See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/228351511

Comparison of TOF, FMCW and Phase-Shift Laser Range-Finding Methods by

Simulation and Measurement

Article · January 2006

CITATIONS READS

14 6,868

4 authors, including:

Saeed Olyaee
Shahid Rajaee University
123 PUBLICATIONS   796 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

2nd West Asian Colloquium on Optical Wireless Communications, 27-28 April 2019, S Beheshti Univ., Tehran, IRAN View project

A High Quality Factor and Wide Measurement Range Biosensor Based on Photonic Crystal Nanocavity Resonator View project

All content following this page was uploaded by Saeed Olyaee on 23 November 2018.

The user has requested enhancement of the downloaded file.

 Quartarly Journal of
Technology & Education
Vol.1, No.1 Autumn 2006
Comparison of TOF, FMCW and
Phase-Shift Laser Range-
Finding Methods by Simulation
and Measurement
Shahram Mohammad Nejad
University of Science and Technology
[email protected]
Saeed Olyaee
Shahid Rajaee Teather Training Uaiersity
[email protected]

Abstract
Since 1970s the use of laser based range-finders has become a
necessary part of many military and domestic systems.
According to the needs and the circumstances, many different
types of such systems have been developed and their
technologies are adapted to the local needs. Distance or
displacement measurement in the medium range is mainly
divided to three groups; time of flight, frequency modulation
continuous wave, and phase-shift laser range finder. In this
paper we present the design, implementation and/or simulation
of three such systems and compare their characteristics.

Keywords
Time-Of-Flight, Laser Range
Finder, FMCW-Like distance
measuring system, phase
shift, heterodyne technique.
Comparison of TOF, FMCW and Phase-Shift Laser Range-Finding Methods by Simulation and Measurement Shahram Mohammad Nejad et.al

1. Introduction reflection and 75% for the transmission losses of the

media, and assuming a perfectly round optical lens of
Three major techniques are usually being used for range 5cm radius, in a maximum distance of 20Km
finding processes. These techniques are: Time of Flight measurement by a 1.5MW laser source, the reflecting
(TOF), frequency modulation continuous wave power will be 70.3nW only.
(FMCW) and Phase-Shift [1-5]. The phase-shift and It can also be seen from fig.1 that, a START
other techniques such as FMCW or the combination of signal is simultaneously introduced to the processing
interferrometry and phase-shift techniques could also be circuit through a p-i-n diode. As the START is signal
used to measure the intermediate distances with high received, a counter is activated, until the STOP signal
accuracy [6-8]. But, none of these techniques could be arrives. After travelling a distance of 2r , the laser
used to measure distances above 1km, since in these beam, trough the optics, arrives on the APD having
techniques continuos wave (CW) or burst lasers are high responsivity and a short rising time. In order to
used. stop the counter, the APD photocurrent is converted into
On the other hand, to detect the reflecting a voltage signal and is amplified. If the clock pulse
beam from long distances, lasers having relatively high frequency is f clk , the distance between the target and
powers are required. Also, the use of high power lasers the transmitter is,
with continues wave or burst is impossible. Hence, for C.N
r=
measuring distances above 1km, the use of pulsed 2 f clk
lasers are necessary. As the repetition of the transmitter (2) where N accounts for the counted digits of the
power increases, the system becomes complicated and counter (between the START and the STOP signals)
also cooling becomes inevidable. If however, the and C is the speed of light. In accordance to equation
repetition rate decreases, the detection of a moving (2), the resolution of the system related to a clock pulse
target become very difficult. It is therfore necessary to of the counter becomes,
compensate between the maximum distance to be
C
measured, the repetition rate and the system’s r = (3)
complication. 2 f clk
The main function of the transmitting section is
2- TOF Laser Range Finder to produce a relatively powerful short period pulse,
adjust its repetition rate and also produce the START
In a TOF laser range finding system, the round trip time signal. The schematic drawing of the transmitter circuit
measurement of a short powerful laser pulse is used to is shown in fig.2. The main sections of the system are:
determine the distance [10-13]. With reducing the pulse a) The laser (including the Nd:YAG laser media, the
width of the signal, the output power of the laser signal polarizer, the Q-Switch crystal, the cavity mirrors, the
could be increased to several MW and the signal to flash lamp and a p-i-n diode).
noise ratio could also be increased considerably. The b) The 4.5KV supply.
most important sections of a TOF system are: c) The driver and repetition rate regulator.
a) The transmitting section including, the Q-switch, d) The laser cooling system.
the HV source to drive the flash tube, the repetition rate As can be seen from fig.2, using a Q-Switch
controlling part, the cooling system and the transmitter and the related driving and oscillating circuits, the
optics. repetition rate is adjusted in the range of 1-20Hz. On the
b) The receiver analog section including, the APD other hand, the laser output pulse variation is adjusted to
detection parts, the low-noise and wide-band the range 10-15ns, which in turn, causes the laser cavity
preamplifier, the converter, the limiter and the receiver output power to vary. In accordance to the measured
optics. 15mJ energy, the output power could be adjusted from
c) The receiver digital section including, the isolating 1.0 to 1.5MW.
circuit, the processing and the interface circuits. The Nd:YAG laser was pumped by a Flash
As can be seen from fig.1, the laser beam travels a Tube. For the Flash Tube execitation a 4.5KV source
distance of 2r. The received power is usually very small. was used. Since the cavity power generation is high, the
The relation between the received power at the optic laser cavity need to be cooled by an especial cooling
system. The starting signal was produced by a silicon
section, Pinc , and the distance r, is defined as [9]:
PIN diode having dark current I d =10nA and a rising
. .S
Pinc = Popt , = c . opt (1) time of
.r 2 r =5ns. When triggering the Q-Switch and

