2023 17th International Conference on Open Source Systems and Technologies (ICOSST)

Low-Cost Multi-Layer Approach for Motion

Detection System for Area Monitoring
2023 17th International Conference on Open Source Systems and Technologies (ICOSST) | 979-8-3503-8132-0/23/$31.00 ©2023 IEEE | DOI: 10.1109/ICOSST60641.2023.10414204

M Farhan Shahid Adnan Shah

College of Aeronautical Engineering Faculty of Computer Sciences and Engineering
National University of Sciences and Technology GIK Institute, Topi-Swabi, Pakistan
Islamabad, Pakistan [email protected]
[email protected]

M Waqas Bashir M Saad Ikram

College of Aeronautical Engineering College of Aeronautical Engineering
National University of Sciences and Technology National University of Sciences and Technology
Islamabad, Pakistan Islamabad, Pakistan
[email protected] [email protected]

Abstract—The RCWL-0516, a remarkably cost- information, such as range, direction, or velocity [2] with
effective yet enigmatic Continuous Wave (CW) radar only a small amount of OEM documentation. In contrast,
module, has posed a compelling challenge for analysis most low-power radar modules have modulation and pro-
due to limited information from its original equipment
manufacturer (OEM). Operating solely on the Doppler cessing capability to measure other essentials of moving
principle, this module detects movement in its entire targets. However, we want to solve the riddles surrounding
surroundings without providing any other contextual the RCWL radar module through this demonstration. Our
data. We endeavored to transform it into directional goal is to convert RCWL into a directional radar that
radar by applying the FFT (Fast Fourier Transform) integrates with other sensors and is capable of mechanical
algorithm and directivity, approximate direction, and
velocity, adding an entirely new dimension to its func- azimuthal scanning for the determination of the direction
tionality. In conjunction, our system integrates Passive of detected motion. Additionally, we work to maximize
Infrared (PIR) sensor and camera, forging a multi- the module’s built-in doppler capabilities by using the
layered approach to motion detection that enhances ac- sophisticated Fast Fourier Transform (FFT) algorithm to
curacy and reliability in threat identification. It further analyze the doppler spectrum. Apart from Radar module,
augments the anti-jamming capability to detect target
with radar, IR and image-based detection. To bolster Passive Infrared (PIR) sensors are also fundamental in
the system’s real-time responsiveness, GSM (Global detecting changes in heat and motion. Cameras serve as
System for Mobile Communications) is also integrated visual sensors, capturing images and videos of the moni-
in addition to audio / visual warning. This approach has tored area. All three sensors work independently. However,
potential to be deployed in various domains to enhance an alert is generated by processing platform, indicating
security, automation, and safety.
Index Terms—cw radar, sensor, motion detection, that the sensor has detected motion. Such incorporation of
conical antenna, infrared, camera detection, Arduino. sensors forms the foundation of a multi-layered approach
FFT to area monitoring. This not only improves accuracy and
reliability but also provides anti-jamming capability to the
I. Introduction system. To enhance the system’s real-time responsiveness
Variations in illumination, clothing deformations, occlu- and alerting features, we’ve seamlessly incorporated simple
sions on the appearance of human body parts, and the audio / visual warning and Global System for Mobile Com-
dimension of the individual in the scene often make it more munications (GSM) technology. This integration enables
difficult to perceive visual motions. Radar has long been the prompt delivery of SMS alerts to specified recipients.
employed for surveillance due to its extensive range, great Additionally, a built-in buzzer functions as part of the
measuring accuracy, good day and night performance, audio-visual warning, promptly alerts occupants to any
and capacity to penetrate walls. It also works well for potential security breach within the property when motion
human motion analysis, tracking, and detection up to 100 is detected by two sensors.
m away [1]. The Continuous Wave (CW) radar module
RCWL has become one of the most popular sensors used A. Limitations
in motion detection. Although incredibly affordable, this The RCWL-0516 sensor presents an intriguing yet chal-
module exploits the doppler effect to detect motion in lenging proposition for developers and engineers to fully
its environment without providing any extra contextual understand its potential. One notable hurdle is the absence

