2022 International Conference on Broadband Communications for Next Generation Networks and Multimedia Applications (CoBCom) | 978-1-6654-8598-2/22/$31.

00 ©2022 IEEE | DOI: 10.1109/COBCOM55489.2022.9880629

DynaSens: Dynamic Scheduling for IoT Devices

Sustainability
Neda Maleki, Arslan Musaddiq, Daniel Toll, Francis Palma, Tobias Olsson, David Mozart,
Mustafa Omareen, and Fredrik Ahlgren( )
Applied IoT Lab, Department of Computer Science and Media Technology,
Linnaeus University, Sweden
[email protected]

Abstract—The Internet of Things (IoT) have shown nu- IoT plays an important role in the smart city by
merous potential applications that can enhance our quality of providing applications to monitor real-time data of city
life. IoT is becoming a core technology to bring smart homes, processes. One example of an IoT-based smart city ap-
smart cities, and smart industries into reality. However, with
potential benefits comes a challenge of sustainability, and plication is a smart waste collection process [7]. Waste
one major concern is to minimize energy consumption. In a disposal is a major challenge for cities. According to the
citywide area, managing the operation of such large-scale World Bank, the average daily amount of waste generated
IoT networking is one of the complex tasks. One of the worldwide is more than two billion tonnes, and the amount
ways is to utilize dynamic sensing scheduling where the IoT is increasing every year [8]. Nowadays, in Sweden, waste
device goes to the sleep mode and prevents unnecessary data
transmission. In this paper, we propose a dynamic sensing collection systems based on IoT technology is an emerging
(DynaSens) algorithm for an IoT-based waste management technology and is installed across various locations in
system. This algorithm helps to reduce the waste bin over- the country [9]. By utilizing a smart waste collection
flowing, thus, provides better sanitation, and it is also helpful management system, it is possible to check the amount
in reducing the fuel cost of waste collection vehicles. Our of garbage in near real-time, thus preventing overflow of
work utilizes measured values such as current consumption,
LiDAR measurement time, and LoRa transmission time as garbage and increase the collection efficiency.
the input data for the simulation experiment to evaluate In a system such as smart waste collection, IoT sen-
energy consumption. We also assessed DynaSens using a sors act periodically leading to higher sensing cost and
real dataset obtained from a recycling house. We use Py- minimized network lifetime. Striking a balance between
com LoPy4 micro-controller as a development board. For device performance and energy efficiency is a challenging
a number of garbage-thrown scenarios, DynaSens enables
longer battery longevity by reducing the repeated execution task in optimizing the energy efficiency. Avoiding repeated
of the same tasks. executions of the same tasks are highly desirable for IoT
Index Terms—IoT, Dynamic Sensing, Scheduling, Energy systems.
Efficiency In this paper, we propose dynamic sensing (DynaSens)
algorithm which provides an efficient sensing scheduling
I. I NTRODUCTION for IoT devices. Using the DynaSens, we performed an
experimental analysis of an IoT-based waste collection
Internet of Things (IoT) allows communication between
system to provide an efficient dynamic sensing scheduling
devices to connect our physical and digital world. These
algorithm. The proposed algorithm provides waste bin-
devices contain a radio chip, micro-controller and one or
related data to the concerned organization. Using the data,
multiple sensors and actuators [1]. The use of IoT devices
organizations responsible for waste collection can plan the
for numerous applications have increased significantly
disposal process more efficiently. The proposed solution
after the development of low-power radio communication
does not only increase sustainability of the IoT network
protocols such as Long Range (LoRa), Narrowband IoT
for the long term but it is also useful in reducing carbon
(NB-IoT), Sigfox, Bluetooth Low Energy (BLE), and
footprint left by waste collection vehicles for a green
ZigBee [2].
environment.
From wireless personal area networks to wide-area
The main contributions of this study are as follows:
networks, IoT has shown numerous potential applica-
tions. These applications include smart homes, smart • We proposed DynaSens algorithm that provides effi-

cities, smart grid, and smart industries [3]. IoT devices cient sensing capabilities to the IoT devices.
have limited computational capabilities and for most of • We applied and evaluated DynaSens to the waste

the application scenarios, these IoT devices are battery- monitoring system.
powered. These resource-constrained sensor-based devices The rest of the paper is organized as follows: Sec-
often operate in lossy and harsh environments [4]. The tion II provides an overview of the related work. Section
lossy environment and limited energy capacity may hinder III describes the waste monitoring use case. Section IV
the true potential of IoT devices. Maintaining battery- explains the proposed framework. Section V highlights the
operated devices for city-wide area network coverage is experimental results followed by Section VI to conclude
one of the main challenges IoT network faces [5], [6]. the paper and provides future work.