where, is the transmitting coefficient (which is equal directing the laser pulse out of the cavity, a current
to the multiplication of transmitting coefficient of the pulse is produced by the diode and implied to the
optics and the transmitting coefficient of the processing section to start the counting.
media), is the scattering coefficient of the target The range finder receiver consists of two
analog and digital sections. The block diagram of the
( =1 for a perfect mirror and is equal to zero for a
analog section is presented in fig.3. As is shown in the
black body), Popt is the laser output power and S is the figure, the reflected beam is being detected by an APD
receiver optic area. Considering a 1% for the target having responsivity of 28 A / W at the wavelength of

Quartarly Journal of Technology & Education Vol.1, No.1 Autumn 2006

12
Comparison of TOF, FMCW and Phase-Shift Laser Range-Finding Methods by Simulation and Measurement Shahram Mohammad Nejad et.al

= 1060nm . Since the APD is in the current mode of

Optical Filter
Switching Power Supply
operation , the output signal is amplified by a wide band

Lens
APD
preamplifier having an input impedance of 50 . The Preamplifier

HI
and AGC
amplifier output signal is given by,

STOP Signal
2

Vo = Pinc . .Z t . Av (4) 1
Optocupler
Mono
where, Pinc is the received optical power on the APD, Amp. Comp. Stable