979-8-3503-8132-0/23/$31.00 ©2023 IEEE

Authorized licensed use limited to: NUST School of Electrical Engineering and Computer Science (SEECS). Downloaded on February 14,2024 at 13:16:49 UTC from IEEE Xplore. Restrictions apply.
 2023 17th International Conference on Open Source Systems and Technologies (ICOSST)

of OEM-specified documentation, leaving users to explore then transforms the detection into an auditory signal. To
its capabilities through experimental means. Furthermore, make user-friendly smart devices, they have incorporated
the availability of a comprehensive datasheet is limited, a variety of sensors. The goal of this project is to make
making it imperative for users to rely on practical exper- it easier for people who are blind to move around on
imentation. Unlike other sensors that benefit from exten- their own. The stick’s built-in vibration and buzzer sensors
sive libraries and support in popular programming envi- alert the user to obstacles. The work was presented in
ronments like Matlab, Simulink, or Python, the RCWL- IEEE International Conference on Futuristic Technologies
0516 operates in a more isolated environment, necessitat- in 2022 [8]. Apart from above, multiple sensors have been
ing a hands-on, trial-and-error approach. Despite these used in past for enhanced detection in various application
challenges, its unique characteristics and cost-effectiveness scenarios, as discussed in recent reviews [9] [10] [11].
make it a compelling choice for those willing to explore its Nevertheless, there is limited literature that explores the
potential through practical applications. combined use of radar, PIR sensors, and cameras. Addi-
tionally, the RCWL-0516 module has not been utilized for
II. Related Work tasks such as direction finding and speed sensing using
RCWL-0516 due to its low cost and high market pen- doppler frequency analysis.
etration, has been used in most of the research and ex-
III. Detection Mechanism
periments where motion detection is directly or indirectly
involved. Sagar Shinde et al in their work for advanced Motion detecting modules commonly known as motion
adaptive cruise control of vehicles, has utilized this radar detectors are used to spot or detect presence of moving
to detect the presence of vehicles in the blind spot while objects, particularly people. In the majority of cases, such
range and distance were calculated with TF-Luna LiDAR devices were frequently included as part of a whole system
sensor, thus enhancing driving safety [3]. In several ex- that executed certain tasks; they served as a crucial part
periments, this sensor has provided early detection. As of many systems, including automatic lighting control,
part of the Autonomous Air Defence System (AADS) energy efficiency, home control, and small to medium,
development, the RCWL-0516 microwave radar was em- and large-scale (advanced) security. A motion detector
ployed for presence detection in near ranges in order to might be one of the sensors in a burglar alarm that
augment detection by HB100 X-band pulse doppler radar alerts the home’s owner or security company when it spots
module [4]. Similarly, an IOT-based thermal surveillance any movement that might be an intruder. A pre-installed
system was created to identify unusual motions at night security camera with this type of detection might also be
and alert the user. Thermal sensors were used to detect programmed to record a live feed of any potential intrusion
hot bodies by infrared (IR) sensing and other movements [12]. Several types of motion detection are in wide use.
were detected using microwave motion sensor. Further- These include:
more, motion sensors helped prevent erroneous warnings A. Passive infrared (PIR)
brought on by environmental conditions [5]. An article
Unlike all other things in the background at ambient
published from Widyatama University, Indonesia, where
temperature, PIR sensors only identify people by the
author discussed controlling various equipment in a certain
radiation they emit, which is generated from their skin
building. Such control was automated through movement
temperature. The sensor is referred to as a passive infrared
of human by integrating Arduino board and RCWL sensor
sensor since it does not emit any energy [13].
thus providing a complete architecture of smart building
[6]. Which not only helped in saving energy but also B. Microwave radar
provided improved security and comfort. Building machine These devices operate in microwave frequencies and
learning models for tracking, detecting, and recognizing can transmit either continuous wave, modulated / un-
movement and their automated processing of vast volumes modulated signals. It resembles a radar speed gun in some
of sensory data presents several challenges. Dejan and ways since it uses the Doppler radar technique to detect
Gjorgji developed a data collection system to observe motion. Continuous microwave radiation is released, and
human movement through less expensive modules based when an item or a person moves in front of the transmitter,
on passive infrared and microwave principles of operation. the reflected microwaves undergo a shift in phase and
In their work, they utilized the analog output of both frequency, producing a low frequency signal [14].
the microwave radar sensor RCWL-0516 and the passive
infrared sensor HC-SR501. The pin 12 of BISS0001 IC was C. Ultrasonic
utilized to obtain the raw analog signal of the detected By using its transducer to create ultrasonic waves at
motion which was logged in Arduino microcontroller [7]. a frequency higher than the human ear can hear, an
An ESP32 microcontroller, an ultrasonic sensor, GPS ultrasonic sensor collects reflections of the waves from
module and a radar sensor (RCWL-0516), were all used nearby objects. The tiniest blip in the received field, sim-
by Bhosle et al. to create an assistance system for those ilar to the doppler radar, shows mobility. Low modulated
who are blind. The object is detected by smart stick, which frequencies (for walking speeds) also creates doppler shift