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SURATHKAL. Downloaded on January 19,2024 at 14:11:45 UTC from IEEE Xplore. Restrictions apply.
 II. R ELATED W ORK Imam et al. [15] presented a smart waste monitoring
and management system which can smartly handle the
IoT has been implemented in various areas, such as
waste management problem. The proposed system uses
management (e.g., waste collection and disposal) system,
one ultrasonic sensor and three gas sensors to detect the
segregation (e.g., separation of wet and dry wastes),
garbage level inside and the irritating gas level around
scheduling and routing (e.g., when and where the waste
the dustbin, respectively. However, by applying Wi-Fi as
bin trucks start and pick up the wastes), optimization (e.g.,
the communication the system cannot support the city
waste monitoring and warning), and design and evaluation
coverage (scalability) and is not power-efficient.
(e.g., design and configuration of the system architecture,
both software and hardware components, according to the III. U SE C ASE
identified functionality and evaluation of the performance
City-wide waste management is a major challenge faced
of the entire system, based on data from years of opera-
by municipalities. An IoT-based remote waste manage-
tion).
ment system has been proposed by numerous researchers
In [10], authors proposed a top-k query-based dynamic to manage waste collection efficiently [16]. Sustaining
scheduling model to address the challenges of near real- such a system for the long term is another challenge
time scheduling driven by sensor data streams. An Android that needs to be solved. To increase sustainability we can
app along with a user-friendly GUI is developed and pre- improve the sensing efficiency of the devices by preventing
sented to prove feasibility and evaluate a waste collection unnecessary data transmissions.
scenario using experimental data. However, their proposed In a traditional setup, municipal waste bins are mon-
algorithm is not adaptive, i.e., the sensing interval does not itored in a regular interval, which might not be energy
change based on how frequently the waste level changes, efficient for the operating agency. In a smart monitoring
and it locks the lids when the bin threshold is reached. solution, sensors can be used to read waste bin level and
An automated smart garbage monitoring and disposal various gas particles to perceive collection time. However,
support system (SGMDSS) is proposed by [11]. SGMDSS in this context, although the energy consumption can be
monitors the waste bins placed at different locations and minimized for the operating agency, the energy efficiency,
notifies the cleaning personnel about the level of garbage thus, sustainability of the IoT devices are not considered.
accumulated through an android mobile application. Using Thus, a smart solution is required where a sustainability
this notified information, the personnel takes the shortest concern is present both for the operating agency and the
path to the waste bin location that is almost filled. This IoT devices.
information is also sent to a webpage and the entire data In this study, we focus specifically on the waste collec-
is stored and accessed through the cloud. Also, an alert tion use case and apply DynaSens to the waste monitoring
message is sent to the worker. However, the area coverage system.
and the system cost are the flaws compared to LoRa tech- In the following section, we propose a dynamic sensing
enabled systems. algorithm where the system can be adjusted to save energy
Husni et al. [12] proposed an IoT-based warning system based on the waste throwing activity.
where three robots are used to sense and send data to a
mobile application and notify the corresponding authority IV. P ROPOSED S OLUTION
if the bins are filled. However, the proposed system is In this section, we present the overview of the proposed
based on Wi-Fi which is not usually available at all solution and the details of DynaSens algorithm for the
locations. The power consumption and the expenditure energy efficiency for the low-power and lossy IoT devices.
of implementation are also relatively high compared to
LPWAN networks. A. Overview of the Proposed Solution
Ramson et al. [13] proposed an IoT system to monitor Energy consumed by IoT devices remains a topic of
the trash level and geo-location of waste bins efficiently. interest because of its long-term sustainability impact. The
Authors have evaluated the accuracy of the sensor em- proposed DynaSens algorithm uses an adaptive sensing in-
ployed, maximum transmission distance between an IoT terval to enhance device battery lifetime. DynaSens detects
device and a gateway, life expectancy of the IoT device, the garbage level and schedules the next sensing interval
battery charging time, and cost. However, the proposed so- based on the change in garbage level. In our proposed
lutions utilizes ultrasonic sensor which provides inaccurate solution, the IoT device is equipped with light detection
readings due to temperature variation in the environment. and ranging (LiDAR) sensor [17]. A LiDAR sensor emits
In [14], authors proposed an IoT-based smart trashcan pulsed light waves, these waves bounce off the nearby
monitoring system, which is composed of smart trashcans objects and return to the sensor. The sensor then measures
and a cloud-based monitoring platform. The proposed the time it takes for each pulse to return to the sensor to
system can detect the current garbage capacity of the trash measure the distance between the sensor and the object.
cans. The proposed smart trash can automatically turn on The LiDAR sensor is connected to the Pycom LoPy4
if the hand is sensed to prevent users from unnecessary development board. The LoPy4 is a multi-communication
contact. However, the proposed system does not consider technology-enabled device (i.e., LoRa, Sigfox, WiFi, and
the dynamic sensing and motor used in the system adds Bluetooth) and is programmed by MicroPython [18]. In
an extra cost and energy consumption. DynaSens, the node is utilizing the LoRa communication