is the APD responsivity, Z t is the transimpedance

0 0
gain of the preamplifier and Av is the amplifier voltage Refrence Voltage

gain. From the designed circuits, the ratio of the output Fig.3 Schematic presentation of the analogue
section for the receiver.
voltage to the power received by the APD equals
1.4MV / W . Furthermore, it is required that the preamplifier
The signal comparison level due to the gain to be proportional to the received optical power.
minimum received power ( 10nW ) is considered to be This can be accomplished by variation of the APD
14mV. This in turn will reduce the system’s noise responsivity with respect to the biasing voltage. By
effects. The pulsewidth of the signal is then increased varying the bias in the range 320-351V for this APD,
by a monostable and the STOP signal of the counter is the responsivity varies between 18.5A/W to 123.1A/W.
implied to the opto-coupler. It then is implied to the If is the responsivity of the APD and the trans-
digital section of the circuit.
impedance gain is Z t , then,
Vo
LASER
HV A' = = .Z t (5)
POWER
SUPPLY
Pinc
Osc.
PIN DIODE
From this equation, the total gain will vary in the range
of 920 KV/W to 6.17MV/W.
The main internal noise is the one created in
APD Pre AMP. Comp. the receiver STOP channel. Due to the isolation of the
analogue section from other parts, the noise effect of the
other parts is being curtailed. The switching noise of the
Mono Stable
4.5KV flash lamp of the laser is being eliminated by
Stop EMI filters. The noise related to the digital section is
CLK. Microcontroller Start
eliminated by separating the source and the ground from
the analogue section of the receiver. However, the main
LCD noise related to the receiver channel is the noise of the
preamplifier, which is given by,
Fig.1 Schematic representation of a time-of-flight range
2 4 KT
finder system. in = ( fi ) (6)
Reqi 2
where, k is the Boltzman constant, T is the absolute
Osc. 1-20Hz temperature, Reqi is the preamplifier’s input impedance,
PIN Diode
Polarizer Polarizer and f i is the system bandwidth, which is 100MHz.
Considering the 50 impedance for the preamplifier’s
Calibration input, the input current noise is 228nA. The equivalent
circuit
optical power will then become,
Laser
in
Beam Pinc = (7)
Cooler
Q-Switch
Crystal
Considering the 28 A / W responsivity for the APD, the
PS 4.5KV
minimum detecting power becomes 12.7nW.
It is well known that signal-to-noise ratio (S/N)
Fig.2 Schematic representation of the transmitter circuit
for a TOF laser range finder. is one of the main elements determining the resolution
of the system. For our system, the relation between the
resolution of the system and the S/N could be presented
by
C n 0.35C
d = . = (8)
2 dU / dt 2 B.SNR
where dU / dt is the slope of the time dependent pulse
and n is the rms noise value. For repetition rate of N
(1<N<15), Equation (8) changes into,

Quartarly Journal of Technology & Education Vol.1, No.1 Autumn 2006 13

13
Comparison of TOF, FMCW and Phase-Shift Laser Range-Finding Methods by Simulation and Measurement Shahram Mohammad Nejad et.al

0.35C (9) 2 kHz and the VCO works linearly from 55 MHz to 130
=
d
2 N .B.SNR MHz. At the end of the transmitter section, a laser
On the other hand the signal to noise ratio could be driver converts the VCO output to the light power. In
presented as, receiver section, an APD photodetector collects the
iS M Pinc
back scattered light.
SNR = = (10)
[2q(k )]
1
in
TP Pinc M F ( M ) + i
2 2
na
2
_ processing data
where kTP = 0.5 , M is the internal gain of the optical + primary data
detector and F(M) is the excess noise factor. The excess
noise factor in terms of the effective ionization ratio can
be presented as
2
M 1
F ( M ) = M 1 (1 k eff ) (11)
M
In this design an appropriate algorithm is
designed to reduce the measurement error. Considering
the sampling rate 20Hz and the 2sec for refreshing the
display, there will be 40 samples in each display. By
omitting the out of range data, the measuring error is
considerably reduced. Fig.4 presents the measured data
for approximate distance of 1.17Km. For primary data
the average and standard deviation are 1185.57m and Fig. 4 Measured data for distance of 1.17Km.
114.95, respectively. After processing the data change
to 1175.23m and 2.92 respectively.

