micromachines

Review
Review of Ultrasonic Ranging Methods and Their
Current Challenges
Zurong Qiu 1, *, Yaohuan Lu 1 and Zhen Qiu 2

1 State Key Laboratory of Precision Measuring Technology and Instruments, Tianjin University,
Tianjin 300072, China; [email protected]
2 School of Engineering, Faculty of Engineering and Technology, Liverpool John Moores University,
Liverpool L3 3AF, UK; [email protected]
* Correspondence: [email protected]

Abstract: Ultrasonic ranging has been widely used in automobiles, unmanned aerial vehicles (UAVs),
robots and other fields. With the appearance of micromachined ultrasonic transducers (MUTs), the
application of ultrasonic ranging technology presents a more extensive trend. This review focuses
on ultrasonic ranging technology and its development history and future trend. Going through the
state-of-the-art ultrasonic ranging methods, this paper covers the principles of each method, the
signal processing methodologies, the overall system performance as well as key ultrasonic transducer
parameters. Moreover, the error sources and compensation methods of ultrasonic ranging systems
are discussed. This review aims to give an overview of the ultrasonic ranging technology including
its current development and challenges.

Keywords: ultrasonic ranging; transducer; pulse echo; time of flight; error compensation





 1. Introduction
Citation: Qiu, Z.; Lu, Y.; Qiu, Z.
An ultrasonic transducer is a device that can realize the mutual conversion between
Review of Ultrasonic Ranging
high-frequency electrical energy and mechanical energy. It is generally divided into the
Methods and Their Current
piezoelectric type, capacitive type and magnetoelastic type and is widely used in measure-
Challenges. Micromachines 2022, 13,
ment fields, including distance measurement (in the air) [1], sonar [2], medical imaging [3,4],
520. https://doi.org/10.3390/
ultrasonic therapy [5], non-destructive testing [6,7], flow monitoring [8,9], etc. Ultrasonic
mi13040520
ranging has the advantages of relatively low hardware requirements over radio frequency
Academic Editor: Aiqun Liu ranging and laser ranging in short-range measurement due to ultrasound’s slower trans-
Received: 27 February 2022
mission speed. Besides, ultrasonic is not sensitive to ambient light, electromagnetic inter-
Accepted: 22 March 2022
ference, dust or other factors. In addition, the relatively low cost of ultrasonic transducers
Published: 26 March 2022
is user-friendly for engineering applications such as reversing radar and construction
surveys. Therefore, ultrasonic ranging is suitable for small range (the distance is usually
Publisher’s Note: MDPI stays neutral
less than 10 m) and high-precision (the accuracy can reach the level of millimeter generally)
with regard to jurisdictional claims in
non-contact distance measurement, and the derived two-dimensional positioning [10] and
published maps and institutional affil-
three-dimensional positioning [11,12], object shape recognition [13] and multi-sensor fusion
iations.
trajectory measurement [14], etc.
Ultrasonic ranging originated with underwater sonar in the 1940s, and as it was de-
veloped for use in the air in the 1960s, possibilities for contactless distance measurement
Copyright: © 2022 by the authors.
from 0.2 m to 2 m in the air opened up [15]. In the 1980s, the research and application of
Licensee MDPI, Basel, Switzerland. ultrasonic ranging systems gradually began and was first applied to robot obstacle avoid-
This article is an open access article ance [16–19]. Since the 1990s, the research on ultrasonic ranging methods has gradually
distributed under the terms and deepened, and not only does the traditional time of flight (ToF) method continues to be
conditions of the Creative Commons studied [20,21], but two frequency continuous wave (TFCW) method [22], multi-frequency
Attribution (CC BY) license (https:// continuous waves (MFCW) method [23], binary frequency shift keying (BFSK) method [24]
creativecommons.org/licenses/by/ and amplitude modulation (AM) method [25] have also been proposed, different new
4.0/). measurement methods continue to emerge. Since the beginning of this century, MEMS

Micromachines 2022, 13, 520. https://doi.org/10.3390/mi13040520 https://www.mdpi.com/journal/micromachines

Micromachines 2022, 13, 520 2 of 33

ultrasonic devices have been gradually applied to ultrasonic ranging, and different sig-
nal processing methods have been gradually developed including thresholding method,
curve-fitting method, sliding-window method and optimum correlation detection in [26].
Different coding methods have also been gradually applied, developing ultrasonic ranging
systems towards miniaturization, high precision and high processing speed. At present, in
addition to improving the performance of ultrasonic devices and ultrasonic systems, there
is also optimization of the processing algorithm aimed at the high performance and wide
applications of ultrasonic ranging systems.
Nowadays, ultrasonic ranging is widely used in the automotive industry, UAVs, robots,
and industrial auxiliary measurement equipment. In the automotive industry, it is mainly
used for the detection of obstacles to assist the driving system with achieving anti-collision
function, including distance measurement to obstacles [27,28] and acquisition of kinematics
information for obstacles moving around the vehicle [29]. Additionally, the type of road
surface can be identified through reflected signals [30] to assist in safe driving. Similarly,
ultrasonic sensors can also be applied to obstacle avoidance of UAVs [31,32] and robots [33].
In addition, it can also be used to detect the landing state [34] to ensure the flight safety of
UAVs. In the field of robots, indoor positioning is also a common application. Whether
position coordinate measurements [35] or attitude measurement [36], it is realized through
the fusion of multiple transmitters in the space and multiple receivers on the robot, and
the positioning accuracy can reach the level of centimeters [37]. Meanwhile, the auxiliary
navigation system can realize the path planning of the robot [38]. In the industrial field,
ultrasonic equipment can be used to monitor the field environment such as the tower crane
in [39] to ensure the normal operation of industrial equipment.
Conventional bulk piezoelectric ultrasonic transducers are not practical in mobile ap-
plications due to their large size and high power consumption compared to micromachined
ultrasonic transducers (MUTs) [40]. MUTs can be integrated into portable products, such
as smartphones, wearable devices, etc. As a result of research and technical development,
the performance of MUTs has been improved, and they are hoped to replace the buck
piezo-composite ultrasonic transducers. For example, multiple ultrasonic sensors can be
integrated into mobile devices for distance estimation, which could then be combined
with other information of reflected ultrasonic signals to realize gesture recognition and
classification [41,42], to achieve contactless human–computer interaction. In addition, a
sliding-window based method [43] can be used to improve accuracy. Therefore, MUTs can
be applied to smartphones, VR devices and smart home products.
In this study, the status of ultrasonic ranging is reviewed from different aspects. Sec-
tion 2 introduces the characteristic parameters of ultrasonic transducer related to ranging
application, including frequency and impedance, energy conversion characteristics, and
ultrasonic propagation characteristics, and also analyzes the influence on ranging systems.
Section 3 presents the principle, composition and evaluation parameters of ultrasonic
ranging systems. Section 4 summarizes the various measurement methods and signal
processing methods, as well as the corresponding measurement range, measurement accu-
racy and measurement rate. Section 5 focuses on the error source and error compensation
method of ultrasonic ranging. In the final section, conclusions are drawn, existing problems
are summarized, and predictions of the future development trend are provided.