Authorized licensed use limited to: NUST School of Electrical Engineering and Computer Science (SEECS). Downloaded on February 14,2024 at 13:16:49 UTC from IEEE Xplore. Restrictions apply.
 2023 17th International Conference on Open Source Systems and Technologies (ICOSST)

which is used to calculate distance and speed. Ultrasonic regions are separated through SIW (Substrate Integrated
sensors have the potential drawback of being sensitive Waveguide) lines. A 3.175 GHz oscillation is produced by
even in locations where their detection is not required, an NPN transistor, capacitors, and an inductor with a
for example, as a result of sound waves reflecting off of wavelength of 9.45 cm and half wavelength of 4.72 cm.
corners of walls. For lighting control, when the objective is The microwave antenna is integrated on the PCB itself
to detect any movement in the area, such broad detection with the front end and processing region, the RCWL-0516
coverage may be useful. However, a sensor that selectively operates as a self-contained radar unit. Figure 1 shows the
faces the way towards the entrance is more likely to be basic workings of this sensor based on components. [17]
suitable for automatic door opening [15].
D. Video camera detection
With the development of inexpensive surveillance cam-
eras with video recording capabilities, it is now possible to
use the output of one of these cameras to employ software
to identify motion in its field of view. This technique
is particularly appealing when the device is configured
to start recording video when motion is detected. The
camera and computer are actually the only pieces of
hardware required. This may nevertheless be regarded
as a passive technology since the coverage path may
typically be lighted. To detect motion in the dark, hybrid
video cameras with infrared capability are installed as an
alternative [16].
Fig. 1. RCWL-0516 Schematic diagram [17]
IV. Proposed detection mechanism
Proposed detection mechanism exploits doppler feature
and directional scanning for enhanced range and direction. C. RCWL-0516 Radiation Pattern
It incorporates additional IR sensing and optical capability The RCWL-0516 module underwent radiation testing
making it suitable for variety of applications in diverse at antenna test lab facility, yielding valuable insights into
fields of real world. Detail is given in subsequent para- its performance characteristics. During the assessment, it
graphs. was observed that the radiation pattern of this module
A. Radar Sensor exhibits a nearly spherical shape [18]. However, the pres-
ence of nulls at specific angles represents directions where
In order to detect a doppler change from an object
the emitted power is significantly reduced up to 25 dB.
that is moving, microwave radar sensors calculate fre-
Coupled with its monopole antenna design, this module
quency shift by comparing the frequency of the echoes
achieves its maximum directivity at an angle of θ=90°.
with the radiation frequency. Such devices typically have
Such pattern infact gives nearly isotropic coverage in 360°.
inexpensive, die-cast metal horn antennas. In our work
We simulated a quarterwave monopole antenna design in
we have employed low-cost RCWL-0516 microwave motion
CST Studio 2019 in Figure 2, which shows a uniform
detector module. It includes MMBR941M high frequency
distribution of radiation intensity in azimuth at θ=90°
NPN transistor. This is the heart of its microwave region
with two nulls perpendicular to broadside radiation.
and functions as a microwave frequency emitter, receiver,
and mixer. For monocratic functioning, a circular micro-
strip patch is employed as the monopole antenna. RCWL-
9196 IC is also included on FR4 PCB for signal processing
and amplification [7]. The sensor incorporates a doppler
shift in received frequency. If the doppler is higher than
a certain threshold of the sensor (threshold > 0.7 v), it
triggers the Vout interface with logic high.
B. Sensor Schematic Circuit
This schematic is divided into two sections. A consider-
ably lower frequency section is based on an IC (RCWL- Fig. 2. Pattern before modifi- Fig. 3. Dimensions of proto-
9196) that has similar functions to the BISS0001 IC used cation type
in most PIR motion detectors for signal processing. The
microwave region of the sensor is based on microwave
frequency transmitter, receiver, and mixer sections. Both

