ITC-CSCC2022 685

Design and Experiment of FMCW Radar System

for Drone Detection
2022 37th International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC) | 978-1-6654-8559-3/22/$31.00 ©2022 IEEE | DOI: 10.1109/ITC-CSCC55581.2022.9895083

Seksan Eiadkaew Kittisak Phaebua Krit Athikulwongse

Akkarat Boonpoonga Research Center of Innovation Digital National Electronics and Computer
Research Center of Innovation Digital and Electromagnetic Technology Technology Center, National Science
and Electromagnetic Technology (iDEMT), Department of Teacher and Technology Development Agency,
(iDEMT), Electrical Engineering, Training in Electrical Engineering, Pathumthani, 12120, Thailand
Department of Electrical and Computer Faculty of Technical Education, King [Link]@[Link]
Engineering, Faculty of Engineering, Mongkut’s University of Technology
King Mongkut’s University of North Bangkok, Thailand.
Technology North Bangkok, kittisak.p@[Link]
Bangkok, Thailand
[Link]@[Link]
akkarat.b@[Link]

Nattakarn Shutimarrungson
Department of Business Computer,
Mahasarakham Business School,
Mahasarakham University, Maha
Sarakham, Thailand
nattakarn@[Link]

Abstract— Design and experiment of FMCW radar This makes it difficult to detect drones in video surveillance.
signal for drone is proposed in this paper. FMCW radar However, the drone detection through video surveillance or
with operating frequency of 10 GHz is designed. The photographs in [1] and [2] has a limitation. It is unable to
experimental study of the radar signal is done in anechoic detect drones in some situations such as at night, cloud
chamber. A metal plate and commercial drone are situation. The RF radar technology can solve this problem.
employed as the radar targets. The triangle wave control Radar system for detecting small objects has continued to
signal with frequency of 300 kHz is employed. The voltage evolve in terms of circuit size and wider bandwidth [3]. The
controls oscillate (VCO) with the bandwidth of 76 MHz is frequency modulated continuous wave radar (FMCW) is used
employed to generate the radio frequency (RF) radar [3]. It provides a compact size and low transient power than
signal. High frequency RF mixer is used to extracts the pulse radar. The nonlinearity of FM chirp and antenna leakage
intermediate frequency (IF) of the target. The high gain is solved. The system provides a compact size, low cost, low
parabolic antenna is employed to increase the reflected power and high range resolution. Operating frequency of 94.7
radar signal which provides long-range detection. It is GHz with bandwidth of 800 MHz is employed. Range
found that the designed radar system provides a correct resolution less than 0.3 m is obtained. Based on the radar
position of the targets. The targets provide difference testing in the anechoic chamber, the results showed that the
spectrum of radar signal. radar was able to detect a weak signal of target at a distance of
300 m with SNR 20 dB. A digital signal processing FMCW
Keywords— Frequency modulated continuous wave radar, radar design on an FPGA board for use at 24 GHz and a DSP
X-band, Drone detection module for automotive applications is proposed in [4]. The
detection distance from 0.5 meter and up to 3.4 meters with
I. INTRODUCTION maximum error of 3 cm is achieved. The design of FMCW
radar at X-band to detect small objects was proposed in [5].
Unmanned Aerial Vehicle (UAV) aircraft namely drone Using an 2 array transmitting (Tx) and receiving (Rx) horn
is an aircraft that can be controlled to fly from a far distance antennas provides 23 dBi gain. The object detection testing
using radio control from ground station. Drone is now widely was carried out in an anechoic chamber. Metal rod with 3cm
used due to lower price, small size and easy to fly. Sometime, diameter is employed. Horizontal and vertically alignments of
an unauthorized drone is causing problems in some locations a metal rod are studies. Also, a square plate of size 30x30 cm2
such as in airport area and other. Thus, an unauthorized drone is employed. The results are showed that the radar was able to
detection system is very important. Drone detection detect a metal rod in both positions with a beat frequency of
technologies in [1] offer machine learning-based drone 500 kHz, while the plate detection test was at 1.5 m and 4 m.
detection. Machine learning recognizes images and videos of It has a beat frequency of 300 kHz and 500 kHz.
drones in different ways so the system can discern the position
of the manufacturer and model of the drone. The results of the This paper presents design of FMCW radar for drone
detection and classification performance of the drones in [1] detection application. Operating frequency of X-band is
are up to 89% accuracy. Moreover, the video surveillance employed. The radar targets are a metal plate and commercial
system with artificial intelligent (AI) was used to detect a drone. The experimental will be conducted in anechoic
drone [2]. In paper [2], the drones and birds targets provide chamber room. In section 2, describes FMCW radar principle
similar video characteristics between small drones and birds. including modulated signal and equation for range calculation

