursts of eye movements. Breathing during REM
sleep is somewhat discordant. Diaphragmatic activity corre-
spondingly increases during REM sleep.
Sleep analytics
Due to the space constraints of the memory units required and
the transmitting channel constraints, the analytics is classified
to two sections
(a) Edge analytics
(b) Server/Cloud analytics
In the Edge analytics, the voltage values sensed from the bed
sensor are sampled at an appropriate sampling rate. These
samples are aggregated and the aggregates are mapped onto
specific symbols according to their range. The symbols are
then transmitted to the server side for further analysis. This
will give an idea about the status of sleep, i.e. deep sleep, light
sleep or out of bed condition.
The Server/Cloud analytics serves the purpose of calculating
the sleeping efficiency and the sleep score on the basis of the
data obtained from edge analytics.
Sleep efficiency is a measure of how effective a person’s sleep
is. It equals the combined time of deep and light sleep divided
by the total time spent in bed. A sleep efficiency of 85% is
considered to be normal and really good sleep efficiency is
above 90%.
Sleep score can be calculated using two methods on the
basis of:
(a) The recommended hours of sleep of a person(Score I)
(b) Previous sleep data of the same person(Score II)
In the first method, we compare the person’s overall sleep to 8
hours. Sleeping hours slightly less than 8 hours is given
positive index and that greater than 8 hours is given negative
index. 6 to 9 hours of sleep is acceptable.
The other method checks the previous week’s sleeping pattern
of a person and compares the median value of the previous
week with the present sleeping pattern to calculate the sleep
score.
Hence the overall sleep score is drawn from Score I & Score II.
Results
Bed Sensor Plots
Fig 4 : Actual Data obtained from Bed Sensor
Fig 5 : Approximation of actual data
From these plots, it is observed that Fig 5 is a close approxi-
mation of Fig 4, i.e. approximations done on the actual values
obtained from the bed sensor are close to the actual values.

Fig 6 : Plot of breath pattern obtained from (a)
Fig 7 : Normal Breathing Pattern
Fig 8 : Sleep screen I
Fig 9 : Sleep screen II
Breath pattern obtained
Fig 6 gives the breathing pattern during sleep obtained from Fig
4.
Fig 7 is a plot of the normal breathing pattern which is obtai-
ned from reliable sources. On comparison of Fig 6 and Fig 7,
we see that there is a close resemblance of the results
obtained.
Intuitive mobile applications that present the
sleeping data as meaningful insights
Availability of intelligent mobile applications that will present
the sleeping pattern and analysis in easy to understand
dashboards, make it really quick to recognize what could be
negatively affecting your sleep and how to improve it
Factors like your sleep score and sleep efficiency will be
displayed along with details of your daily sleep hours. By
analyzing the sleep chart you can understand how exactly you
are sleeping.
Conclusion
With increasing public awareness of the negative impact of
poor sleep and insomnia on health, the time has come to take
a much more proactive stance in promoting sleep as a critical
determinant of health. Our work acts in this direction by giving
an idea about how effective our sleep is and how much one
should move on to have a good sleeping pattern. It is observed
that, the results obtained from our work using bed(force)
sensors satisfies the actual sleeping pattern of individuals and
are accurate to certain extent. The work can be extended
further to recognize the different stages of sleep and to detect
sleeping disorders like sleep apnea, movement disorders etc.
For business enquiries write to us at
biz.commsbu@nestgroup.net

Abbreviations
• N- Newton
• ADC-analog to digital converter
• REM- Rapid eye movement
• NREM- Non rapid eye movement
• SSRI- Selective serotonin reuptake inhibitor
Reference
[01] http://www.convertunits.com/from/N/to/kg
[02] https://www.sparkfun.com/datasheets/Sensors/Pressure/fsrguide.pdf
[03] http://www.howsleepworks.com/types_nonrem.html
[04] https://en.wikipedia.org/wiki/Sleep
[05] https://sleepfoundation.org/sites/default/files/STREPchanges_1.png
[06] http://www.dailymail.co.uk/sciencetech/article-3042230/Slee-
ping-habits-world-revealed-wakes-grumpy-China-best-quality-shut-eye-South-Africa-wakes-earliest.html
[07] https://www.researchgate.net/publication/224004070_Biological_Rhythms_During_Residence_in_Polar_Regions
[08] http://www.sleepcouncil.org.uk/wp-content/uploads/2013/02/The-Great-British-Bedtime-Report.pdf
[09] http://www.sleephealthfoundation.org.au/pdfs/sleep-week/SHF%20Sleep%20Survey%20Report_2015_final.pdf
[10] http://www.news.com.au/lifestyle/real-life/news-life/whe-
re-you-live-affects-how-much-you-sleep-according-to-new-study/news-story/992551dd7b89cdc806fe0ef5e5ce8399
[11] http://www.helpguide.org/articles/sleep/sleep-disorders-and-sleeping-problems.htm
[12] https://sleepfoundation.org/sleep-disorders-problems
[13] https://www.scribd.com/doc/22904067/Factors-Affecting-Sleep-The-Quality-and-Quantity-of-Sleep-Are
[14] http://www.dailymail.co.uk/health/article-84207/More-factors-affects-sleeping-patterns.html
[15] https://en.wikipedia.org/wiki/Sleep_and_breathing
[16] http://www.vaughns-1-pagers.com/medicine/sleep-apnea.htm

Applying IoT technologies to analyse sleeping pattern

More Related Content

What's hot (18)

Similar to Applying IoT technologies to analyse sleeping pattern (19)

Recently uploaded (20)

Applying IoT technologies to analyse sleeping pattern