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SURATHKAL. Downloaded on January 19,2024 at 14:11:45 UTC from IEEE Xplore. Restrictions apply.
 TABLE I Algorithm 1 DynaSens: Dynamic Sensing Scheduling
L IST OF N OTATIONS Require: Stmin = 30 s, Stmax = 86 400 s, T hmax =
90 %, bin size = 100 cm, Wl = 0
Parameters Labels
1: while Device is ON do
St Current sleeping interval 2: Wl , isConsistent ← waste level(Wl )
Stmin Minimum sleeping interval
Stmax Maximum sleeping interval 3: if isConsistent = TRUE then
T hmax Waste level maximum threshold 4: St ← St × 2
Wl Current waste level 5: if (St > Stmax ) then
6: St ← Stmax
7: end if
technology, which is an open network for long-range IoT 8: else if isConsistent = FALSE then
devices communication [19]. In our testbed, we used USB 9: St ← Stmin
to serial transistor-transistor logic (TTL) to program the 10: if Wl >= T hmax then
LoPy4. 11: send notification()
B. DynaSens: Dynamic Sensing Scheduling Algorithm 12: end if
13: end if
This subsection introduces the DynaSens algorithm to 14: sleep()
improve the energy consumption of sensor nodes. Table I 15: end while
describes the list of notations used in this algorithm.
Sleep scheduling based on waste level consistency or
inconsistency enhances energy efficiency significantly. Al- Simulated Waste Throw Patterns
gorithm 1 describes the proposed DynaSens scheduling. 1400 60s Throw
Cyclic
DynaSens initiates its operation by initializing minimum 1200 NoThrow
Total Simulated Waste Thrown

sleeping time Stmin = 30 s, maximum sleeping time

Stmax = 86 400 s , and maximum garbage level threshold 1000
T hmax =90 %. In the waste level function the LiDAR 800
sensor measures the distance to the garbage to calculate 600
the garbage level Wl of the bin (line 2). Internally, the Wl
is compared with the previous measurement and this is 400
returned and stored as True or False in isConsistent. If 200
isConsistent is True it means the same waste level has
0
been measured again.
0 20000 40000 60000 80000
If there is no change in garbage level compared to Time passed seconds
the previous iteration (line 3), the sleeping interval St is
doubled for the next iteration (line 4). The St defines the Fig. 1. Frequency of trash thrown over time in the three scenarios.
amount of time in seconds for which the sensor and micro-
controller remain in sleep mode. In case St reaches above
its maximum value (line 5), it is kept at Stmax (line 6). that sleep with a fixed sleep interval, i.e., Sleeper-30s and
During each iteration, If the waste level changed, the sleep Sleeper-60s.
time St will be reset to its minimum value (line 9). In case Since the sleep pattern depends on whether the sensing
the Wl is more than T hmax (line 10), the algorithm sends value changed or not, three different scenarios were cre-
the notification to the concerned agency so they can empty ated, each with different waste throwing patterns. These
the bin (line 11). We evaluate DynaSens in the following can be seen in Figure 1. In the first N oT hrow scenario,
section. nothing is thrown. This is the scenario where DynaSens
should outperform algorithms with fixed intervals. In the
V. P ERFORMANCE E VALUATION next scenario, trash is thrown at a regular intervals of
We evaluate DynaSens in two parts: 1) simulation using 60 s. The scenario was chosen as a worst-case scenario
synthesized data, where we measure time and current for the DynaSens Algorithm since an algorithm that sleeps
on micro-controllers running the DynaSens and then use for the same interval should behave better compared to
these measurements to simulate different scenarios. 2) DynaSens. In the final cyclic scenario, trash is thrown in
simulation using real dataset, where we use real data from regular intervals, and then there are intervals when nothing
an environmental house with six recycling bins equipped is thrown. Such a scenario shows the benefit of being able
with the sensors in Kalmar, Sweden. The data is collected to adopt the sleep pattern.
from December 2021 to the end of May 2022 [20]. Using an Otii Arc, we measure the average awake time
power consumption as well as the awake time for running
A. System Setup the algorithms. Since these values are different depending
The source code of the DynaSens algorithm can be if the algorithms use the LoRaWAN or not, we have
downloaded from [21]. A detailed analysis of the proposed separated measurements for that. We also measure the idle
technique is done by comparing DynaSens to algorithms power when the device is sleeping, which is the same for