Authorized licensed use limited to: NUST School of Electrical Engineering and Computer Science (SEECS). Downloaded on February 14,2024 at 13:16:49 UTC from IEEE Xplore. Restrictions apply.
 2023 17th International Conference on Open Source Systems and Technologies (ICOSST)

its own IDE (Integrated Development Environment) Ver

2.

V. Experimental Setup
RCWL sensor was mainly intended for connectivity with
computer through Arduino as no sufficient information of
this sensor was available, and thus maximum information
can be extracted for its operations. We performed multiple
experiments to observe the behaviors of sensor. We used
CST Studio Suite for analysis of antenna-related features,
Matlab 2023 for signal processing, and Arduino IDE V2
Fig. 4. Pattern after modification for integrating sensors and monitoring detection.

A. Basic setup
D. Pyroelectric infrared sensor (HC-SR501)
In this basic setup, we integrated a doppler radar sensor
Passive sensors that can detect shifts in temperature are and an LED into USB powered Arduino UNO configured
pyroelectric infrared (PIR) sensors. They are frequently on a breadboard. The configuration was made to enable
employed as a presence trigger since they are well suited the LED to react when motion is detected as depicted
to detect human motion using body temperature. The HC- in Figure 5. [23]. Indoor detection was observed at short
SR501 PIR detector is a low-cost and convenient gadget. distance of 1-5 cm. This may be due to interference or
The LHI778 Inactive Infrared Detector and the BISS0001 orientation of the sensor.
IC are used in this motion detection module to regulate
how motion sensing. The module has a sensitivity setting
that allows it to detect movement in a 110° cone-shaped
region at a range of three to seven meters. The module
also features trigger choice and time delay settings that
allow for fine tailoring in accordance with its application
[19].

E. Motion Camera
In order to demonstrate proof of concept (POC), we
used a low cost internet protocol (IP) enabled, off-the- Fig. 5. Initial Checks Fig. 6. Conical Horn (Narrow
Beam)
shelf solution. The camera is portable, compact, and 150°
wide to cover a wider azimuth. It is powered by an Arduino
board and runs on 5V DC. It takes a photo of the detected
region and sends an alarm notification to the owner’s B. Directivity Enhancement Approach
PC or mobile device over WiFi [20]. Since the detection The present detector’s radiation pattern is nearly omni-
is pixel-based, thus requires better lightening condition. directional, featuring various lobes with lower intensity.
By using optical technology, this adds another layer of To enhance its performance, increase its range, and mini-
detection. IR lights are utilized to cover a 5 meter range mize false alarms which are generated due to interference
for night vision. For integration with Arduino platform and multi-path reflections, it is necessary to focus the
and for additional configuration, we can also use ESP32- microwave radiations [23] [24]. This entails improving the
CAM WiFi Board with camera module OV2640 [21]. detector’s directivity and effective radiated power (ERP)
while maintaining maximum directivity at θ = 90°. So we
F. Data Processing and analysis Platform carried out various experiments by using different reflec-
Arduino UNO was used as data processing platform. tors, including an aluminum conical horn, an aluminum
This is equipped with ATmega328P-based microcontroller. parabolic reflector, and a rectangular copper reflector, as
It is capable of 6 analogue inputs, 14 digital input/output shown in Figures 6-8.
pins (six of which can be used as PWM), 32 KB of flash Following observations were made:-
storage, a USB port, a power jack, an in-circuit serial • Conical horn with aperture diameter less than λ/4
programming (ICSP) header, a 16 MHz ceramic resonator, gives poor performance.
and a reset button [22]. It has all the components required • Placing reflector close to sensor (less than 1 cm) gives
to support the microcontroller and functions properly with no oscillations.
a 5-volt DC supply from a computer or other external • Rectangular copper reflector with distance 1 cm apart
source. Configuration and coding is carried out through gives good transmission in 0-180°.