978-1-6654-8559-3/22/$31.00 @2022 IEEE

Authorized licensed use limited to: Dr. Ambedkar Institute of Technology. Downloaded on September 03,2025 at 02:18:21 UTC from IEEE Xplore. Restrictions apply.
 ITC-CSCC2022 686

and velocity calculation. In section 3, describes hardware III. HARDWARE DESIGN

design. In section 4 and 5, experimental setup and result will
be presented, respectively. Conclusion is presented in section A. Hardware design
6. The proposed FMCW radar employed the RF bandwidth
of 76 MHz with 10 GHz starting RF frequency which provides
II. FMCW RADAR PRINCIPLE the resolution of 65.22 cm. The block diagram of the proposed
Frequency modulated continuous wave (FMCW) radar is FMCW radar is shown in Fig. 2. The frequency of triangle
a radar that the transmitting (Tx) frequency increased and control signal of 300 kHz is employed. The proposed FMCW
decreased as a function of time [1]-[6]. The sawtooth and radar system is showed in Fig. 7. The Tx and Rx parabolic
triangular signals are commonly used to control a high reflector antennas are employed as shown in Fig. 3. The NI-
frequency radar signal. The Tx RF electromagnetic wave is USB oscilloscope is employed and interface to computer. The
generated by the Tx antenna. The transmitting signal fast Fourier transform (FFT) algorithm is used to transform
propagates through the air and hit the object. Tx signal is time domain signal to frequency domain signal. The low-cost
reflected from object and return to receiver as shown in Fig. 1 RF equipments consist low-noise amplifiers, RF power
(a). Figure 1 (a) shows chirp triangular frequency of Tx and divider, RF mixer, and intermediate frequency (IF) amplifier.
Rx versus time. Rx chirp triangular frequency has a time delay
and shifted frequency comparing with Tx chirp triangular
frequency due to the distance of the target and doppler effect.
Tx and Rx signal will be mixed together by RF mixer. The
high and low intermediate frequency (IF) signal will be
created. However, high-frequency IF component is filtered
out. The remaining low-frequency component is employed to
calculate the target distance.

Fig 2. FMCW radar system diagram

B. Tx and Rx Antennas
The parabolic reflector antennas are employed for both
Tx and Rx parts as shown in Fig. 3. The summary of antenna
dimensions is shown in Table 1. The main reflector diameter
(D1) of 65 cm is used. The parameters F/D1 ratio of 0.35, sub
reflector diameter (D2) of 13.6 cm with focus distance (F2) of
18.7 cm and linearly polarization are employed. The proposed
antenna provides gain at 10 GHz (d=5cm) of 33.8 dBi. The
Fig 1. FMCW radar waveform signal (a) Triangular signal between Reflection coefficient of |S11| dB is shown in Fig. 4. At
Tx and Rx signal (b) The corresponding beat frequency operating frequency of 10 GHz, the |S11| dB is less than -10
dB.
The IF frequency signals are obtained by subtraction
between Tx and Rx frequency diagram as shows in Fig.1 (b). F2
The range (R) and velocity (V) can be calculated by using
equation (1) and (2) that presented in [7]. Where and
are the IF frequency for both up-ramp and down-ramp,
respectively. The B and T denote bandwidth of FMCW
modulated signal and chirp signal period respectively. d
D2 D1

(1)

(2)

Fig 3. Parabolic antenna for FMCW radar

Authorized licensed use limited to: Dr. Ambedkar Institute of Technology. Downloaded on September 03,2025 at 02:18:21 UTC from IEEE Xplore. Restrictions apply.
 ITC-CSCC2022 687

Table I. Antenna Dimensions

Parameter Value
Main reflector diameter; D1 65 cm

F/D1 ratio 0.35

Sub reflector diameter; D2 13.6 cm
Focus distance; F2 18.7 cm
Polarization Horizontally linear
Gain at 10 GHz (d=5cm) 33.8 dBi (c) (d)
Fig. 5. Target setup (a) Metal plate is mounted vertically on PVC
stand (b) Metal plate is mounted horizontally on PVC stand (c) The
IV. EXPERIMENTAL SETUP drone is placed horizontally on the table (d) The drone is placed
vertically on the table.
Experimental for proposed FMCW radar is carried out
in anechoic room to prevent external interference. Target
objects are divided into four cases. Case 1, the metal plate
with the width and length of 5x70 cm! . It is mounted
vertically on PVC stand as show in Fig. 5 (a). Case 2 using
the same metal plate as Case 1, but it is installed
horizontally on PVC stand as show in Fig. 5 (b). Case 3 and
Case 4, using agricultural drone as target object, it has
dimensions (WxDxH) of 135x100x60 cm" and it is placed
on a table with a height of 70 cm. RF Absorber is placed in
front of table to reduce the reflection effect from table. In
the Case 3, drone is placed horizontally, while Case 4,
drone is placed vertically using foam padding supporter as
show in Fig 5 (c) and 5 (d), respectively. The FMCW radar
is setup of 5 meter away from the target object as shows in
Fig. 6.