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SURATHKAL. Downloaded on January 19,2024 at 14:11:45 UTC from IEEE Xplore. Restrictions apply.
 TABLE II 5V
M EASURED VALUES FROM E XPERIMENTAL S ETUP I LiDAR
Vin
A sensor t
jec
1 28

ob
2 27

Parameters DynaSens Sleeper-30/60 3

4
26
25
Distance
5 24 NPN
Algorithm Sleep time 30 s - 86 400 s 30 s or 60 s 6 23
Transistor
Awake Time Not Sending 4.292 s 4.22 s V
Otii V 7 Pycom 22
8 LoPy4 21
data
Awake Time Sending 11.226 s 11.35 s 9
10
20
19
A : Ampere meter
Sleep Mode Current 1.1 mA 1.1 mA 11 18 R V : Voltage meter
12 17
Average Awake Not Sending Current 64.5 mA 64.7 mA 13 16 1 KΩ
14 15
Average Awake Sending Current 44.1 mA 44.4 mA GND
Awake Time Joining LoRaWAN 18.712 s 16.816 s

Fig. 2. Overview of system circuit diagram.

TABLE III
S IMULATION OF WASTE -T HROWN F REQUENCY S CENARIOS
consumption analysis.
Parameters No Throw 60s Throw Cycle
Trash throw frequency n/a 60 s 60 s
B. Analysis of Results
Cycle length n/a n/a 6h In this subsection, we present analysis results of Dy-
Run time 24 h 24 h 24 h
naSens in comparison to other sleeping patterns in dif-
ferent garbage-thrown scenarios. During sleep mode, the
TABLE IV LiDAR sensor and all the radios are OFF as well as CPU
E XPERIMENT PARAMETERS and RAM. For result analysis, we utilize the measured cur-
rent values for energy consumption calculation as follows:
Parameters DynaSens Sleeper
Micro-controller Pycom LoPy4 Pycom LoPy4
Sensor Type TF-Luna LiDAR TF-Luna LiDAR
Communication Technology LoRaWAN LoRaWAN Etotal = Esleep + Eawake + Etransmit (1)
Waste Bin Size (heightß) 100 cm 100 cm
Sensing Interval Min Stmin 30 s 30 s Esleep = Vcc + Isleep + Tsleep (2)
Sensing Interval Max Stmax 86 400 s -
Waste-level Max Threshold T hmax 90 % - Eawake = Vcc + Iawake + Tawake (3)
Voltage Supply 5V 5V
Etransmit = Vcc + Itransmit + Tawake (4)

