Professional Documents
Culture Documents
Submitted by:
Hamza Ahmad- F2019019050
Badar Ashraf Rana - F2019019039
Submitted to:
Dr. Faran Awais Butt
January 2023
Design of a Transmitter-Receiver loop for
FMCW Radar for Advanced Driver
Assistance System
Submitted by:
Hamza Ahmad - F2019019050
Supervised by:
Dr. Faran Awais Butt
Advisor
(Dr. Faran Awais Butt)
School of Engineering
Contents
List of Figures .........................................................................................................................
iv
Abbreviations ......................................................................................................................... 2
Abstract ....................................................................................................................................
3 1 Introduction ........................................................................................................................
Matlab .......................................................................... 14
References…………………………………………………………………………………………………………...30
iii
Contents iv
List of Figures
2.1 Structure of an FMCW Radar System……………………………………………………………..6
3.1 Spectrogram of Saw-tooth Wave Chirp for FMCW Radar………………………………...8
3.2 Spectrogram of FMCW saw-tooth Signal Model…………………………………………….10
5.1 Time-Domain Representation of Saw-tooth Modulated FMCW Transmitted
Signal………………………………………………………………………………………………………….14
5.2 Spectrogram of Saw-tooth Modulated FMCW Radar……………....……………………..15
5.3 Received Dechriped Signal of Saw-tooth Modulated FMCW Radar………………...15
5.4 Range-Doppler Response Pattern of Saw tooth Modulated FMCW Radar……….16
5.5 Spectrogram of Triangular Modulated FMCW Radar……………………………………..17
5.6 Received Dechriped Signal of Triangular Modulated FMCW Radar………………...17
5.7 Range-Doppler Response Pattern of Triangular Modulated FMCW Radar………18
6.1 Block Diagram of 5.8GHz FMCW System……………………………………………………….19
6.2 Link Budget on Receiver Side for 5m Range of 5.8GHz FMCW System…………….21
6.3 Input Parameters of Receiver Side for 5m Range of 5.8GHz FMCW
System………………………………………………………………………………….……………………..21
6.4 Link Budget for 5m Range of 5.8GHz FMCW System………………………………………21
6.5 Link Budget on Receiver Side for 102m Range of 5.8GHz FMCW System........…..22
6.6 Input Parameters of Receiver Side for 102m Range of 5.8GHz FMCW
System………………………………………………………………………………………………………..23
6.7 Link Budget for 102m Range of 5.8GHz FMCW System………………………………….23
List of Tables
2.1 FMCW Radar System Parameters…………………………………………………………………….7
2.2 FMCW PARAMETERS………………………………………………………………………………………7
1
Chapter 2. Problem Statement 2
Abbreviations
DSP Digital Signal Processing
Many types of radars have been used in automotive applications, such as adaptive
cruise control systems. This Complex Engineering Problem (CEP) required us to design
a unique Continuous Wave (CW), radar system where the frequency of the transmitted
signals changes with time. This CEP requires us to develop an FMCW radar system for
advanced driver assistance systems (ADAS) in automotive applications. As every
industry in the automotive sector is moving towards automation and greater safety for
their customers, this field has been a hot topic for many years. They adapt quickly to
these changes, but there is still much to do in this area.
3
Chapter 2. Problem Statement 4
Chapter 1
Introduction
There are many kinds of radars available for automotive applications, such as
automotive adaptive cruise control systems. We use Frequency Modulated Continuous
Wave radar (FMCW) because it is cheap and has three types of modulation models. This
is the future of automotive, with every industry moving towards safety and automation.
They adapt quickly to these changes, but there is still much to be done in this area. This
CEP will design a radar system based on Saw-tooth and Triangular modulated FMCW
radars. We'll also simulate the design in Matlab to compare its performance.
The 24GHz FMCW radar has been created with a single receiver antenna and
transmitter. FMCW radar can be achieved using the tuning range of 100MHz with the
8dBm output powers and the 9dB conversion losses at the frequency of 24GHz. The
FMCW radar has been successfully implemented in vehicles to provide vehicle collision
avoidance and autonomous direction for the vehicle [1].
The author presents the Frequency Modulation Continuous wave (FMCW), Radar
operating at 24 GHz with a single transmitter and receiver array. This is a novel, low-
cost, and low-profile antenna. The author also includes a Gunn Voltage-Controlled
Oscillator, (VCO), and the MMIC. These are both driver amplifiers that act as transmitters
for the FMCW radar. The FMCW radar receiver is provided by the Rat race, which is a
hybrid mixer. An antenna with a 100-patch micro-strip antenna is used by the author. It
has a narrow beam width and high gain in FMCW radar. The antenna's gain measured at
24GHz is 22dB [1].