2. Transducer and Ultrasonic Characteristics Related to Ranging

2.1. Comparison of Different Transducers
An ultrasonic transducer is a kind of energy conversion device, which converts me-
chanical energy to electrical energy or vice versa. Although conventional bulk piezoelectric
transducers can generate high output power, acoustic impedance mismatchs between the
devices and transmission medium severely reduce transduction efficiency [44]. A typical
structure of a conventional piezoelectric ultrasonic transducer is shown in Figure 1a. In
order to reduce the mismatch of acoustic impedance between the piezoelectric layer and
the propagation medium, a matching layer usually is added at the transducer’s front
Micromachines 2022, 13, x FOR PEER REVIEW 3 of 35

the devices and transmission medium severely reduce transduction efficiency [44]. A typ-
Micromachines 2022, 13, 520 ical structure of a conventional piezoelectric ultrasonic transducer is shown in Figure 3 of1a.
33
In order to reduce the mismatch of acoustic impedance between the piezoelectric layer
and the propagation medium, a matching layer usually is added at the transducer’s front
face
face [45–49].
[45–49]. For
For the
the transducers
transducers operating
operating with
with anan air
air load,
load, the
the best
best properties
properties of
of aa match-
match-
ing
ing layer are observed from the materials of polyether sulfone and nylon membranes [47].
layer are observed from the materials of polyether sulfone and nylon membranes. [47].
A matching
A matching layer
layer based
based on
on aa combination
combination of of aa porous
porous material
material with
with aa low-density
low-density rubber
rubber
material canachieve
material can achievean animprovement
improvement in in received
received signal
signal amplitude
amplitude of 30ofdB
30when
dB when com-
compared
paredthe
with with the unmatched
unmatched caseIn[48].
case [48]. In addition,
addition, transducers
transducers basedbased on ferroelectrets
on ferroelectrets have have
been
been reported
reported with reduced
with reduced impedance
impedance mismatchmismatch with
with the airthe air [50].
[50].

(a) (b) (c)

Figure 1. Typical cross-sectional structures of (a) piezoelectric ultrasonic transducers; (b) CMUTs
Figure 1. Typical cross-sectional structures of (a) piezoelectric ultrasonic transducers; (b) CMUTs and
and (c) d31-mode PMUTs [51].
(c) d31-mode PMUTs [51].

In
In contrast
contrast toto conventional
conventional ultrasonic
ultrasonic transducers,
transducers, the the MUTs
MUTs employ
employ aa flexural
flexural mem-mem-
brane for generating and receiving ultrasound waves, which
brane for generating and receiving ultrasound waves, which is conducive to better is conducive to impedance
better im-
pedance
matchingmatching
and higher andtransmitting
higher transmitting
efficiencyefficiency [51]. Therefore,
[51]. Therefore, MUTs are MUTs
mainly aredivided
mainly
divided into capacitive micromachined ultrasonic transducers
into capacitive micromachined ultrasonic transducers (CMUTs) [52] and piezoelectric mi- (CMUTs) [52] and piezoe-
lectric micromachined
cromachined ultrasound ultrasound
transducers transducers
(PMUTs)(PMUTs) [53]. CMUTs’[53]. CMUTs’
operation operation
is basedison based
the
on the flexural vibrations caused by a field-induced electrostatic
flexural vibrations caused by a field-induced electrostatic attraction between the suspended attraction between the
suspended membrane and the substrate (Figure 1b), whilst
membrane and the substrate (Figure 1b), whilst PMUTs is based on flexural vibrations PMUTs is based on flexural
vibrations
caused by d31- causedor by d31- or d33-mode
d33-mode excitation of excitation of a piezoelectric
a piezoelectric membranemembrane
(Figure 1c). (Figure 1c).
CMUTs
CMUTs have have the advantage of high bandwidth (often over 100%) [51], which is more
conducive to the modulation of the ranging ranging system.
system. However,
However, the output pressure of
CMUTs depends
CMUTs depends on on the excitation voltage
excitation voltage and the inverse of the capacitor gap [54]. Hence, Hence,
the CMUTs usually have submicrometer gaps [55] which leads to not only a complicated
fabrication process but also a small linear linear vibration
vibration amplitude.
amplitude. To To overcome
overcome this this problem,
problem,
high bias voltages of hundreds of volts can be used [56], which, however, would lead to
further increase in the system complexity and higher power power consumption.
consumption.
Compared with with CMUTs,
CMUTs,PMUTs PMUTsdo donot notrequire
require DC DC bias voltage
bias voltage [57,58],
[57,58],andandthe lin-
the
linear
ear displacement
displacement range
range is a is a function
function of theofmembrane
the membrane thicknessthickness
[59]. For[59].
PMUTs,For PMUTs,
the ac-
the active
tive piezoelectric
piezoelectric layer,
layer, e.g. leade.g., lead zirconate
zirconate titanate titanate
(PZT) and (PZT) and aluminum
aluminum nitride (AlN),nitrideis
(AlN),
depositedis deposited with nanofabrication
with nanofabrication techniques, techniques, e.g., plasma-enhanced
e.g. plasma-enhanced chemicalchemical
vapor deposi-vapor
deposition
tion (PECVD), (PECVD), sol-gel process,
sol-gel process, and sputtering,
and sputtering, on passive onelastic
passive elastic
layers suchlayers such
as Si, Si xNy as
or
Si, Si
SiO 2. x Ny resonant
The or SiO2 . frequency
The resonant frequency
of the PMUTs does of thenot PMUTs
directly does not directly
depend depend on
on the thickness of
the piezoelectric
thickness of the piezoelectric layer. Instead, the flexural mode
layer. Instead, the flexural mode resonant frequencies are closely relatedresonant frequencies
arethe
to closely
shape, related to the shape,
dimensions, boundary dimensions,
conditions, boundary
intrinsic conditions, intrinsic stress
stress and mechanical and
stiffness
mechanical stiffness of membranes [53]. In practical applications,
of membranes [53]. In practical applications, PMUTs can generate sufficient ultrasonic PMUTs can generate
sufficient
power ultrasonic
from a sub-mW power from adrive
electrical sub-mW signalelectrical
for target drive
rangessignal
up for
to atarget ranges[60]
few meters up and
to a
few meters
can meet the [60] and can meetofthe
requirements requirements
output pressureofinoutput pressure
the range. in the range.
Therefore, PMUTs Therefore,
are ex-
PMUTs are expected to be a better solution for rangefinders [61] due
pected to be a better solution for rangefinders [61] due to their superior power efficiency to their superior power
efficiency even though that is less than conventional bulk transducers.
even though that is less than conventional bulk transducers. As an example, a power dis- As an example,
a power of
sipation dissipation
400 μ W atof30400 fpsµW for at
a 1m30 fps for a 1 range
maximum m maximumhas been range has been
achieved withachieved
an AlN-
with an AlN-based ultrasonic array transducer [62]. The comparisons of bulk piezoelectric
based ultrasonic array transducer [62]. The comparisons of bulk piezoelectric transducers,
transducers, CMUTs and PMUTs are shown in Table 1.
CMUTs and PMUTs are shown in Table 1.
Micromachines 2022, 13, x FOR PEER REVIEW 4 of 35

Table 1. The characteristics of bulk piezoelectric transducers, CMUTs and PMUTs [45,51–54,56–59].
Micromachines 2022, 13, 520 4 of 33
Bulk piezoelectric CMUTs PMUTs
Mechanical machining, e.g., Wafer bonding and microm- Micromachining and wafer
Fabrication methods
dicing,
Table 1. Thelapping and casting
characteristics achining CMUTs andtransfer
of bulk piezoelectric transducers, diaphragm formation
PMUTs [45,51–54,56–59].
Matching layer Required No matching layer No matching layer
2 Bulk Piezoelectric CMUTs PMUTs
k eff * and bandwidth Low High Low
Mechanical machining, e.g., Wafer bonding and Micromachining and wafer
Fabrication
CMOS methods
compatible and flip- dicing, lapping and casting micromachining transfer diaphragm formation
No Yes Yes
chip integration
Matching layer Required No matching layer No matching layer
DC
k 2 bias requirement
* and bandwidth Low
No High
Yes Low
No
ef f
Millimeter level size of trans- Millimeter level size of trans-
CMOS compatible and
No to millimeter
Centimeter Yes
ducer arrays Yes
ducer arrays
flip-chip integration
General Size
level size Hundred microns level size Hundred microns level size
DC bias requirement No Yes No
of a single CMUT of a single PMUT
Millimeter level size of Millimeter level size of
2