both algorithms since it depends on hardware as well as where E, V , I, and T represents the energy consumption,
the sleep function used. voltage, current, and time, respectively.
To get accurate measurements, we connect the UART Figure 3 depicts the DynaSens algorithm clearly is
of the micro controller to the Otii Arc and use print favorable when the scenario includes long periods when
statements to find the start and stop measurements. The no waste is thrown. Similarly, in Figure 4, when garbage is
measured time can vary when joining the LoRaWAN thrown in a cyclic pattern, DynaSens still performs better
network. We give examples of that also in Table II. compared to Sleeper-30s and Sleeper-60s. However, in a
However, joining the LoRaWAN network is normally only scenario when waste is thrown every 60s interval, as shown
done once and is independent of the algorithm used to in Figure 5, we can see that DynaSens consumes energy
schedule sleep. The values shown in Table II are fed into more than the Sleeper-60s algorithm since DynaSens will
a simulation where the different scenarios are running for restart with 30s sleep every time the sensor value changes.
24 h of simulated time for each scenario and algorithm. Nevertheless, DynaSens energy consumption is less than
During simulation, the different functions waste level, Sleeper-30s because DynaSens doubles the sleeping period
send notif ication, and sleep are replaced by simulation after the 30 s when no garbage activity is detected.
functions that interact and get their outputs from the From the worst-case scenario (see Figure 5), we see that
scenario. DynaSens may consume more energy than an algorithm
Table III shows the simulated scenarios where there are that is in sync with the waste throwing pattern. From
different garbage-thrown frequencies. this, we conclude that DynaSens is a good choice for
In all scenarios, the waste bin is of 100 cm in height, adopting to the No Throw and Cyclic waste throwing
and the TF-Luna LiDAR module measures the garbage pattern. However, only optimizing an algorithm for battery
level by emitting pulsed light waves that bounce off the consumption does not work efficiently, as seen in Figure
object surface. The LiDAR sensor uses the time difference 6. In this figure, the waste bin overflows since DynaSens
between transmitted and received pulse to measure the oversleeps, which is not expected for sanitation. This
distance. Table IV shows the experimental parameters used problem can be solved by reducing the T hmax from 90 %
in this study. to 70 %. Figure 7 shows the empty pattern of DynaSens
The circuit schematic of DynaSens is shown in Figure algorithm using T hmax of 70 % and 90 %. We can see
2. As illustrated in Figure 2, we utilize Otii Arc, which from the graph that DynaSens faces one overflow if
is a power supplier and power analyzer tool for energy T hmax is 90 %. However, there is no overflowing if a

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SURATHKAL. Downloaded on January 19,2024 at 14:11:45 UTC from IEEE Xplore. Restrictions apply.
 No Throw Scenario off is that the bin might be emptied more often, thus
4000 increasing the overall cost on the environment by requiring
DynaSens
3500 Sleeper-30s more frequent empty actions.
Sleeper-60s
Energy Consumption (Joule)

3000
2500 Empty Pattern in Cyclic Scenario
2000 DynaSens
100 Sleeper-30s
1500 Sleeper-60s
80

Waste Level in Bin %

1000
500 60
0
0 20000 40000 60000 80000 40
Time passed seconds
20
Fig. 3. No Trash Thrown scenario. Energy consumption of DynaSens
in is favorable to sleeping in regular intervals. 0
0 20000 40000 60000 80000
Cyclic Scenario Time passed seconds
4000
DynaSens
3500 Sleeper-30s Fig. 6. In the Cyclic scenario, DynaSens misses an empty cycle and the
Sleeper-60s waste bin overflows due to oversleeping.
Energy Consumption (Joule)

3000
2500
2000 70% Threshold Cyclic Scenario
DynaSens Threshold 70
1500 100 DynaSens Threshold 90
1000
80
Waste Level in Bin %

500
0 60
0 20000 40000 60000 80000
Time passed seconds 40
Fig. 4. In the Cyclic scenario, DynaSens can sleep longer during the 20
period when no waste is thrown, thus can energy be saved.

0
60s Throw Scenario 0 20000 40000 60000 80000
4000 DynaSens Time passed seconds
3500 Sleeper-30s
Sleeper-60s
Energy Consumption (Joule)

Fig. 7. With reduced threshold, DynaSens causes more empties.

3000
2500
2000 Empty Pattern in 60s Throw Scenario
1500 DynaSens 70% Threshold
1000
Sleeper-30s
80 Sleeper-60s
500
Waste Level in Bin %

0 60
0 20000 40000 60000 80000
Time passed seconds
40
Fig. 5. 60s throw scenario, show similar behavior between Sleeper-60s
and DynaSens, despite DynaSens minimal sleep time of 30s. 20

0
T hmax is reduced to 70 %. Similarly, in Figure 8, we can 0 20000 40000 60000 80000
see that the bin empty pattern for 60 s Throw scenario Time passed seconds
for DynaSens having T hmax of 70 % in comparison with
Sleeper-30s, and Sleeper-60s. In the 60s Throw scenario, Fig. 8. With reduced threshold, DynaSens causes more empties.
DynaSens still faces no overflowing. However, the trade-