Authorized licensed use limited to: NUST School of Electrical Engineering and Computer Science (SEECS). Downloaded on February 14,2024 at 13:16:49 UTC from IEEE Xplore. Restrictions apply.
 2023 17th International Conference on Open Source Systems and Technologies (ICOSST)

Where DB is directivity-beamwidth product, which is

33709 deg 2 for uniform circular aperture.
D. Mechanical Scanning approach
Typically, physical scanning involves circular move-
ments, achieved by rotating the entire antenna to follow
a predetermined pattern. In the case of circular scanning
antennas, the antenna undergoes continuous rotation, ei-
ther clockwise or counterclockwise, covering a full 360◦
Fig. 7. Aluminum Parabolic Fig. 8. Copper with FR4 sheet
around a vertical axis. The scan time ω is the interval
between successive signal sweeps that holds true for a cir-
C. Reference Antenna reflector cular scan. We employed a circular scan at proportionate
rotational speed (I) using a TowerPro servo motor and a
Test results of conical horn by antenna test lab are found
5V DC supply (V) from an Arduino. The no-load speed,
superior with gain of 12 dBi and directional radiations.
or maximum speed the motor can achieve in the absence
Conical horn is also compatible and gives better results
of torque, is the speed that is typically mentioned in shaft
if optimized geometry is used [25]. For the purpose of
speed specifications. Shaft speed is frequently expressed
test and trial, we employed feasible dimensions, consid-
in revolutions per minute (rpm). The number of radians
ering the wavelength and keeping in view the observations
that correspond to these rotations is expressed in radians
made earlier. We prepared wave-guide like structure as
per second (rad/s). The rotational speed of a perfect DC
truncated conical horn and used in the proposed system.
motor is proportional to the input voltage [30] [31].
A commercial off the shelf discarded product was used
with Aluminum foil radiator and source (sensor) facing
I =J ×V (6)
z-axis was placed inside at distance and height of λ/4
[26]. Figure 3 shows prototype antenna with dimensions. Where j is the constant of proportionality with units
We further carried out simulation on CST Studio for rad/(s-v). For numerical calculations, this unit of rota-
calculations of antenna directivity, beamwidth and radi- tional speed is more convenient
ation patter. Figure 4 shows simulation results of our
prototype. The simulated results indicate a lower level of ωrad/sec = ωrpm × 2π/60 (7)
directivity compared to the analytical solution, primarily We set rotation speed to RPM = 5 which gives ARP =
because material losses were taken into account. Following 12 s and a scan rate of ωdeg/sec = 30. We set scan area
approximations are useful for horn antenna with uniform from 0 - 180° which gives three non-overlapping contiguous
field distribution [27] [28]. beam positions, each with HPBW of 60° as calculated
earlier. This way, the sensor will be able to carry out one
4π C detection per beam position. If there are number of moving
Dc (dB) = 10log10 [ϵap (πa2 )] = 10log10 ( )2 −L(s) (1)
λ2 λ objects in 60°, only the initial moving object above the
Where λ is wavelength (9.4 cm), C is the circumference threshold will be detected. An increasing scan rate will
of aperture (34.5 cm) and L(s) is loss figure (1.59 dB) have no effect on detection as the re-trigger time of the
which is calculated using following equation; sensor is 2 seconds.
E. Direction of target
2 3 Through employing scanning in azimuth, the direction
L(s) = 0.5030 + 5.1123s − 7.1138s + 23.1401s , L(s) ≤ 3λ
(2) of detection can be calculated. When antenna is rotating
at 5 RPM and motion is detected at certain angle, Arduino
gives a text warning on the serial monitor and indicates
L(s) = 0.7853 − 0.3976s + 13.112s2 + 3.901s3 , L(s) > 3λ the current position of the servo in degrees, which is the
(3) boresight of the sensor. Detection starts from X°, and
Coefficient S (phase deviation) is calculated as; with 1° stepping, it keeps reporting till the detection is
d2m over due to change of antenna direction to Y°. So the
S= (4) approximate direction is calculated as Z°= X° + (Y°-
8λl
X°)/2 as illustrated in Figure 9.
Using dimensions of the prototype reflector as depicted
in Figure 3, we get directivity of 9.8 dBi. Half Power VI. Discussion & Results
Beamwidth (HPBW) is 60◦ calculated using following The PIR sensor, when linked to one of the Arduino’s
approximation [29]; input pins, is capable of identifying changes in the infrared
DB radiation it perceives within its field of view (FOV). When-
D≈ (5) ever motion is detected, the PIR sensor transmits a high
HPE HPH