General Size
*
eff
: electromechanical
kCentimeter to millimeter coupling coefficient
transducer arrays transducer arrays
level size Hundred microns level size of Hundred microns level size of
a single CMUT a single PMUT
2.2. Transducer Characteristics
* k2 e f f : electromechanical coupling coefficient.
The following subsections describe some characteristics of transducers related to the
ultrasonic
2.2. ranging
Transducer system’s performance.
Characteristics
The following subsections describe some characteristics of transducers related to the
2.2.1. Frequency
ultrasonic ranging Characteristics
system’s performance.
The operating frequency of the ultrasonic ranging transducer is often chosen to be
2.2.1. Frequency Characteristics
near the series resonant frequency which has minimal impedance. The frequency band-
Theof
width operating
PMUTsfrequency
used for of the ultrasonic
ranging ranging
in the air transducer
is generally is oftene.g.,
narrow, chosen
25%toinbe[63],
near com-
the series resonant frequency which has minimal impedance. The frequency
pared with CMUTs, >100% reported in [61]. Current ranging systems generally use a cer- bandwidth
of PMUTs used for ranging in the air is generally narrow, e.g., 25% in [63], compared
tain frequency of the transducer rather than frequency modulation. The double resonant
with CMUTs, >100% reported in [61]. Current ranging systems generally use a certain
frequency transducer [64] is also under study, which can obtain a broadened frequency
frequency of the transducer rather than frequency modulation. The double resonant
response. Broader bandwidth allows the adoption of the frequency modulation operation
frequency transducer [64] is also under study, which can obtain a broadened frequency
and is conducive
response. to the improvement
Broader bandwidth of ranging
allows the adoption technology.
of the frequency In addition,operation
modulation the operating
and is conducive to the improvement of ranging technology. In addition, the operating the
frequency determines ranging accuracy. In general, the higher the working frequency,
higher the
frequency ranging accuracy.
determines However,
ranging accuracy. In this will the
general, be at the expense
higher of the
the working ranging range
frequency, the as
the attenuation
higher the rangingofaccuracy.
ultrasound in the propagation
However, this will be atmedium willofincrease
the expense proportionally
the ranging range as to
theattenuation
the frequency of[65]. Therefore,
ultrasound the propagation
in the choice of operating
mediumfrequency
will increaseshould be matched
proportionally to with
specific
the measurement
frequency requirements.
[65]. Therefore, the choice of operating frequency should be matched with
specific measurement requirements.
2.2.2. Impedance
2.2.2. Impedance
Impedance analysis of transducers can evaluate the dynamic characteristics of trans-
Impedance analysis of transducers can evaluate the dynamic characteristics of trans-
ducers [66]. At present, most ultrasonic transducers are made of polycrystalline piezoe-
ducers [66]. At present, most ultrasonic transducers are made of polycrystalline piezo-
lectric ceramic materials. High-frequency electrical signals, > 20 kHz, are applied to pie-
electric ceramic materials. High-frequency electrical signals, >20 kHz, are applied to
zoelectric materials
piezoelectric materialsand
andconverted into ultrasonic
converted into ultrasonicsignals.
signals.The
Theequivalent
equivalent circuit
circuit model of
model
ofthe
thepiezoelectric
piezoelectricultrasonic
ultrasonic transducer
transducer isisshown
shownininFigure
Figure2.2.

Figure 2. Equivalent circuit model of transducer.

Figure 2. Equivalent circuit model of transducer.
Micromachines 2022, 13, 520 5 of 33

Where R0 is the parallel resistance of medium loss, C0 is the static capacitance, mea-
sured far below the resonant frequency, Cd is dynamic capacitance, Ld is dynamic induc-
tance, R L is the load resistance. In general, R0  Rd . In order to simplify the model, R0 can
be ignored during analysis. In addition, R L is short-circuited during analysis. Then, the
equivalent impedance of the piezoelectric transducer can be expressed as Equation (1) [67]
( Rd + jωLd + 1/jωCd ) · 1/jωC0
Z= (1)
Rd + jωLd + 1/jωCd + 1/jωC0

To maximize the power transmission efficiency of the ranging system,

√ make the
dynamic branch in resonance, and the resonant frequency ω = ωs = 1/ Ld Cd . Then
the impedance of the transducer can be shown as Equation (2), as a combination of both
resistance and reactance.
Rd ωs C0 R2d
Z = R + jX = − j (2)
1 + (ωs C0 Rd )2 1 + (ωs C0 Rd )2
Matching circuits are commonly included in the ultrasonic system, aiming to make
the transducer closer to pure resistance in order to achieve higher power output and
better efficiency [68,69]. An improvement of 300% in the amplitude of the received signal
was reported with and without the matching circuit [68]. This improvement leads to an
improved signal to noise ratio (SNR) and, therefore, better accuracy in measurements which
is desirable in the ultrasonic ranging system.

2.2.3. Electromechanical Coupling Coefficient

The electromechanical coupling coefficient k2 e f f is a parameter of ultrasound trans-
ducers to reflect energy conversion efficiency. A higher k2 e f f is desired to have more energy
to be converted between electrical and mechanical energy within the transducer [4]. The
k2 e f f can be calculated using Equation (3) [70]
 2
fr
k2 e f f = 1 − (3)
fa
where f r is the resonance frequency and f a is the anti-resonance frequency.
In general, the k2 e f f of PMUTs used is less than 5%, and the method called ClarinetTM
to increase the coefficient is also being studied [71]. Due to the high attenuation in ul-
trasound waves along the travelled path in the air, the ultrasonic ranging system would
benefit from a transducer with a higher coupling coefficient to improve its signal sensitivity
and measurement accuracy.
Apart from optimizing for a higher coupling coefficient of the transducer, other ap-
proaches are used to improve the SNR of the received signals as well for the ranging system.
In [72], the authors reported that the signal amplitude of the bent transducer was enhanced
10 times bigger by designing its active diameter. The signal quality can also be improved
by optimizing the excitation voltage of the transmitting transducer [73], exciting the trans-
mitter transducer with a direct modulation system [74] and adopting the broadband tuning
of the receiving circuit [75].