3. FMCW Laser Range Finder

In FMCW-like method, the laser power is frequency

modulated with a ramp or sinusoidal wave. The
received signal has a time delay with respect to emitted
one. If these two signals are mixed, beat frequency will
be obtained (fIF=fe(t)-fr(t)).
As is shown in fig. 5, beat signal (IF) has a
transient behavior when slope of the ramp wave
changes from upward to downward or vise versa.
Except for the unstable zones, the beat frequency can be
measured at other points. The distance can be
approximated by following equation:
F .c.T
d= a r (12) Fig. 5 Ramp wave modulating VCO frequency from Fmin
4. f to Fmax (top) and resultant beat frequency (bottom).
where, Fa is the beat frequency, c is the speed of the
light, Tr being the period of the ramp wave and f For maximum and minimum allowable
stands for modulation frequency bandwidth. distances we have:
The electronic components also introduce an 4. f .d max
additional delay in the received signal which can be Fa max = = 60 KHz
c.Tr (13)
accounted for during the calibration process. According 4. f .d min
to the measuring time intervals, minimum detectable Fa min = = 4 KHz
c.Tr
distance corresponds to a half period of beat signal, and We need a band pass filter with cutoff frequencies lower
c than 4kHz and higher than 60kHz. The remaining is IF
d min = . On the other hand, time delay between
4. f signal and there is one more step to attain the distance
emitted and received signals can’t exceed data. There are two ways to measure this frequency and
Tr T c.T convert it to the distance. The first is counting signal
( d max = r ), so d max = r . Practically, the cycles during the measuring time interval, and the
2 2 4
maximum distance is limited by other factors and second is measuring width of the signal to attain period
becomes less than this value. and then Fa = 1 / Ta . Note that the beat frequency must
Figure 6 shows the block diagram of a system be measured out of transition zones, and of course a
that measures distances from 2m to 30m. As shown, time delay ( d ) is needed to receive the transmitted
frequency modulation is produced by a VCO whose signal.
input voltage is a ramp wave. The ramp wave period is

Quartarly Journal of Technology & Education Vol.1, No.1 Autumn 2006

14
Comparison of TOF, FMCW and Phase-Shift Laser Range-Finding Methods by Simulation and Measurement Shahram Mohammad Nejad et.al

employed. If se is the transmission (emission) signal,

s r is the receiving signal, and s 2 being the reference
(auxiliary) signal, then,
se = S e cos( 1t )

s r = S r cos( 1t + d )

s 2 = S 2 cos( 2 t+ )
(15)

Fig.6 Block diagram of the designed and simulated system

In what follows, we simulate our designed

system and obtain resolution for counting and timing
methods. Figures 7 and 8 represent waveform and
spectrum of the transmitted and mixed signals,
respectively, for d=15m.
After the filtering, the beat frequency is
converted into distance by using a range finding
algorithm. Figures 9 represent distances obtained by
counting and timing methods for d=2m and d=30m. We
can see that the distance error for counting algorithm is Fig. 7 The waveform (top) and Freq. spectrum (bottom) of
transmitted signal.
less than 1m for both figures as proved earlier by
theoretical principles. As is shown, in counting method,
one half period is required to establish the distance data.
On the other hand, timing algorithm gives us an error
less than 0.036m for d=30m.

4. Phase-Shift Laser Range Finder

In the phase-shift method, the transmitted light intensity

is modulated sinusoidally, and the round-trip time is
turned into phase shift [4,5,7]. If f1 is the small signal
modulation frequency, the phase shift is presented by,
2d
= 1 d = 2 f1
c
(14)
There are several methods of measuring the resultant
phase shift In this research, because the modulation Fig. 8 The waveform and Frequency spectrum of the
frequency is high, heterodyne technique is being mixed signal.

Quartarly Journal of Technology & Education Vol.1, No.1 Autumn 2006 15

Comparison of TOF, FMCW and Phase-Shift Laser Range-Finding Methods by Simulation and Measurement Shahram Mohammad Nejad et.al

Fig. 10 The block diagram of a phase shift laser altimeter

system.
Fig. 9 Distance obtained by full-wave counting (top) and
timing (bottom) for d=30m.