A second FMCW Radar system with an operating frequency of 24GHz was designed and
implemented using Field Programmable Gate Arrays (FPGA) as well as Digital Signal
Processing (DSP). A 24GHz FMCW radar system was also created by the author. It has a
single transmitter antenna array and one transmitting antenna array, which can
measure distances with high accuracy. A test was conducted to verify the accuracy and
range resolution of the radar system. The distance measured was approximately 3cm
within the target range of 0.5m to 3.4m.
FMCW radar is used for automotive applications. DSP chips were used to create a signal
processing system that can be used for FMCW radar. The author verified the radar using
testing and accurate measurement of target distance. The experimental results show the
range estimation accuracy of < 3 cm [2].
Chapter 2
5
Chapter 2. Problem Statement 6
Problem Statement
2.1 Introduction
Frequency Modulated Continuous Wave radar system (FMCW), is a special type of
Continuous Wave radar system. The frequency of the transmitted signals is constantly
changing with time. This Complex Engineering Problem (CEP) requires us to design an
FMCW radar system for advanced driver assistance systems (ADAS).
The FMCW radar system parameters are define in the Table 2.1 which is used to design
and simulate the complete linear FMCW radar.
The FMCW radar's basic waveform transmits and receives the sawtooth waveform.
Triangular waveforms offer additional benefits. We must simulate FMCW radar for
sawtooth and triangular waveforms, and then generate Range-Doppler response
patterns for FMCW linear chirp patterns FMCW using MATLAB. Also, we need to drive
the FMCW radar received signal and transmitted signal time-domain equations. These
expressions are described in Table 2.2
Also, we need to calculate Link Budget for the receiver path for targets at 5m and 102m
using ADIsimRF Analog Devices for 5.8GHz FMCW radar. We have to propose a
hardware circuit to implement the FMCW radar system.
Chapter 3
7
Chapter 2. Problem Statement 8
Derivation of Transmitted-Received
Signal in Time Domain
3.1 Introduction
The first task is to derive both the FMCW radar's transmitted and received signals in the
time domain and it can be used to further simulate the FMCW radar and determine the
object distance and relative position.
As you can see, the transmitted frequency increases linearly with time during Sweep (T).
In our case, the operating frequency or we can say the starting frequency fo is 77GHz
which is given in the FMCW radar system parameters in Chapter 2.
We can determine the frequency of our transmitted signal at any time t by using the
Equation 3.1.
(3.1)
Chapter 3. Derivation of Transmitted-
Received Signal in Time Domain
Here the is called the chirp rate, it gives us the information of the speed of frequency
change.
(3.2)
The instantaneous phase in the Frequency Modulation (FM) can be determine by the
given Equation
o (3.3)
By putting the Equation 3.1 in the Equation 3.3 to get the instantaneous phase.
o (3.4)
In the Equation 3.4, ϕo is the initial phase of the transmitted signal. The FM modulated
transmitted signal can be determine by the Equation 3.5.
We can get the FMCW radar transmitted signal in the first sweep by putting the
instantaneous phase in the FM modulated signal in Equation 3.5.
(3.6)
The Above Equation 3.6 gives the transmitted signal s(t) = xtx(t) for only first sweep.
But we transmit the N number of sweep at shown in the Figure 3.2.
To determine the N number of Sweep’s transmitted signal which is in our case N = 64.
The T will be modified in the Equation 3.6.
Chapter 3. Derivation of Transmitted-
9
Received Signal in Time Domain 10
t = nT + ts (3.7)
Here 0 < ts < T, so we get the final transmitted signal which is given below:
(3.8)
Here the ts is the starting time of nth sweep.
(3.9)
(3.10)
Here the R is the object distance and the v is the relative speed of the object and the c is
the speed of light. So, the received signal will be
Chapter 3. Derivation of Transmitted-
Received Signal in Time Domain
Equation 3.11 shows the Received Signal. FMCW radar uses the received signal
multiplied by the transmitted signal to obtain the message signal. The beat frequency is
the difference between the received and transmitted signals. The beat frequency fb can
be used to determine the distance of an object.