2.2.4. Directivity
The directivity of transducer and transducer array is a characteristic that the amplitude
of transmitting response or receiving response changes with azimuth angle. It determines
the range of the azimuth angle of the measurable range in the space. Depending on their
applications, the requirements for directional characteristics of transducers are different. For
a transmitting transducer, the sharpness of its directional characteristic curve determines
the concentration of its transmitted energy, and for a receiving transducer, it determines
the range of azimuth to explore space. In the ultrasonic ranging system, the transducers
can be used as either the transmitter, the receiver or both depending on the configurations.
 ferent. For a transmitting transducer, the sharpness of its directional characterist
determines the concentration of its transmitted energy, and for a receiving trans
determines the range of azimuth to explore space. In the ultrasonic ranging sys
Micromachines 2022, 13, 520
transducers can be used as either the transmitter, the receiver or both6 ofdependin 33
configurations.
The sound field established by the transducer in the medium is related to th
andThe
sizesound
of the transducer,
field establishedthe vibration
by the mode,
transducer in thethe working
medium parameters
is related such as fr
to the shape
and
andsize
theoftypes
the transducer, the vibration
of the medium [76].mode, the working parameters such as frequency
and theThe
types of the medium [76].
directivity response diagram can be obtained by drawing the direct
The directivity response diagram can be obtained by drawing the directivity response
sponse diagram in decibel (dB) compared to its maximum in the sound field.
diagram in decibel (dB) compared to its maximum in the sound field. Figure 3 presents the
presents diagram
directivity the directivity diagram
of a transmitting of a transmitting
ultrasonic transducer. ultrasonic transducer.

Figure
Figure Typical
3. 3. directivity
Typical diagram
directivity of a transmitting
diagram ultrasonic transducer.
of a transmitting ultrasonic transducer.
As shown in Figure 3, there are the main lobe and sidelobes in the acoustic field [77],
Asmain
with the shownlobe in Figure
mainly could3, there
be usedareforthe mainInlobe
ranging. and sidelobes
applications where the in ranging
the acoustic fi
with theis main
direction fixed, thelobe mainlyiscould
transducer be used
generally for ranging.
configured In applications
in an orientation to align withwhere
its the
main lobe’s direction.
direction is fixed, the transducer is generally configured in an orientation to align
When the width of the main lobe becomes narrower, the energy of the beam is more
main lobe’s direction.
concentrated and its directivity is higher. The parameter, beam spread angle θ, can be used
Whenthe
to represent the width of
directivity theultrasound
of the main lobetransducer.
becomesBeam narrower, the energy
spread angle measuresofthe
the beam
concentrated
width of the beamand its directivity
in degrees, from side toisside
higher. The lobe
of the main parameter,
where thebeam spread
ultrasonic energy angle θ
intensity drops to −
used to represent the directivity of the ultrasound transducer. Beam sprea
3 dB. In general, the ranging system takes the corresponding space
within the beam
measures the spread
widthangleof the of the
beamtransducer as the detectable
in degrees, from siderange.
to side of the main lobe w
When the beam spread angle of a single-element ultrasonic transducer cannot meet
ultrasonic energy intensity drops to −3dB. In general, the ranging system takes th
the range requirement, transducer array and beam forming techniques can be adapted to
sponding
change space within
the directivity the beam
of transducers [78]spread angle
and inhibit theof the transducer
sidelobe as the
level [79], and detectable
can also
improveWhen the beam
the ultrasonic spread
emission angle and
intensity of aincrease
single-element
the effectiveultrasonic
measurabletransducer
distance. cann
the range requirement, transducer array and beam forming techniques can be ad
2.3. Ultrasonic Propagation Characteristics
change the directivity of transducers [78] and inhibit the sidelobe level [79], and
When the ultrasonic wave propagates in the medium, its sound pressure intensity
improve the ultrasonic
gradually decreases, emission
known as intensity
attenuating. and increase
Attenuation the
is a result of effective
several measurable
factors from d
interaction with medium and interfaces, including scattering, absorption, reflection and
2.3. Ultrasonic
diffraction. In the Propagation Characteristics
ultrasonic ranging system where the transducer operates in pulse-echo
mode, the transducer emits the ultrasonic wave and is reflected by the target obstacle, the
When the ultrasonic wave propagates in the medium, its sound pressure i
ultrasonic overall loss formula can be expressed as shown in Equation (4), including the
gradually decreases,from
attenuation contribution known as attenuating.
the propagation Attenuation is a result of several fact
path [76]
interaction with medium and interfaces, Pr
including scattering, absorption, reflec
a −2αr
diffraction. In the ultrasonicG ranging=
Pt
= Gsystem
ac
4r
10 where the transducer operates (4) in pu
mode, the transducer emits the ultrasonic wave and is reflected by the target obst
where G is ultrasonic attenuation coefficient, Pr is the pressure of the ultrasonic wave re-
ultrasonic
ceived overall loss
by the transducer, Pt isformula canofbe
the pressure theexpressed as shown
ultrasonic wave in the
emitted by Equation (4), inclu
transducer,
Gattenuation contribution
ac is the acoustic gain, depends from the
on the propagation
size of the target, apath [76] membrane radius, r
is effective
is the range to the target, α is attenuation coefficient which increases with the frequency
P
of sound waves, and also depends on humidity, temperature a -2α r ambient pressure. The
G = r =Gac 10and
ultrasound attenuation in the air can be as high as P7 dB/m 4 atr 215 kHz at room temperature
t
and 60% relative humidity in the air [63]. In general, the ultrasonic ranging system tends to
use transducers that operate at frequency range < 300 kHz, and typically at frequencies
around 40 kHz, in order to obtain a balance between the measurement distance of interest,
the measurement resolution and the signal quality, e.g., SNR.
 system measures the ultrasonic wave transmitted and received by the transducer
verts it into distance measurement to complete the function of the ranging system

Micromachines 2022, 13, 520

3.1. Principle of Ultrasonic Ranging 7 of 33
The ultrasonic transducer transmits ultrasonic waves and receives echo
from the object by either the same transducer, pulse-echo mode or by another tra
3.asUltrasonic Ranging
pitch-catch mode. System
Whenandusing
Its Evaluation Parameters
the pulse-echo mode, the distance s between t
sonicThetransducer
ultrasonic transducer
and theisobject
the keyis
component of the ultrasonic
to be calculated ranging system.
by measuring The t betw
the time
system measures the ultrasonic wave transmitted and received by the transducer and
transmitted signal and received signal as shown in Equation (5)
converts it into distance measurement to complete the function of the ranging system.

3.1. Principle of Ultrasonic Ranging

s = ct 2
The ultrasonic
where transducer
c is the sound transmits
velocity ultrasonic
in the medium.waves and receives echo reflected from
the object by either the same transducer, pulse-echo mode or by another transducer as
Figure
pitch-catch 4 shows
mode. a pair
When using theofpulse-echo
the transducers
mode, theare configured
distance s betweeninthe
pitch-catch
ultrasonic config
the distance
transducer between
and the object is tothe plane where
be calculated ultrasonic
by measuring transducers
the time t between theare located and th
transmitted
object
signal d is
and received signal as shown in Equation (5)
s = ct/2 (5)
where c is the sound velocity in the medium. d= s - h 2 ( 2
( )
2
)
Figure 4 shows a pair of the transducers are configured in pitch-catch configuration,
the distance between the plane where ultrasonic transducers are located and the target
In practice, d is much greater than the distance between the two transduce
object d is r
therefore h can be treated asdnegligible. )2 means d=s and Equation (6)
That (6) can al

= s2 − (h/2
in this configuration.

Figure
Figure 4. Schematic
4. Schematic diagram
diagram of Ultrasonic
of Ultrasonic Ranging
Ranging with with
a pair of a pair of
transducers. transducers.

3.2.InComposition
practice, d is much greater than the distance between the two transducers h and
of Ultrasonic Ranging System
therefore h can be treated as negligible. That means d = s and Equation (6) can also apply
A typical ultrasonic ranging system would be as shown in Figure 5 and com
in this configuration.
transducers for transmitting and receiving, transmitting circuits, receiving circu
3.2. Composition of Ultrasonic Ranging System
croprocessor, a temperature compensation module and a display module [80–83
A typical ultrasonic ranging system would be as shown in Figure 5 and composed
Micromachines 2022, 13, x FOR PEER REVIEW 8 of 35
of transducers for transmitting and receiving, transmitting circuits, receiving circuits, a
microprocessor, a temperature compensation module and a display module [80–83].