Authorized licensed use limited to: NUST School of Electrical Engineering and Computer Science (SEECS). Downloaded on February 14,2024 at 13:16:49 UTC from IEEE Xplore. Restrictions apply.
 2023 17th International Conference on Open Source Systems and Technologies (ICOSST)

signal to the Arduino via pin D6, signaling the presence of

an object or heat source. In response, the microcontroller
activates a red LED that is connected to pin D7, serving
as a visual warning. Simultaneously, it outputs a message
to the Arduino IDE’s serial monitor, stating, "Motion
detected by the PIR sensor" When the RCWL-0516 sensor
senses motion, it illuminates a blue LED and sends a
warning message to the Arduino IDE, stating, "Motion
detected by the radar sensor" Additionally, it provides
directional information, offering an approximate direction Fig. 9. Multi-layer mechanical Fig. 10. Motion detection by
scanning and coverage Prototype
within a 60° azimuth. The camera operates independently
of these sensors. It is connected to an IP network, which
transmits photos of the detected area to a mobile device. Hz. The discrete fourier transform (DFT) of collected
The camera’s detection range is observed up to a range of baseband data was obtained using the FFT algorithm with
8 meters, and notifications are monitored over WiFi. This a sampling rate of 1 KHz and 600 samples. The resulting
can be integrated with processing platform to ascertain frequency domain spectrum is shown in Figure 11. Velocity
whether the detected object is human or not through is then calculated at peak value of doppler frequency as
resource-efficient algorithms. [32]. Furthermore, the sys- 1.8 m/s, which is normal human walking speed, using the
tem is integrated with GSM technology, generating SMS following relation:
alerts that convey information about detection events,
V = (f d × λ)/2 (8)
sensor details, and direction. When detection is made
by both sensors, it gives text warning on serial monitor
indicating that motion is detected by PIR and radar. With
this, a visual and audio warning is successfully initiated
through the buzzer attached to the board, and call is
generated to given mobile number by GSM module. PIR
signal was also monitored on an Arduino IDE serial plotter
to analyze behavior during detection. A smooth transition
was observed, which triggers logic high for visual warning,
which indicates that the sensor may be using adaptive
thresholding for noise rejection. With multiple detections
carried out in outdoor environment, the FOV of PIR was
observed to be wider around 100° with an effective range
of 3 meters.
The analog signal of radar sensor was collected from
pin 12 of IC which is the output of 2nd stage of Op
amplifier [33] which is a baseband signal. It was analyzed
in the Arduino IDE serial plotter and Matlab (Arduino
Fig. 11. Spectrum of base-band data
app). When a movement is detected, it creates a doppler
shift that increases voltage above 0.7 VDC. This results
in triggering D3 with high logic for 2 seconds. Noise is VII. Conclusion
observed in radar returns as compared to PIR sensors. In The unit exhibits a detection range of a few centimeters
an open-air environment, good detection was observed at a when placed close to the ground, but this range increases
longer range ( 5 m–15 m) due to higher gain w.r.t isotropic to approximately 3 meters when positioned at a higher
radiation. elevation. Notably, when directional adjustments were
Matlab was utilized to collect real-time data through made, a range exceeding 12 meters was achieved, and the
a serial port. Fast Fourier Transform (FFT) was applied detection capability did not significantly diminish with
on collected data via serial port, and the resulting trans- increasing distance. The baseband spectrum of the return
formed data is plotted. The FFT is a computational tech- echo holds good for velocity measurement, provided accu-
nique used to convert a signal from the time domain, such rate data on signal bandwidth and sample parameters are
as a sample buffer, into the frequency domain, called as the available. Additionally, phase shifting measurements can
spectrum which does not represent individual frequencies offer precise range detection within 4.6 centimeters (λ/2)
but instead indicates the distribution of energy across with resolution of sub-millimeters, making it suitable for
different frequencies in a given range [34]. The maximum monitoring vital signs and achieving precision in indus-
detectable velocity by microwave module is 2 m/s, as trial measurements. This multi-layered system, harness-
per the datasheet. which gives maximum doppler of 86 ing radar, infrared, and optical technologies, significantly