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SURATHKAL. Downloaded on January 19,2024 at 14:11:45 UTC from IEEE Xplore. Restrictions apply.
 C. Real Data Evaluation Paper packaging Scenario
The six months of data obtained from the environmental DynaSens TH90
100000 Sleeper-15m
recycling house [20] is shown in Figure 9. The data depicts Sleeper-60m

Energy Consumption (Joule)

the filling and empty pattern of the recycling house paper 80000
packaging bin from Dec 2021 to June 2022. Smaller
LiDAR sensor value depicts the bin is more full and vice 60000
versa. The sensor transmits the data with 15 min interval
using LoRaWAN. 40000

20000
Paper Packaging Raw Data from Recycling Center
Paper packaging 0
0 25 50 75 100 125 150 175
1600 Time passed days
Lidar/Sonar Sensor value mm

1400 Fig. 10. DynaSens saves more energy compared to the two other sleeping
algorithms in real dataset scenario.

1200 Waste level in Bin With Paper Packaging Data

100
DynaSens TH60
1000 Waste level in Bin (Percent of full bin)
80
800
60
600 40

2021-12 2022-01 2022-022022-03 2022-042022-05 2022-06 20

Date
0
Fig. 9. Real waste throwing pattern from six months of Paper Packaging
recycling. 0 25 50 75 100 125 150 175
Time passed days
Since the data is transmitted at 15 min intervals, we
compare DynaSens with an algorithm that sleeps for Fig. 11. Waste level in bin with DynaSens with 60% threshold.
15 min each interval, i.e., Sleeper-15m. For further inves-
tigation, we compared DynaSens with a longer sleeping
the device’s energy consumption however, the algorithm
interval algorithm, i.e., Sleeper 60min. From Figure 10,
is sensitive to overflowing the trash bin when sleep time
we can see for waste bin real data, DynaSens is still
becomes longer. Thus, in the future, we plan to develop
favorable. DynaSens saves more energy compared to the
the algorithm further so that both sleep-time and threshold
two other sleeping algorithms. Similarly, Figure 11 depicts
adapt to waste throwing patterns and how full the waste
the waste bin level pattern of 175 days using the DynaSens
bin is.
algorithm. The proposed algorithm with a 60 % threshold
We plan to extend and apply DynaSens using a real
is able to achieve no overflowing over the period of 175
dataset from a real-world scenario where multiple bins are
days.
located in different parts of the city. Besides, in this case,
VI. C ONCLUSION AND F UTURE W ORK researchers can work on routing algorithms for garbage
IoT facilitates numerous tasks, such as automation, trucks to find the optimized route for garbage collection.
data collection, information sharing, and reporting. It In the future, we also also plan to extend the DynaSens
is necessary for battery-operated-based IoT devices to algorithm to other IoT use cases.
schedule and sense the environment in an efficient way
to sustain the operation for a longer period. In this paper, ACKNOWLEDGMENT
we proposed DynaSens algorithm for IoT devices. The This work was supported by the European Regional
proposed DynaSens algorithm monitors waste bin levels Fund, Kalmar Region, Kronoberg Region and Linnaeus
and schedules the device sensing intervals accordingly. University.
DynaSens detects the garbage level and schedules the next
sensing interval based on the change in garbage level. In R EFERENCES
this way, sensors remain in sleep mode for a longer interval
[1] S. Kumar, P. Tiwari, and M. Zymbler, “Internet of things is
and avoid unnecessary sensing. The experiment results a revolutionary approach for future technology enhancement: a
show the proposed algorithm is effective in improving review,” Journal of Big data, vol. 6, no. 1, pp. 1–21, 2019.