Figure5.5.An
Figure Anoverall
overalldiagram of of
diagram ultrasonic ranging
ultrasonic system.
ranging system.

Transducers include the transmitting transducer and the receiving transducer. The
transmitting circuit amplifies the programmed pulses to high voltage pulses that can drive
the transducer, while the preamplifier of the receiving circuit amplifies the received sig-
Micromachines 2022, 13, 520 Figure 5. An overall diagram of ultrasonic ranging system. 8 of 33

Transducers include the transmitting transducer and the receiving transducer. The
transmitting circuit amplifies the programmed pulses to high voltage pulses that can drive
Transducers include the transmitting transducer and the receiving transducer. The
the transducer, while the preamplifier of the receiving circuit amplifies the received sig-
transmitting circuit amplifies the programmed pulses to high voltage pulses that can drive
nals because they are generally weak [84]. Then, the signal passes through the bandpass
the transducer, while the preamplifier of the receiving circuit amplifies the received signals
filter to remove interfering noise signals and improve the SNR of received signals. To
because they are generally weak [84]. Then, the signal passes through the bandpass filter
avoid the high requirements of the hardware, the bandpass sampling theory can be ap-
to remove interfering noise signals and improve the SNR of received signals. To avoid
plied to digitalize the echo signal [85].
the high requirements of the hardware, the bandpass sampling theory can be applied to
Microprocessor generally adopts a micro controller unit (MCU) which is high-per-
digitalize the echo signal [85].
formance and low-power [86–88]. The microprocessor controls the programming of the
Microprocessor generally adopts a micro controller unit (MCU) which is high-
transmitting signal. At the same time, it also processes the received signal to obtain the
performance and low-power [86–88]. The microprocessor controls the programming of
information of interests and transmits the measurement results to the display module.
the transmitting signal. At the same time, it also processes the received signal to ob-
Meanwhile, the power supply module is the basis of the normal work of the ranging sys-
tain the information of interests and transmits the measurement results to the display
tem.
module. Meanwhile, the power supply module is the basis of the normal work of the
As for
ranging the processing of received signals by the microprocessor, the key is to obtain
system.
the starting
As for the T1 of echoofsignals,
timeprocessing thesignals
received ideal and noisy
by the ultrasonic echothe
microprocessor, signals
key isare
to shown
obtain
theFigure
in starting
6. time T1 of echo
In addition, thesignals, the ideal
amplitude and
of the noisy ultrasonic
received echowith
signal varies signals
the are shown
measured
in Figure 6. In addition, the amplitude of the received signal varies with the measured
distance [89]. Different methods can be used to obtain T1 , which will be elaborated in
distance [89]. Different methods can be used to obtain T1 , which will be elaborated in detail
detail in Section 4.
in Section 4.

Figure 6.
Figure Ideal and
6. Ideal and noisy
noisy ultrasonic
ultrasonic echo
echo signals
signals [90].
[90].

The temperature compensation module compensates for the influence of temperature

The temperature compensation module compensates for ◦the influence of tempera-
on the propagation velocity of ultrasonic waves [80]. For every 1 C increase in temperature,
ture on the propagation velocity of ultrasonic waves [80]. For every 1 °C increase in tem-
the speed of sound increases by about 0.607 m/s [91]. The circuit is equipped with a
perature, the speed of sound increases by about 0.607 m/s [91]. The circuit is equipped
temperature sensor and connected with the microprocessor to realize the acquisition of
with a temperature sensor and connected with the microprocessor to realize the acquisi-
real-time ultrasonic velocity.
tion of real-time ultrasonic velocity.
Display module: For the measurement data obtained by the microprocessor, they
Display module: For the measurement data obtained by the microprocessor, they can
can be directly connected to a liquid crystal display (LCD) for real-time display [92]
be
or directly connected
transmitted to microcontrollers
to other a liquid crystal display
through (LCD) for real-time
wireless display [92]
communication or trans-
to avoid the
mitted to other microcontrollers through wireless communication
inconvenience of reading the measurement data directly [93]. to avoid the inconven-
ienceSeveral
of reading the measurement
research data directly
groups had developed [93].ultrasonic ranging system [80,82,94–96],
their own
and we extract and summarize the program flow of the ultrasonic ranging system and
show in Figure 7. The value of T is set in direct proportion to the furthest measurement
distance, that is, the single measurement period is inversely proportional to the value of T.

3.3. Evaluation Parameters of Ranging System

3.3.1. Measurement Range of Distances and Angles
The sector in a two-dimensional space or the conical shape in a three-dimensional space
can be represented as the effective range of ultrasonic ranging as shown in Figure 8 [97].
In the figure, θ is the beam spread angle which relates to the directivity of the transducer;
Lmin and Lmax are the nearest and the farthest distance that can be measured, respectively.
Lmin is determined by the dead zone of the transducer, which is related to the near field of
the ultrasound propagation. The dead zone is also related to the length of the excitation
pulses [98]. Although the longer the excitation signal is, the more energy delivered and
therefore the better signal quality at reception, its longer ringing down time will lead to
a larger blind area. Lmax is determined by the propagation attenuation characteristics of
 Micromachines 2022, 13, x FOR PEER REVIEW 9 of 35

Micromachines 2022, 13, 520 Several research groups had developed their own ultrasonic ranging system 9 of 33
[80,82,94–96], and we extract and summarize the program flow of the ultrasonic ranging
system and show in Figure 7. The value of T is set in direct proportion to the furthest
measurement
the distance,
ultrasonic wave that is, in
described theSection
single measurement period is inversely
2.2.4. The measurement range ofproportional
an ultrasonic to
the value
ranging of T. is 2 cm–5 m in general to ensure adequate echo signal at reception [99].
sensor

Micromachines 2022, 13, x FOR PEER REVIEW 10 of 35

Figure7.7.AAprogram
Figure programflow
flowofofan
anultrasonic
ultrasonicranging
rangingsystem.
system.