Authorized licensed use limited to: NUST School of Electrical Engineering and Computer Science (SEECS). Downloaded on February 14,2024 at 13:16:49 UTC from IEEE Xplore. Restrictions apply.
 2023 17th International Conference on Open Source Systems and Technologies (ICOSST)

enhances accuracy while exhibiting strong resistance to [14] Surg-Won Sohn. Design and implementation of a microwave
conventional jamming techniques. Although the system motion detector with low power consumption. , 20(7):57–64,
2015.
encounters challenges related to reliability and compat- [15] Kirtan Gopal Panda, Deepak Agrawal, Arcade Nshimiyimana,
ibility, meticulous optimizations can render it a robust, and Ashraf Hossain. Effects of environment on accuracy of
low-cost solution for comprehensive area monitoring. Im- ultrasonic sensor operates in millimetre range. Perspectives in
Science, 8:574–576, 2016.
proved integration can be attained by incorporating data [16] Asif Ansari, TC Manjunath, and C Ardil. Implementation of
fusion and employing deep learning methods for better a motion detection system. International Journal of Computer
estimation of range, bearing, and velocity. Exploring an and Systems Engineering, 2(9):1979–1990, 2008.
[17] Handson Technology. RCWL-9196, RCWL-0516 Doppler Radar
object’s radar cross-section (RCS) measurement analysis Sensor Datasheet, 2023. Rev. D.
through signal processing and utilizing spectrometry for [18] LabTest Technology. Antenna Tests Results BY LabTest, 2021.
classification represents another promising avenue for fu- Rev. D.
[19] PIR. HC-SR501 PIR Sensor Datasheet, 2020. Rev. D.
ture research. [20] Wifi. Low-cost wifi motion camera specifications, 2023. Rev. D.
[21] AI-Thinker. ESP32-Cam WiFi SOC Module V.1, 2023. Rev. 1.
[22] Arduino. Arduino® UNO R3 1 / 13 Product Reference Manual,
References 2023. Rev. 1.
[23] Laxmi Samyuktha Manchineella. Motion detection using mi-
[1] Victor C Chen. Detection and analysis of human motion by crowave radar sensor. Available at SSRN 3918233, 2021.
radar. In 2008 IEEE Radar Conference, pages 1–4. IEEE, 2008. [24] Alexis Herrera-Angulo and Efrain Zenteno-Bolanos. A low-cost
[2] Alexis Herrera-Angulo and Efrain Zenteno-Bolanos. A low-cost microwave system for pedestrian speed estimation. In 2018
microwave system for pedestrian speed estimation. In 2018 IEEE MTT-S Latin America Microwave Conference (LAMC
IEEE MTT-S Latin America Microwave Conference (LAMC 2018), pages 1–3. IEEE, 2018.
2018), pages 1–3. IEEE, 2018. [25] LabTest Technology. Reference Antenna Tests BY LabTest,
[3] Sagar Shinde, Lalitkumar Wadhwa, Sheetalkumar Rawandale, 2021. Rev. D.
Nitin Sherje, Shubham Jadhav, Ausmita Biswas, and Kalyani [26] W.W.S. Lee and E.K.N. Yung. The input impedance of a coaxial
Gupta. Advanced adaptive cruise control and vehicle to vehicle line fed probe in a cylindrical waveguide. IEEE Transactions on
communication using lidar. Multidisciplinary Science Journal, Microwave Theory and Techniques, 42(8):1468–1473, 1994.
6(3), 2024. [27] Giuseppina Monti, Laura Corchia, and Luciano Tarricone. Fab-
[4] Fazle Rabby Khan, Md Muhabullah, Roksana Islam, Moham- rication techniques for wearable antennas. In 2013 European
mad Monirujjaman Khan, Mehedi Masud, Sultan Aljahdali, Radar Conference, pages 435–438. IEEE, 2013.
