Experimental Approach of FMCW Signal Generation

using Direct Digital Synthesizer Module

Kusmadi† and Achmad Munir
Radio Telecommunication and Microwave Laboratory
School of Electrical Engineering and Informatics, Institut Teknologi Bandung, Indonesia
Bandung, Indonesia
†[email protected]

Abstract— In this paper, the process of generating a signal

frequency modulated continuous wave (FMCW) using direct II. BRIEF OVERVIEW OF FMCW SIGNAL GENERATION
digital synthesizer (DDS) will be presented starting from the
design process simulation using the software on the computer until A. Direct Digital Synthesizer
the realization and measurement of the generation of such signals Direct Digital Synthesizer (DDS) is a frequency generation
on a module DDS, the kits from Analog Devices. Generation techniques using digital data processing blocks are processed
conducted a baseband signal with a bandwidth of 50 MHz in the such that the output frequency and phase produced can be
frequency range of 50-100 MHz, the output power generated from
regulated based on the reference clock that is set as the source.
-3.90 up to -7.80 dBm. FMCW signal is raised using Chirp
In essence, the reference clock divided in blocks of internal DDS
waveform with the reference clock of 30 MHz and can be applied
as a signal generator on the system Radar.
uses a particular scale factor in binary. In general block diagram
and the advantages of the use of DDS can be shown in Fig. 1 [5].
Keywords—FMCW; DDS; baseband; chirp; radar.

I. INTRODUCTION
Based on the form of signals used in a radar device, is
divided into two types of pulse radar and radar continuous wave
(CW) [1]. The technique used in radar continuous wave least
two kinds are already very common and has the potential to
continue to be developed, namely FMCW and stepped
frequency continuous wave (SFCW).
FMCW radar technique that uses a radar Chirp waveform in
the form of a continuous wave. This continuous wave is very
good at detecting moving objects, where the reflected waves
generated by a moving target will generate a frequency shift due
to the Doppler effect [2]. However, with the continuous-wave
radar is not able to provide the target distance information from
the transmitter. This is due to calculate the distance, use the Fig. 1. Blok system Direct Digital Synthesizer.
delay of the signal send and signal pantulnya. The weakness of
the radar with this technique can be overcome by making some DDS advantages :
sort of marker of time to modulate other continuous wave which
is transmitted [3]. · Micro-Hertz tuning resolution of the output frequency and
sub-degree phase tuning capability, all under complete digital
There are several components required for the control.
implementation of FMCW techniques in the generation of · Extremely fast “hopping speed” in tuning output frequency
signals used in radar systems, one of which is commonly used (or phase), phase-continuous frequency hops with no
and is an important part that should be there is a voltage over/undershoot or analog-related loop settling time
controlled oscillator (VCO). however, there is a loss of the VCO anomalies.
is the kind of active components that are not linear. On the other · The DDS digital architecture eliminates the need for the
hand, the advantage is the use of VCO can be used for wide manual system tuning and tweaking associated with
bandwidth and high operating frequency. To overcome the component aging and temperature drift in analog synthesizer
losses obtained when using VCO selection, by the use of DDS solutions.
be a better option to overcome these problems. The ability to · The digital control interface of the DDS architecture
produce hundreds of mega herts bandwidth, linearity also facilitates an environment where systems can be remotely
frequency modulation precision becomes more advantage [4]. controlled, and minutely optimized, under processor control.


ª
*&&&

 DDS output frequency value is obtained from the equation III. SIMULATION AND EXPERIMENTAL OF FMCW SIGNAL
(1). GENERATION
$ %&'
!"# (1) A. Simulation Design
2(
Schematic diagram of DDS module simulation design
shown in Fig. 3.
where M is tunning words, fclk is a reference clock, and n is the
number of bits phase accumulator. DDS output signal needs
attention, mainly due to the emergence of other signals that are
unwanted. These signals are emitted when participating along
with the original signal, it can lead to ambiguous that can cause
errors in the detection process on a radar system. Therefore, the
signal processing includes filtering is required.

B. Principle Operation of FMCW Radar

FMCW radar using Chirp signal, shown in Fig. 2, then the
beat frequency fb can be obtained using the equation (6). Fig. 3. Schematic diagram of DDS design simulation.
∆
), (2) The simulation results of the DDS module, referring to the
∆+
circuit in Fig. 3 shown in Fig. 4.
where Δf is the frequency range of chirp and ΔT is the duration
of one period of time to reach chirp. In the process, the
processing of the FMCW radar signal requires conversion from 6

analog to digital form and the fast Fourier transform (FFT) for mod_out
calculating range of the spectrum. The maximum beat frequency vout, volt 4

that can be clearly measured according to the Nyquist criterion, 2

is half the sampling frequency fs. So that the maximum
unambiguous distance is also determined by how much the 0
sampling frequency is used, as in equation (3). 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
time, usec
1∆ 3 time,
(a)usec
-./0 = (3)
2 ∆+ 2
10
vout, volt

-10
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
time, usec
time,
(b) usec

Fig. 4. Chirp signal of simulation result; (a) voltage controller (b)

sinusoidal waveform
Fig. 2. Block diagram of FMCW radar [7].
The shape of the spectrum of DDS signal generated from the
In general the components needed to build a FMCW radar simulation are shown in Fig. 5.
system shown in Fig. 2. Chirp signal generator which is a
0
complex signal generated using DDS, after filtering on a specific
-10
frequency range and the reinforcement is then transmitted via a
-20
transmitter antenna. The complex signal emitted by a transmitter
Pout, dBm