3.3. Evaluation Parameters of Ranging System

3.3.1. Measurement Range of Distances and Angles
The sector in a two-dimensional space or the conical shape in a three-dimensional
space can be represented as the effective range of ultrasonic ranging as shown in Figure 8
[97]. In the figure, θ is the beam spread angle which relates to the directivity of the trans-
ducer; Lmin and Lmax are the nearest and the farthest distance that can be measured, re-
spectively. Lmin is determined by the dead zone of the transducer, which is related to the
near field of the ultrasound propagation. The dead zone is also related to the length of the
excitation pulses [98]. Although the longer the excitation signal is, the more energy deliv-
ered and therefore the better signal quality at reception, its longer ringing down time will
Figure 8.
Figure 8. Schematic diagram of measuring range.
lead to a larger blind area. Lmax is determined by the propagation attenuation character-
isticsIn
3.3.2. oforder
the ultrasonic
Measurementto reduce wave described
the dead
Accuracy zone in
of Ultrasonic Section
effect 2.2.4.
on the
Ranging The measurement
measurable
System range, the range of an ul-
pitch-catch
trasonic
mode canranging
The be adopted
measurementsensor is 2separate
with cm–5of
accuracy mtransmitters
in general
distance toand
ensure
s depends adequate
receivers
on the[100]. echo signal
In this
accuracy at reception
of configuration,
both transmit
[99].
the pulse’s ringing down will not mask the echo received, therefore,
time t and sound velocity c. Among them, the acquisition of transmit time is more reducing the dead
critical,
zone In
L order
. to reduce
Another
min below.
as detailed the dead
approach iszone
to effect
optimize on the
the measurable
transmit range,
pulse and the pitch-catch
receiving mode
filter by
can First,
be adopted
customizing when the direct transit time method is used, the accuracy of transmit timethe
the with separate
transmit pulses transmitters
of chirp and
80–65 receivers
kHz with [100].
Cauer In this
high-pass configuration,
filter to reduce is
the receive
pulse’s pulse
ringing duration
down will and
not combining
mask the with
echo partial
received, ringing
therefore,
determined by the accuracy of identifying the received time of ultrasonic waves. removal
reducing by using
the deadringing
zone
look-up-tables
. Anotherto
Lmin Secondly, reduce the to
approach minimum
optimizedetectable range down to 3 cm [101]. In addition, to
phases of is
the propagated the transmit
ultrasound pulse
waves and
canreceiving filter
also be measured by custom-
to ex-
increase Lmax and angle range, a transducer array can be adopted [102].
tract
izingtiming information
the transmit pulsesforof distance
chirp 80–65 calculation.
kHz with When
Cauer thehigh-pass
phase methodfilter to and the fre-
reduce the
quency
receive modulated
pulse continuous
duration and waves
combining (FMCWs)
with
3.3.2. Measurement Accuracy of Ultrasonic Ranging System partialmethod
ringing are adopted,
removal by the
using transit
ringing time is
look-
essentially obtained indirectly. The accuracy of distance is also related
up-tables to reduce the minimum detectable range down to 3 cm [101]. In addition, to to the range reso-
The measurement accuracy of distance s depends on the accuracy of both transmit
lution or axial resolution [103]. Ita transducer
is the minimal range difference (the axial distance be-
time t and Lsound
increase max and angle range,
velocity c. Among them, the array can be
acquisition adopted
of transmit [102].
time is more critical,
tween the measured points B and C in Figure 8) needed to distinguish the movement of a
as detailed below.
target along one bearing. Therefore, the range resolution is the highest accuracy that can
First, when the direct transit time method is used, the accuracy of transmit time is
be achieved by ultrasonic ranging systems.
determined by the accuracy of identifying the received time of ultrasonic waves.
When the phase method is adopted, the distance corresponds to the phase value, and
its range resolution can be expressed as
c θ 0
σ1= (7)
f  360°
 The measurement accuracy of distance s depends on the accuracy of both transmit
time t and sound velocity c. Among them, the acquisition of transmit time is more critical,
as detailed below.
First, when the direct transit time method is used, the accuracy of transmit time is
Micromachines 2022, 13, 520 determined by the accuracy of identifying the received time of ultrasonic waves. 10 of 33
Secondly, phases of the propagated ultrasound waves can also be measured to ex-
tract timing information for distance calculation. When the phase method and the fre-
quency modulated continuous waves (FMCWs) method are adopted, the transit time is
Secondly, phases of the propagated ultrasound waves can also be measured to extract
essentially obtained indirectly. The accuracy of distance is also related to the range reso-
timing information for distance calculation. When the phase method and the frequency
lution or axial resolution [103]. It is the minimal range difference (the axial distance be-
modulated continuous waves (FMCWs) method are adopted, the transit time is essentially
tween the measured points B and C in Figure 8) needed to distinguish the movement of a
obtained indirectly. The accuracy of distance is also related to the range resolution or axial
target along one bearing. Therefore, the range resolution is the highest accuracy that can
resolution [103]. It is the minimal range difference (the axial distance between the measured
be achieved
points B and by ultrasonic
C in Figure 8)ranging
neededsystems.
to distinguish the movement of a target along one
bearing. Therefore, the range resolution isthe
When the phase method is adopted, thedistance
highestcorresponds
accuracy thatto can
the phase value, and
be achieved by
its range resolution
ultrasonic can be expressed as
ranging systems.
c θ 0 corresponds to the phase value, and
When the phase method is adopted, the distance
its range resolution can be expressed as σ 1 = f  360° (7)
c · θ0
Where σ 1 is the distance resolution σ1 =of the ◦phase method, θ is the phase resolu-
f ·360 0
(7)
tion, c is the sound velocity and f is the transducer frequency.
where σ1 is the distance resolution of the phase method, θ0 is the phase resolution, c is the
soundFMCWs, also fcalled
velocity and is thechirps, are commonly
transducer frequency. used to enlarge the range of interest and
improve measurement resolution, first
FMCWs, also called chirps, are commonly in the radar [104],
used to and the
enlarge laterrange
in ultrasonic nonde-
of interest and
structive testing applications [105,106], ultrasonic medical imaging [107]
improve measurement resolution, first in the radar [104], and later in ultrasonic nonde- and ultrasonic
ranging testing
structive [108]. The transmitted
applications and received
[105,106], signals
ultrasonic can beimaging
medical expressed as shown
[107] in Figure
and ultrasonic
9.
ranging [108]. The transmitted and received signals can be expressed as shown in Figure 9.

Figure9.9.The
Figure Thetransmitted
transmittedand
andreceived
receivedchirp
chirpsignals
signals(T(sTsisisthe
thescanning
scanningperiod
periodofofone
onechirp
chirpwhich
whichisis
the difference between t1 and t2 , Bc is the scanning bandwidth which is the difference between f 1
the difference between t1 and t2 , Bc is the scanning bandwidth which is the difference between
and f 2 , τ is time delay of received signals compared to transmitted signals and f b is the difference
frequency between the transmitted signals and received signals).

The range resolution of the FMCWs method is

c
σ2 = (8)
2Bc

where σ2 is the distance resolution of the FMCWs method, c is the sound velocity and Bc is
the scanning bandwidth.
The range resolution of the FMCWs method depends on the scanning bandwidth.
Therefore, high bandwidth transducers are required for high range resolution.
It is important to note that range resolution is the measurement accuracy in an ideal
world. The accuracy in practice will be affected by many other aspects such as the hardware
and the echo signal processing method.

3.3.3. Measurement Rate

The measurement rate is the reciprocal of the time required for a single measurement.
When a single transducer is used in the ranging system to perform in pulse-echo mode,
the transducer would usually be set up in a way that it will not fire the next pulse until a
reflected echo is received.
The maximum measurement rate of the pulse-echo method is as follows:
1 c
Maximum measurement rate = = (9)
tto f 2· d
 When a single transducer is used in the ranging system to perform in pulse-echo mode,
the transducer would usually be set up in a way that it will not fire the next pulse until a
reflected echo is received.
The maximum measurement rate of the pulse-echo method is as follows:

Micromachines 2022, 13, 520 1 c 11 of 33

Maximum measurement rate= = (9)
ttof 2d

The measurement rate decreases inversely with the distance, and the relationship be-
The measurement rate decreases inversely with the distance, and the relationship
tween measurement rate and measurement distance is shown in Figure 10. For a target of
between measurement rate and measurement distance is shown in Figure 10. For a target
5 m, the measurement rate is only 34 Hz, which cannot meet the requirements of meas-
of 5 m, the measurement rate is only 34 Hz, which cannot meet the requirements of
urement speed on all occasions such as the dynamic measurement of the blade tip distance
measurement speed on all occasions such as the dynamic measurement of the blade tip
between between
distance the upper blades
the upperand the lower
blades blades.
and the lower blades.