Fig 6. Experimental setup in anechoic chamber.

V. EXPERIMENTAL RESULT
From experimental, mounted metal plates on PVC stand
for both cases of vertical and horizontal alignments are shown
in Fig 5 (a) and 5 (b). The frequency domain result is shown
in Fig. 7. IF frequencies of both cases are the same of 800 kHz
where range (R) of 5.263 meters is calculated by Eq. (1). An
error of 26.3 cm from the actual range is obtained. Amplitude
of IF signals from vertical alignment metal plate case is 754.54
mV and horizontal alignment metal plate case is 843.86 mV.
Moreover, the drone is mounted on table as shown in Fig. 5
(c) and 5 (d). It is detected by FMCW radar system. The result
in frequency domain is shown in Fig. 8. IF frequencies of both
cases are the same of 800 kHz and 5.263 meter range is
calculated by Eq. (1). Amplitude of IF signals from the
Fig. 4. Reflection coefficient of |S11| dB. vertical drone alignment is 857.32 mV and horizontal drone
alignment is 907.47 mV. The amplitude from horizontal
alignment metal plate is slightly higher than vertical metal
plate alignment. Moreover, the amplitude of the radar signal
in the case of horizontal drone alignment is higher than
vertical drone alignment due to the horizontally linear
polarization of the Tx and Rx antennas.

(a) (b)

Authorized licensed use limited to: Dr. Ambedkar Institute of Technology. Downloaded on September 03,2025 at 02:18:21 UTC from IEEE Xplore. Restrictions apply.
 ITC-CSCC2022 688

REFERENCES
[1] D. Lee, W. Gyu La and H. Kim, "Drone Detection and
Identification System using Artificial Intelligence," 2018
International Conference on Information and Communication
Technology Convergence (ICTC), 2018, pp. 1131-1133, doi:
10.1109/ICTC.2018.8539442.

[2] M. Nalamati, A. Kapoor, M. Saqib, N. Sharma and M.

Blumenstein, "Drone Detection in Long-Range Surveillance
Videos," 2019 16th IEEE International Conference on
Advanced Video and Signal Based Surveillance (AVSS), 2019,
pp. 1-6, doi: 10.1109/AVSS.2019.8909830.

[3] A. Y. Nashashibi, J. Lee and K. Sarabandi, "Compact FMCW

design for short range millimeter-wave radar imaging
applications," 2011 IEEE RadarCon (RADAR), 2011, pp. 885-
888, doi: 10.1109/RADAR.2011.5960664.

[4] Yeonghwan Ju, Youngseok Jin and Jonghun Lee, "Design and
Fig 7. Metal plate detection result. implementation of a 24 GHz FMCW radar system for
automotive applications," 2014 International Radar
Conference, 2014, pp. 1-4, doi:
10.1109/RADAR.2014.7060385

[5] T. Tongboonsong, A. Boonpoonga, K. Phaebua, T.

Lertwiriyaprapa and L. Bannawat, "A Study of an X-band
FMCW Radar for Small Object Detection," 2021 9th
International Electrical Engineering Congress (iEECON),
2021, pp. 579-582, doi:
10.1109/iEECON51072.2021.9440348.

[6] E. Hyun, W. Oh and J. -H. Lee, "Multi-target detection

algorithm for FMCW radar," 2012 IEEE Radar Conference,
2012, pp. 0338-0341, doi: 10.1109/RADAR.2012.6212161.

Fig 8. Drone detection result.

VI. CONCLUSION
Design and experiment of FMCW radar signal for drone
is proposed in this paper. A metal plate and commercial drone
are employed as the radar targets. It is found that the designed
radar system provides a correct position of the targets with an
error of 26.3 cm. The horizontally linear polarization of the Tx
and Rx antennas make amplitude of the radar signal from
horizontally targets are higher than vertically target

ACKNOWLEDGMENT
This research was funded by National Science, Research
and Innovation Fund (NSRF), and King Mongkut’s
University of Technology North Bangkok with Contract no.
KMUTNB-FF-65-26 and this research was funded by King
Mongkut’s University of Technology North Bangkok and
National Science and Technology Development Agency
(NSTDA), Thailand, Contract no. GC 010/2020.

Authorized licensed use limited to: Dr. Ambedkar Institute of Technology. Downloaded on September 03,2025 at 02:18:21 UTC from IEEE Xplore. Restrictions apply.