As transmitted and received signals are the cos signals. So, we have to apply the cos
multiplication trigonometric formula which is given below:
(3.13)
After the multiplication of the received and transmitted signal, we get a higher frequency
component that is filtered by the band pass filter. After filtering, we get the message
signal as follow:
After some simplifications in the Equation 3.14 we get the simplified form of the
massage signal.
(3.15)
Chapter 3. Derivation of Transmitted-
11
Chapter 4. Expressions for a Linear FMCW Radar System’s Parameters
Chapter 4
Pr GtGrσλ2
= (4.1)
Pt (4π)3r4
Here Pt is the radar transmitted power and Pr is the radar received power & Gr and Gt are
the gain of receiver and transmitter antenna and σ is the radar cross-sectional area in
(m2) and λ is the wavelength of the radar in (m).
4.2.2 Range
The range of the target is obtained from the beat frequency as shown in the Equation 4.2.
(4.2)
Here fb is the beat frequency which is the frequency difference of the transmitted and
received signal, and c is the speed of light and α is the Chirp rate. We can also get the
12
(4.3)
(4.4)
Here the ∆fb is the frequency resolution. In order to detect the two different object on
minimum range one condition is applied on the frequency resolution which is given
below:
(4.5)
The beat frequency must be greater than the chirp frequency. So, we get the range
resolution in the given Equation:
(4.6)
From the above Equation 4.8 we get the velocity equation of a moving target.
fdc
vr = (4.8)
2fo
It depends on the velocity resolution of the radar which is the minimum relative velocity
that can accurately be measured by the radar of a moving target. The relative velocity of
the moving target also depends on the Doppler frequency fd which is described below:
Chapter 4. Expressions for a Linear FMCW Radar System’s Parameters
Chapter 5
14
Chapter 5. 16
Simulation of FMCW Radar on Matlab
We have also simulate the spectrogram on Matlab to see the saw-tooth wave on the
spectrogram which is shown in Figure 5.2.
We also simulate the 2-D Range Doppler response pattern by simulating the FMCW
radar model on Matlab with a target car distance of 50m and speed of 150Km/h when
the radar speed is 172Km/h then we get the Range-speed response pattern with the
distance and the relative speed of the target with the radar which is shown in the Figure
5.4.
We also simulate the 2-D Range Doppler response pattern by simulating the FMCW
radar model on matlab with target car distance of 50m and speed of 150Km/h when the
radar Simulation of FMCW Radar on Matlab
Chapter 5. 19
speed is 172Km/h the we get the Range-speed response pattern with the distance and
the relative speed of the target with the radar which is shown in the Figure 5.7.
Our task in FMCW radar is to calculate the link budget for received paths for targets at
5m and 102m using ADIsimRF software for Analogue Devices for 5.8GHz FMCW radar.
To calculate the link budget for 5m and 102m, first we need to calculate the received
power Pr in dBm on the Receiver antenna, which can be calculated by using this given
Equation:
The Figure 6.1 shows the total FMWC radar system block diagram in which the value of
Pt, Gt and Gr is given in dBm and dBi. First we have to convent these value into decimal
then calculate the Pr using Equation 6.1.
(6.2)
(6.3)
Now we’re going to find the received power Pr when r = 5m by using Equation 6.1.
Figure 6.2: Link Budget on Receiver Side for 5m Range of 5.8GHz FMCW System.
By putting all the known values on ADISimRF as seen in Figure 6.2 and 6.4 it
automatically calculated the other parameters which are shown in the Figure Now we
will calculate the link budget for the range 102m for that we are again going to use
Equation 6.1
Figure 6.3: Input Parameters of Receiver Side for 5m Range of 5.8GHz FMCW System.
By putting the all the known values on ADISimRF as seen in Figure 6.5 and 6.7 it automatically
calculated the other parameters which are shown in the Figure Below.
Figure 6.5: Link Budget on Receiver Side for 102m Range of 5.8GHz FMCW System.
Figure 6.6: Input Parameters of Receiver Side for 102m Range of 5.8GHz FMCW System.
Chapter 6. Link Budget Calculation 24
Figure 6.7: Link Budget for 102m Range of 5.8GHz FMCW System.