10. Diagram of the measurement

Figure 10. measurement rate
rate and
and distance.
distance.

In summary, the performance of an ultrasonic ranging system depends on the mea-

In summary, the performance of an ultrasonic ranging system depends on the meas-
surement range, measurement accuracy and measurement rate, and should be designed
urement range, measurement accuracy and measurement rate, and should be designed
according to the actual measurement requirements.
according to the actual measurement requirements.
4. Ultrasonic Ranging Methods and Signal Processing
4. Ultrasonic Ranging Methods and Signal Processing
Ultrasonic ranging methods are divided into the time of flight (ToF) method, two fre-
quencyUltrasonic ranging
continuous wave methods are divided
(TFCW) method into the time ofcontinuous
and multi-frequency flight (ToF) method,
waves two
(MFCW)
frequency continuous wave (TFCW) method and multi-frequency continuous
method, binary frequency shift keying (BFSK) method, amplitude modulation method and waves
(MFCW)
signal method,
coding methodbinary frequency
according shift
to the keying (BFSK)
measured method,
parameters amplitude
and the modulation
type of transmitted
method
wave. Theand signal coding
following method
subsections according
describe to methods
these the measured parameters and the type of
in detail.
transmitted wave. The following subsections describe these methods in detail.
4.1. Time of Flight (ToF) Method
ToF method is the most common method. The transmitter sends simple single-
frequency sequence pulses, and the receiver processes the echo signal to get the time
of flight of ultrasonic waves, and then the distance value can be obtained by combining
it with the sound velocity. The key for the ToF method is to obtain the time information
accurately and that can be achieved through the amplitude threshold, envelope fitting and
correlation method.

4.1.1. Amplitude Threshold Method (ATM)

ATM is the most commonly used due to its simplicity and easy implementation, fast
calculation speed and low hardware price [26,109].
Examples of driving and received signals are shown in Figure 11. The initial time of
the driving signal is denoted as T0 , and the amplitude detection of the received signal is
performed according to the amplitude threshold τ. In general, the amplitude threshold is
set at 3 to 5 times the noise level due to the presence of noise [26]. When the amplitude
reaches the threshold τ, the time is denoted as T1 , and the flight time is t = T1 − T0 . In
practice, the receiver may have an electrical signal fluctuation before the ultrasonic echo
arrives due to the crosstalk, as shown in Figure 11, which can be filtered out by hardware
isolation or filtering algorithms.
The performance achieved by the ranging system using the ATM is summarized in
Table 2.
 is set at 3 to 5 times the noise level due to the presence of noise [26]. When the am
reaches the threshold τ , the time is denoted as T1 , and the flight time is t = T1
practice, the receiver may have an electrical signal fluctuation before the ultrason
Micromachines 2022, 13, 520
arrives due to the crosstalk, as shown in Figure 11, which can be filtered out by ha
12 of 33
isolation or filtering algorithms.

Figure Driving
Figure11.11. and received
Driving signals with
and received crosstalk
signals with(the brown line
crosstalk indicates
(the brown theline
waveform of thethe wav
indicates
system clock, the blue line indicates the twelve electrical pulse excitation signals and
the system clock, the blue line indicates the twelve electrical pulse excitation signals the purple line and th
indicates the waveform of the receiver) [110].
line indicates the waveform of the receiver) [110].
Table 2. Performance comparison of several ATM-based ranging systems.
The performance achieved by the ranging system using the ATM is summa
Reference Range Accuracy Transducer Type
Table 2.
[63] 1300 mm 1.3 mm PMUT with 215 kHz
[110]
Table 500 mmcomparison
2. Performance 0.63 mm PMUT withranging
of several ATM-based 97 kHz and 96 kHz
systems.
[111] 1000 mm 4 mm PMUT with 77.34 kHz
Reference
[109]
Range
5000 mm 4 mm
Accuracy Transducer type
Conventional bulk transducers with 35 kHz
[63]
[112] 1300
100 mmmm 0.5 mm1.3 mm PMUT
Conventional bulk transducers with
with 215 kHz
40 kHz
[110] 500 mm 0.63 mm PMUT with 97 kHz and 96 kH
[111]
Compared with 1000 mm
conventional bulk4transducers,
mm the maximumPMUT range ofwith 77.34iskHz
the PMUT
limited, generally
[109] about 1
5000 mm m. As can be seen
4 mm from Table 2, the shorter the measurement
Conventional bulk transducers with
range is, the higher the accuracy can be achieved due to the better SNR of the echo signal.
[112] 100 mmfrom the received
In practice T detected 0.5 mmsignal presents
Conventional bulk
an error and transducers
does not repre- with
1
sent the exact arrival time of the echo signal. This is because of the presence of noise and
the setting of the amplitude threshold. This error cannot be compensated as a systematic
error because
Compared it is different when the propagation
with conventional distance is different.
bulk transducers, In addition,
the maximum thisof the P
range
method is susceptible to noise, and the system will process it as an echo signal when the
limited, generally about 1 m. As can be seen from Table 2, the shorter the measu
noise of a high level occurs occasionally. To solve this problem, ref. [113] proposed the
range threshold
double is, the higher
method, the accuracy
also known ascan
the be achieved
sliding windowdue to the
method. A better
windowSNR of the echo
of width
N shifts along the echo signal one sample at a time. At each window position, calculate
the number of samples exceeding the set threshold τ. If this number exceeds the second
threshold m, then estimate ToF. The advantage of this method is its robustness to noise
peaks because the detection of the target is based on m samples rather than a single sample
with a single threshold. A measurement accuracy of 0.69 mm in the range of 100–600 mm
is reported in [113].

4.1.2. Envelope Fitting Method

In order to solve the problem of the ATM method in detecting the initial time of the
echo signal, the curve fitting method can be used to fit the envelope of the echo signal, to
find the starting point of the echo signal and generate the unbiased estimator of ToF.
The ideal model of the received signal analyzed in [114] is shown in Figure 12a, where
Vr is the received signal and Vrenv is the envelope of the received signal.
 4.1.2. Envelope Fitting Method
In order to solve the problem of the ATM method in detecting the initial time of the
echo signal, the curve fitting method can be used to fit the envelope of the echo signal, to
find the starting point of the echo signal and generate the unbiased estimator of ToF.
Micromachines 2022, 13, 520 The ideal model of the received signal analyzed in [114] is shown in Figure 12a, where
13 of 33
V r is the received signal and V renv is the envelope of the received signal.