-30
antenna, the other antenna functions as a receiver that will
-40
receive the signal reflected from the object is detected and the
-50
form of complex signal as well, so do the separation between the
-60
In-phase (I) and Quadrature (Q) parts, and then conduct the
-70
modulation process on the component of I/Q demodulator to
obtain complex signal which has a phase shift proportional to 30 40 50 60 70 80 90 100 110 120
frequency, MHz
the distance of the object that reflects the transmitted signal by
further signal processing [8]-[9]. Fig. 5. Spectrum DDS generation signal of simulation result.
 Components used in the simulation, as shown in Fig. 3, is a
block of programmable impulse signal source, VCO block, and
terminator output of 50 ohm impedance. Chirp signal generated
from the simulation are shown in Fig. 4. The signals are formed
in the form of a triangular signal at the figure (a) is a control
voltage of VCO in order to produce a continuous signal in the
form of Chirp in a predetermined frequency range, ie at 50-100
MHz. While the figure (b) is shown sinusoidal output waveform
of the VCO signal in the form of a continuous signal along the
two periods
(a)
FMCW spectrum output signal generated from the
simulation circuit shown in Fig. 4. Bandwidth obtained at 50
MHz in the frequency range of 50-100 MHz on output power
ranging from -4 dBm up to -2 dBm. Chirp signal bandwidth to
get a sharper, takes bandpass filter in the frequency range and
other parameters right. Therefore, usually the output of the DDS
module will be followed by the BPF module.

B. Experimental Result
DDS module used in this experiment is a type that is
AD895x products from Analog Devices, which can be obtained
easily in the commercial market. The DDS module can be (b)
programmed with a USB connection using the software have
also been easily obtained. It has five output ports, each of which
can be used vary according to the needs. AD895x DDS module
type shown in Fig. 6.

(c)

Fig. 8. Output baseband spectrum of DDS module; (a) at 59.13 MHz, (b)
at 85.73 MHz, and (c) at 86.90 MHz

llustration DDS module testing process is performed to

measure the resulting output waveform, shown in Fig. 7. Where
in the process, the E1 port of DDS module is connected to an
oscilloscope to analyze waveforms are generated, while the E2
port is connected to spectrum analyzer for analyzing the
Fig. 6. AD895x DDS module resulting spectrum. The second output port is a signal in-phase
and quadrature of the FMCW signal generated by the same
signal power but out of phase from each other in 1800 between
one and the other.
The movement of the measured signal spectrum of the DDS
output module can be observed through the pieces of the picture
at a certain time, is shown in Fig. 8. When observed through
video visualization it will show movement. As an illustration,
Fig. 8 (a) can be observed spectrum signal which occurs at time
t indicated the signals measured at 59.13 MHz, in (b) which is
the spectrum at t+n, the position of spectrum signals have been
shown at a frequency of 85.73 MHz, as well as in (c) position
signal spectrum is at 86.90 MHz. On the whole spectrum of the
signal appear and move increases and decreases ranging from a
frequency of 50 MHz to 100 MHz, as already specified
frequency range and bandwidth of the signal FMCW expected.
Fig. 7. Ilustration of DDS module experimental measurement The period of the wave will be repeated continuously.
 FMCW signal output in the form of in-phase and quadratue The DDS is suitable for linear frequency modulation, and it is
measured from the DDS output port module E1 and E2 are easily controlled by programming binary code. FMCW
shown in Fig. 9. Channel 1 on the oscilloscope is derived from spectrum bandwidth required signal of 50 MHz in the frequency
E1 port that can function as an in-phase signal, while the second range of 50-100 MHz successfully simulated and realized on
channel oscilloscope comes from port E2 has a phase difference DDS module. Output power is generated through simulations
of 1800 with the output signal E1 port, can function as a signal and measurements of the same approach in the range of -3 dBm.
quadratue. The magnitude of the power released through both
ports can be set as desired. REFERENCES
[1] B. R. Mahafza, Radar Systems Analysis and Design Using MATLAB, 3rd
Edition, CRC Press, 2012.
[2] H. J. Kramer, Observation of The Earth and Its Environment Survey of
Missions and Sensors, 4th Edition, Springer-Verlag, 2001.
[3] M. Skolnik, Radar Handbook, 3rd Ed, Mc Graw Hill, 2008.
[4] Yun-Taek Im, Jee-Hoon Lee, and Seong-Ook Park, A DDS and PLL-
based X-band FMCW Radar System, KAIST-ICC, IEEE, 2011.
[5] A Technical Tutorial on Digital Signal Synthesis, Analog Devices, Inc.,
1999.
[6] I. Nicolaescu, P.v. Genderen, V. Dongen, J.v. Heijenoort, P. Hakkaart,
“Stepped frequency continuous wave radar data preprocessing” in Proc.
of 2nd International Workshop on Advanced GPR, May 2003.
[7] Kusmadi and A. Munir, “Simulation Design of Compact Stepped-
Frequency Continuous-Wave Through-Wall Radar,” in Proc. of 5th
International Conference on Electrical Engineering and Informatics, Bali,
Indonesia, Aug. 2015
[8] Azizah, A. B. Suksmono, and A. Munir, "Signal Processing of Range
Fig. 9. I/Q Signal measurement Detection for Radars Using Matlab and GNU Radio", in Proc. of
International Conference on Computer, Control, Informatics and Its
Applications. 2014.
IV. CONCLUSION [9] A. B. Suksmono, A. Pramudita, E. Bharata, A. A. Lestari, N. R. Sambas,“
Improved Bow-Tie Antenna for Pulse Radiation and Its Implementation
n experimental of FMCW signal generation using direct in a GPR Survey” in Proc. of International Conference on Electrical
Engineering and Informatics, Jun. 2007.
digital synthesizer module is designed and tested successfully.