Chapter 7
Conclusion
In this Complex Engineering Problem (CEP), we have designed a Frequency Modulated
Continuous Wave (FMCW) Radar for automotive applications especially for the
Advanced Driver Assistance System (ADAS). It can be used to implement autonomous
braking and radar cruise control. In this CEP we have simulated the two types of
modulated FMCW radar and compared its performance and the Pros and cons of each
type of modulated FMCW radar. We also drive the time domain expression of the
transmitted and the received signal of the FMCW radar system. With the help of the time
domain expression, this task can be implemented on the hardware, and circuit-level
implementation can be proposed on the given specification of the radar.
24
Appendix A
db2pow(min(10*log10(car_dist)+5,20)); cartarget =
phased.RadarTarget(’MeanRCS’,car_rcs,’PropagationSpeed’,c,...
channel = phased.FreeSpace(’PropagationSpeed’,c,...
’OperatingFrequency’,fc,’SampleRate’,fs,’TwoWayPropagation’,true);
Appendix A. Win Spice Coding of MOSFET’s Schmitt Trigger 26
phased.Transmitter(’PeakPower’,tx_ppower,’Gain’,tx_gain); receiver =
phased.ReceiverPreamp(’Gain’,rx_gain,’NoiseFigure’,rx_nf,... ’SampleRate’,fs);
specanalyzer = dsp.SpectrumAnalyzer(’SampleRate’,fs,...
’PlotAsTwoSidedSpectrum’,true,...
’Title’,’Spectrum for received and dechirped signal’,...
’ShowLegend’,true); rng(2012); Nsweep = 64; xr =
complex(zeros(waveform.SampleRate*waveform.SweepTime,Nsweep)); for
m = 1:Nsweep
transmitter(sig);
’DopplerOutput’,’Speed’,’OperatingFrequency’,fc,’SampleRate’,fs,...
’RangeMethod’,’FFT’,’SweepSlope’,sweep_slope,...
’RangeFFTLengthSource’,’Property’,’RangeFFTLength’,2048,...
’DopplerFFTLengthSource’,’Property’,’DopplerFFTLength’,256);
clf;
plotResponse(rngdopresp,xr); % Plot range Doppler map axis([-v_max v_max 0 range_max]) clim = caxis;
rootmusic(pulsint(xr_d,’coherent’),1,fs_d); rng_est =
beat2range(fb_rng,sweep_slope,c)
peak_loc = val2ind(rng_est,c/(fs_d*2)); fd =
deltaR = rdcoupling(fd,sweep_slope,c)
waveform_tr = clone(waveform);
release(waveform_tr); tm = 2e-3;
waveform_tr.SweepTime = tm; sweep_slope = bw/tm;
deltaR = rdcoupling(fd,sweep_slope,c)
transmitter,channel,cartarget,receiver); fbu_rng =
rootmusic(pulsint(xr(:,1:2:end),’coherent’),1,fs); fbd_rng =
rootmusic(pulsint(xr(:,2:2:end),’coherent’),1,fs); rng_est =
beat2range([fbu_rng fbd_rng],sweep_slope,c) fd = -
txchannel = phased.TwoRayChannel(’PropagationSpeed’,c,...
’OperatingFrequency’,fc,’SampleRate’,fs); rxchannel =
phased.TwoRayChannel(’PropagationSpeed’,c,... ’OperatingFrequency’,fc,’SampleRate’,fs);
Nsweep = 64; xr = helperFMCWTwoRaySimulate(Nsweep,waveform,radarmotion,carmotion,...
transmitter,txchannel,rxchannel,cartarget,receiver);
plotResponse(rngdopresp,xr); % Plot range Doppler map axis([-v_max v_max 0 range_max]); caxis(clim);
Appendix A. WinSpice Coding of MOSFET’s Schmitt Trigger 29
sFMCW = phased.FMCWWaveform(’SweepBandwidth’,bw,...
’SampleRate’,fs,’SweepDirection’,’Triangle’,...
’NumSweeps’,4); sig =
step(sFMCW); windowlength
= 32; noverlap = 16; nfft =
32;
spectrogram(sig,windowlength,noverlap,nfft,sFMCW.SampleRate,’yaxis’) title(’FMCW
Transmitted signal spectrogram’);
[2] Yadav, Rekha & Dahiya, Pawan & Mishra, Rajesh. (2016). A high performance 76.5
GHz
FMCW RADAR for advanced driving assistance system. 383-388.
10.1109/SPIN.2016.7566724.
[3] Eun Hee Kim and Ki Hyun Kim. Random phase code for automotive mimo radars
using combined frequency shift keying-linear fmcw waveform. IET Radar, Sonar &
Navigation, 12(10):1090–1095, 2018.
30