According to these equations, it is evident that Since the frequency of the designed circuit is
the transmitting and the receiving frequencies are equal. relatively low, the system error ( d ) due to the cross-
The term d is the signal-phase-delay due to the round talk, as a function of the distance could be expressed as:
C "0
trip. The reference signal frequency is f 2 and its phase d = . 0
=
2 2 f1
difference with the main signal is . Based on the (18)
4
sin[! 0
f 1 (r rmax )]
heterodyne technique, the multiplication of the C C
arctan
4 f1 A0 rmax 2 4
transmitting and receiving signals into the auxiliary .( ) + cos[! 0 f 1 (r rmax )]
B r C
signal results,
In this equation, 0 " 0 is the phase difference
X e = se .s 2 = 12 S e S 2 {cos[( )t ]
1 2
between the working and the measured signals, and
+ cos[( 1 + 2 ) + ]} (16) A0 / B is the ratio of the main signal (the signal related
X r = s r .s 2 = S r S 2 {cos[(
1
2 1 2 )t + ( d )] to the photons) to the leakage one. Selecting the
+ cos[( modulation frequency with respect to the final divider
1 + 2) + ( d + )]}
Both signals have a differentiation and addition as f1 =371.838 kHz, and the related maximum distance
frequencies. Using a low-pass filter to annihilate the of rmax =403m, and assuming a maximum phase
addition part, divination between the working and the leakage signals
ye = Y e cos( m t ) to be ! 0 = 90 o , one could measure the deviation of the
distance due to cross-talk for different distances. A
y r = Y r cos( m t+ d ) graph of the distance error variation for leakages as a
(17) function of distance for the ratio of A o /B equal to
where, d (in radian) is the phase difference between the 10dB, 50dB, and 100dB is shown fig.12. As can be seen
tow final signals. It is equal to that considered for the from the figure, by increasing A o /B, one can decrease
case of the round trip. Figure 10 shows the general
the measured error considerably.
block diagram of a phase-shift laser altimeter system,
By shielding the system properly, and
giving a primary insight to the important sections of the
decreasing the laser diode’s driving voltage amplitude,
system.
the coupling capacitor effect was decreased

16 Quartarly Journal of Technology & Education Vol.1, No.1 Autumn 2006

Comparison of TOF, FMCW and Phase-Shift Laser Range-Finding Methods by Simulation and Measurement Shahram Mohammad Nejad et.al

considerably. Since increasing the cutoff frequency With IF Measurement

could decrease the induced coupling, the input 1200 Without IF Measurement

impedance of the current to voltage converter was also 1000

relatively decreased. Of course, care should be taken to 800

not decrease the detector gain too much. A further
600
decrease of cross-talk was possible by isolation of the
transmitter and receiver power supplies. In general, 400

considering the A/B =100 dB in the simulation, the 200

maximum cross-talk error is less than 0.2 percent in the

worst condition. 0.1 0.3 0.5 0.7 0.9 1.1 1.3 1.5

Another possible error of the system is the Fig. 12 The error measurement difference with and
error created by the small variation in the oscillator without IF measurement.
frequency. To reduce or possibly eliminate this error,
one could create both signals by a single oscillator. For Table 1 Comparison of TOF, FMCW and Phase Shift laser
example, an oscillator with frequency f, is employed as range finding methods.
the main oscillator. Designing a synthesizer with Range TOF FMCW Phase-Shift
coefficient of 1.01, a frequency of 1.01f can be assigned Finding
for the local oscillator. Methods
In this case, the IF frequency becomes 0.01f.
However, it should be noted that, this will increase the Laser Source Nd:YAG, Semiconductor Semiconductor
Q-Switch
hardware considerably. Hence, in this system, the
measurement of IF is used to decrease the error. In Power 1.5MW 3mW 10mW
fig.12 the error measurement difference with and Wavelength 1064nm 650nm 1100nm
without IF measurement is shown. As can be seen, the
error resulted from the frequency shift has approached Detector APD APD APD
to zero value. Hence, one can practically conclude that, Range 300-20km 2-30m 2-400m
in a 2 to 400 m range, the error is less than 1.2 percent.
Sampling 1-20Hz 4kHz 52Hz
Rate
5. Comparison
Minimum 12.7nW 200nW 23pW
Detectable
Table 1 depicts the measured and calculated data of the
Power
three above discused systems. It is clearly seen that the
phase-shift method has the best detection ability. Accuracy 5m 36mm 21mm
However, the accuracy to the maximum range ratio is Oscillator 30MHz 4-60MHz 1MHz
best in TOF. Frequency
Driving Three Phase Single Phase Single Phase
Power
Laser Pulse Ramp Sinusoidal
Driving
Waveform
Cost Expensive Low Cost Low Cost
Structural Simple Complex Complex
Complexity
Distance CN Fa C.Tr C
d= d= d=
Measuring 2 f clk 4 f 4f
Equation