Avinash Kaur, and Parminder Singh. A cost-efficient au- [28] Constantine A Balanis. Antenna theory: analysis and design.
tonomous air defense system for national security. Security and John wiley & sons, 2016.
Communication Networks, 2021:1–10, 2021. [29] Warren L Stutzman and Gary A Thiele. Antenna theory and
[5] AM Bharath et al. IoT based thermal surveillance and security design. John Wiley & Sons, 2012.
system. 2019. [30] Garg Ashish. Security system using esp32. Master’s thesis,
[6] Rian Eka Putra, Kurniawan Rosadi, Asep Kurniawan, Wafi České vysoké učení technické v Praze. Vypočetní a informační
Firgiwansyah, and Ase Suryana. Smart building with speech centrum., 2020.
recognition through android and rcwl-based bluetooth hc05 [31] Hyunmin Jeong and Sangkil Kim. Educational low-cost C-band
(case study: Smks pelita bunga bangsa arjasari). PalArch’s FMCW radar system comprising commercial off-the-shelf com-
Journal of Archaeology of Egypt/Egyptology, 17(4):3316–3335, ponents for indoor through-wall object detection. Electronics,
2020. 10(22):2758, 2021.
[7] Dejan Gjorgjevikj and GJorgji Madzarov. Data collection [32] Khurram Kazmi Farhan, Nayyer. Influence of various ML-
module for human activity recognition. 2018. based binary classifiers on the performance on handwritten
[8] Priyanka Bhosle, Prashant Pal, Vallari Khobragade, digit recognition. In 10th International Conference on Future
Shashank Kumar Singh, and Prathmesh Kenekar. Smart Internet of Things and Cloud (FICloud). IEEE, 2023 (to be
navigation system assistance for visually impaired people. published).
In 2022 International Conference on Futuristic Technologies [33] Youtai Semiconductor Co., Ltd. BISS0001 Micro Power PIR
(INCOFT), pages 1–5. IEEE, 2022. Motion Detector IC Datasheet, 2020. Rev. D.
[9] M Kastek, R Dulski, M Życzkowski, M Szustakowski, W Ciu- [34] Tuppak Bobby Vorlen Sagala and Joko Suryana. Implemen-
rapiński, K Firmanty, N Pałka, and G Bieszczad. Multisensor tation of mechanical scanning and signal processing for fmcw
systems for security of critical infrastructures: concept, data radar. In 2016 International Symposium on Electronics and
fusion, and experimental results. In International Symposium Smart Devices (ISESD), pages 46–50. IEEE, 2016.
on Photoelectronic Detection and Imaging 2011: Advances in
Infrared Imaging and Applications, volume 8193, pages 1069–
1077. SPIE, 2011.
[10] A Vellingiri, K Mohanasundaram, KS Tamilselvan, R Mah-
eswar, and N Ganesh. Multiple sensor based human detection
robots: A review. International Journal on Smart Sensing and
Intelligent Systems, 16(1), 2023.
[11] Luis Patino, Michael Hubner, Rachel King, Martin Litzen-
berger, Laure Roupioz, Kacper Michon, Łukasz Szklarski, Julian
Pegoraro, Nikolai Stoianov, and James Ferryman. Fusion of het-
erogenous sensor data in border surveillance. Sensors, 22(19),
2022.
[12] Ching Yee Yong, Rubita Sudirman, and Kim Mey Chew. Motion
detection and analysis with four different detectors. In 2011
Third International Conference on Computational Intelligence,
Modelling & Simulation, pages 46–50. IEEE, 2011.
[13] Roman Gabl and Florian Stummer. Development of a sensor
to measure physician consultation times. Sensors, 19(24):5359,
2019.

Authorized licensed use limited to: NUST School of Electrical Engineering and Computer Science (SEECS). Downloaded on February 14,2024 at 13:16:49 UTC from IEEE Xplore. Restrictions apply.