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SURATHKAL. Downloaded on January 19,2024 at 14:11:45 UTC from IEEE Xplore. Restrictions apply.
 [2] C. A. Medina, M. R. Pérez, and L. C. Trujillo, “Iot paradigm into [12] N. L. Husni, E. Prihatini, A. Silvia et al., “Garbage monitoring and
the smart city vision: A survey,” in 2017 IEEE International Con- warning system,” in 2019 International Conference on Electrical
ference on Internet of Things (iThings) and IEEE Green Computing Engineering and Computer Science (ICECOS). IEEE, 2019, pp.
and Communications (GreenCom) and IEEE Cyber, Physical and 171–175.
Social Computing (CPSCom) and IEEE Smart Data (SmartData), [13] S. J. Ramson, S. Vishnu, A. Kirubaraj, T. Anagnostopoulos, and
2017, pp. 695–704. A. M. Abu-Mahfouz, “A lorawan iot enabled trash bin level
[3] A. Kirimtat, O. Krejcar, A. Kertesz, and M. F. Tasgetiren, “Future monitoring system,” IEEE Transactions on Industrial Informatics,
trends and current state of smart city concepts: A survey,” IEEE 2021.
Access, vol. 8, pp. 86 448–86 467, 2020. [14] X.-R. Huang, S.-S. Yang, W.-S. Chen, Y.-Q. Zhang, C.-T. Lee, and
[4] A. Musaddiq, Y. B. Zikria, O. Hahm, H. Yu, A. K. Bashir, and L.-B. Chen, “An iot-based smart trash cans monitoring system,”
S. W. Kim, “A survey on resource management in iot operating in 2021 IEEE 10th Global Conference on Consumer Electronics
systems,” IEEE Access, vol. 6, pp. 8459–8482, 2018. (GCCE). IEEE, 2021, pp. 623–624.
[5] P.-M. Mutescu, A. I. Petrariu, and A. Lavric, “Wireless communica- [15] M. S. Imam, M. H. Azad, R. Ahmed, S. Hossain, M. I. H. Patwary,
tions for iot: Energy efficiency survey,” in 2021 12th International and M. A. R. Reyad, “Iot based smart waste monitoring and
Symposium on Advanced Topics in Electrical Engineering (ATEE), management system,” in 2021 5th International Conference on
2021, pp. 1–4. Electrical Engineering and Information & Communication Tech-
[6] N. Maleki, M. Loni, M. Daneshtalab, M. Conti, and H. Fotouhi, nology (ICEEICT). IEEE, 2021, pp. 1–4.
“Sofa: A spark-oriented fog architecture,” in IECON 2019-45th [16] T. Anagnostopoulos, A. Zaslavsky, K. Kolomvatsos, A. Medvedev,
annual conference of the ieee industrial electronics society, vol. 1. P. Amirian, J. Morley, and S. Hadjieftymiades, “Challenges and
IEEE, 2019, pp. 2792–2799. opportunities of waste management in iot-enabled smart cities:
[7] L. Pelonero, A. Fornaia, and E. Tramontana, “From smart city to A survey,” IEEE Transactions on Sustainable Computing, vol. 2,
smart citizen: Rewarding waste recycle by designing a data-centric no. 3, pp. 275–289, 2017.
iot based garbage collection service,” in 2020 IEEE International [17] TF-Luna Instruction Manual. (Accessed on 13/06/2022). [Online].
Conference on Smart Computing (SMARTCOMP), 2020, pp. 380– Available: https://www.robotshop.com/media/files/content/b/ben/
385. pdf/tf-luna-8m-lidar-distance-sensor-instructions-manual.pdf
[8] Trends in Solid Waste Management. (Accessed on 13/06/2022). [18] Lopy4 datasheet. (Accessed on 13/06/2022. [Online].
[Online]. Available: https://datatopics.worldbank.org/what-a-waste Available: https://development.pycom.io/gitbook/assets/specsheets/
[9] U. H. Westergren, K. Jonsson, and O. Velsberg, “Internet of things Pycom 002 Specsheets LoPy4 v2.pdf
in the public sector: Perspectives from northern europe,” 2019. [19] LoRa Alliance. (Accessed on 13/06/2022). [Online]. Available:
[10] T. Anagnostopoulos, A. Zaslavsy, A. Medvedev, and S. Khoruzhni- https://lora-alliance.org/
cov, “Top–k query based dynamic scheduling for iot-enabled smart [20] Project: Collaboration towards a gamified waste sort-
city waste collection,” in 2015 16th IEEE International Conference ing house. (accessed on 13/06/2022). [Online]. Avail-
on Mobile Data Management, vol. 2. IEEE, 2015, pp. 50–55. able: https://lnu.se/en/research/searchresearch/research-projects/
[11] T. Vamsi, G. K. Chakravarthi, P. Lanka, and B. Divakar, “An iot project-collaboration-towards-a-gamified-waste-sorting-house/
based smart garbage monitoring and disposal support system,” in [21] Dynasens algorithm. [Online]. Available: https://github.com/
2021 5th International Conference on Computing Methodologies iot-lnu/DynaSens
and Communication (ICCMC). IEEE, 2021, pp. 438–442.

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SURATHKAL. Downloaded on January 19,2024 at 14:11:45 UTC from IEEE Xplore. Restrictions apply.