6. Conclusion
Fig. 11 A graph of the distance error variation for leakages
as a function of distance for Ao / B =10, 50 and 100dB.
TOF and Phase-Shift range-finding systems were
designed and implemented. FMCW-like system was
also designed and simulated. According to their distinct
characteristics, their applications are classified. Hence,
based on the classification, a user can easily choose the
proper system.

Quartarly Journal of Technology & Education Vol.1, No.1 Autumn 2006 17

17
Comparison of TOF, FMCW and Phase-Shift Laser Range-Finding Methods by Simulation and Measurement Shahram Mohammad Nejad et.al

References

[1] B.Zagar, ‘A Laser-Interferometer Measuring

Displacement with Nanometer Resolution’, [7] M.Vered and A.Aire, ‘Distance Measurement Using
IEEE.Trans.Inst.measu., Vol.43, No.2,PP.332-336 Tow Frequency-Stabilized Nd:YAG Lasers’, Appl.
(1994). Opt. 35, 1996, PP.3010-15.
[2] J.C.Martinez, ‘A robust photo-interferometric [8] T.Kawakami and M.Endo, ‘Adaptive
technique to obtain the refractive index and Multiferequency Modulation Method for an
thickness of non-absorbing stand-alone films’, Advanced Laser Range Finder’, IEEE. Trans. Inst.
Meas.Sci.Tech.11,PP.1138-1144 (2000). measu., Vol.43, No.6, 1994, PP.857-860.
[3] U.Minoni and L.Rovalti , ‘High performance front- [9] Geo-Qing Wei and G.Hirzinger, ‘Active Self-
end electronic for frequency-modulated continuous- Calibration of Hand-Mounted Laser Range
wave interferometers’, IEEE. Trans.Inst.measu., Finders’, Proceeding of the 1997 IEEE, April 1997,
Vol.48, No.6 ,PP.1191-1196 (1999). PP.2789-2794.
[4] I.Fujima, S.Iwasaki and K.Seta , ‘High-Resolution [10] K.maatta, ‘Profiling of hot surfaces by pulsed time
Distance Meter using Optical Intensity Modulation of flight laser range finder techniques’, Applied
at 28 GHz’ ,Meas.Sci.Tech.,PP.1049-1052 (1998). Optics, Vol. 32, No.27, 1993, PP. 5334-5342.
[5] G.Bazin and B.Journet, ‘A New Laser Range-Finder [11] F.Refaat,‘Characterization of advanced avalanche
Based on FMCW-Like Method’, IEEE. Ins. Measu. photodiode for water vapor lidar receivers’,
Tech. Conference, Belgium, June 4-6,PP.90-93 NASA/TP2000, July 2000.
(1996). [12] P.Bhattacharya,‘Semiconductor Optoelectronic
[6] G.Barwood and R.Rowley, ‘High-Accuracy Length Devices and optics’ ,Prentice Hall, Second Edition ,
Meterology using Multiple-Stage Swept-Frequency 1997.
Interferometry with Laser Diode’, Measu. Sci. [13] Ari. Kilpela, ‘Pulsed Time-of-Flight Laser Range
Tech. 9,PP.1036-1041 (1998). Finder Techniques for Fast, High Precision
Measurement Applications’, OULU, 2004.

18 Quartarly Journal of Technology & Education Vol.1, No.1 Autumn 2006

View publication stats