(a) (b)
Figure 12. (a) An ideal model of the received signal, (b) An example of the fitting curve [114]
Figure 12. (a) An ideal model of the received signal, (b) An example of the fitting curve [114].
The envelope curve is simply fitted by the peak amplitude and time of each cycle, as
The envelope curve is simply fitted by the peak amplitude and time of each cycle,
shown
as shownin Figure 12b,12b,
in Figure or adopt the the
or adopt parabolic model
parabolic model[26].[26].
Take thethe
Take time corresponding
time corresponding to
the point where the amplitude of the envelope curve is 0 as the initial time
to the point where the amplitude of the envelope curve is 0 as the initial time of the of the received
signal.
received signal.
With
Withthe
theabove
abovemethod,
method,[114]
[114]reported
reportedthe
themeasurement
measurementaccuracy accuracycan canreach
reach0.7
0.7mm
mm
within
within the range of 3000 mm distance and 0.3 mm within the range of 1000 mm. It canItreach
the range of 3000 mm distance and 0.3 mm within the range of 1000 mm. can
reach
higherhigher
accuracyaccuracy
withinwithin the range
the same same compared
range compared
with ATM withmethod
ATM methodin Tablein2.Table 2. In
In practice,
practice, the received signal envelope has deviation due to the influence
the received signal envelope has deviation due to the influence of device performance, of device perfor-
mance,
noise andnoise andfactors.
other other factors. Therefore,
Therefore, the envelope
the envelope fitting method
fitting method is morein
is more limited limited in
actuality.
actuality.
4.1.3. Correlation Method
4.1.3. The
Correlation Method
correlation method is considered the optimal TOF estimation technique in gen-
eral The
[115].correlation
It performs method is considered the
a cross-correlation optimal TOF
calculation estimation
on the receivedtechnique in gen-
echo signal and
the [115].
eral transmitted
It performssignal, and then determine
a cross-correlation the flight
calculation on time according
the received echoto signal
the maximum
and the
value of thesignal,
transmitted cross-correlation
and then determinesignal. For
theaflight
giventime
sequence of transmitted
according and received
to the maximum value
signals y ( kT
of the cross-correlation
p S ) and y ( kT )
E signal.
S , where T is
For a given
S the sampling period, then the cross-correlation
sequence of transmitted and received signals
ysignal
( kT X
) and
C is y ( kT ) , where T is the sampling period, then the cross-correlation signal
p S E S +∞ S

X C is XC = ∑ y p (kTS )·yE (kTS + nTS ) (10)

−∞

The ToF can be determined according to the peak position of XC , and the peak lag is
proportional to ToF, thus obtaining the measured distance R [116,117].

R = (τmax · Ts − TOE)·c − Rcal (11)

where τmax is the lag of the maximum peak which is proportional to the time of arrival
(TOA), TOE is the time of emission of the ultrasonic signal, and Rcal is a calibration constant
including all the fixed delays of the system which is independent of the range. Using the
correlation method, the accuracy can reach 3.9 mm in the range of 30–450 mm with a PMUT
of 214 kHz in [40,118] and 1.2 mm within the range of 2300 mm with the chirp of 15 kHz
to 40 kHz [119]. In order to further improve the accuracy, the spline interpolation method
can be adopted. However, the processing time will be increased. The accuracy can reach
0.25 mm in the range of 200–1000 mm, and the operation time needs 0.3 s in [120].
In addition, the flight time can be obtained through a combination of methods in
order to improve the accuracy over a longer distance. There are studies that apply cross-
correlation to correct the distance error for one wavelength scale and then use a phase-shift
technique for subwavelength range refinement [121]. Reference [122] reported a 1 mm
ranging resolution for the distance up to 3000 mm and [90] has achieved 0.5 mm accuracy
for the distance up to 5000 mm.
In summary, ATM is simple and fast due to fewer processing requirements when
compared to the envelope fitting and correlation algorithm, and therefore can be applied to
the measurement of moving targets. The envelope fitting method has a processing speed of
Micromachines 2022, 13, 520 14 of 33

6 ms to 8 ms in [114] and is limited by the envelope model. While the processing speed of
correlation method is slower than ATM, the processing time can reach 0.7 ms of correlation
method while 0.07 ms of ATM with the same processing device [123]. What should be kept
in mind is that the measurement accuracy is not only related to the processing method but
also related to the performance of the devices and measurement environment. The method
of obtaining ToF should be selected according to the specific measurement requirements in
measurement accuracy, processing speed and the speed of the measurement target.

4.2. Two Frequency Continuous Wave (TFCW) and Multi-Frequency Continuous Waves (MFCW)
TFCW and MFCW methods obtain the time delay through phase difference measure-
ment [124] and provide higher accuracy in measurement at the expense of measurement
range when compared to the conventional ToF method.

4.2.1. Two Frequency Continuous Wave (TFCW)

To apply TFCW methods, the transmitter sends two excitation signals of the continuous
wave with two frequencies of f 1 and f 2 ( f 1 < f 2 ) respectively, and the phase shifts of the
two signals can be measured respectively when the waves reach the receiver. Then the
distance between transmitter and receiver can be expressed as
 
θ1
d = n1 + λ1 (12)
2π
 
θ
d = n1 + 1 λ2 (13)
2π
where λ1 and λ2 are wavelengths of transmitted signals at frequencies f 1 and f 2 respectively,
θ1 and θ2 are the phase shifts of two received signals compared to transmitted signals
respectively. Therefore, the phase shift can be expressed as

θ2 − θ1 = ∆θ − 2π (n2 − n1 ) (14)

where ∆θ is  
1 1
∆θ = 2πd − (15)
λ2 λ1
The integers n have only two possible values: n2 = n1 and n2 = n1 + 1. The difference
of phase shifts can be defined by the following Algorithm 1:

Algorithm 1
1: If n2 = n1 , θ2 − θ1 = ∆θ
2: If n2 = n1 + 1, θ2 − θ1 = ∆θ − 2π

Therefore, the measured distance d can be expressed as

∆θ c
d= · (16)
2π ∆ f
where ∆ f = f 2 − f 1 . Therefore, in a certain sound propagation medium, the measurement
range depends on the phase shift difference and frequency difference. The range can be
increased to the wavelength corresponding to ∆ f , and the accuracy is determined by the
phase resolution. The highest resolution of the phase depends on the clock frequency of
the hardware system; for example, when using a 40 MHz clock for phase counting of a
40 kHz signal, a maximum resolution of 0.1% is obtained [125]. In the actual measurement,
the distance resolution can reach 1% wavelength [100,126].
References [127–129] all adopted the TFCW method, and the performance achieved
by the measurement system is summarized in Table 3.
Micromachines 2022, 13, 520 15 of 33

Table 3. Performance comparison of several TFCW based ranging systems.

Reference Range Accuracy Signal Frequency

[127] 30 mm~100 mm 1.5 mm 39.85 kHz and 40.6 kHz
[128] 50 mm~200 mm 0.1362 mm 40 kHz and 40.82 kHz
[129] 10 mm~110 mm 2.5 mm 94.21 kHz and 95.59 kHz

Micromachines 2022, 13, x FOR PEER REVIEW 16 of 35

As can be seen from Table 3, TFCW is suitable for measuring a distance within a
short range. In order to expand the range with two different frequencies waves, the phase
detection method can also be combined with other methods such as the amplitude of
method
the waveform cannotwhich meet the canrequirements of minimum
achieve the resolution resolution
of 1.5% and maximum
wavelength over therangedistance at the
of
same time.
550–1450 mm [130,131].While the MFCW method derived from the TFCW method which uses three
frequencies ultrasonic waves for range measurement can meet the requirements simulta-
4.2.2. Multi-Frequency Continuous Waves (MFCW)
neously.
The
Accordingworking to principle
the ranging of MFCW
principle in of[132] is shown
TFCW, in Figurerange
its minimum 13. Supposeresolutionthatisthec/
(∆ f ·360 ) (m/
frequencies ◦ of the◦ ), which
transmittedmeanscontinuous
that a larger waves of different
frequency frequencies
difference can obtain f f
are 1a, higher
2 and
resolution but results in the decrease in the measuring range. Therefore, the TFCW method
f3 ( f1＞f2＞f3 ) respectively, the phase shifts of the received signal and the transmitted sig-
cannot meet the requirements of minimum resolution and maximum range at the same time.
nal
While arethe ϕ1 ,MFCW
ϕ2 and ϕ3 respectively.
method derived from Δϕthe
1 isTFCW
the phasemethoddifference
whichbetween
uses three phase ϕ1 and
frequencies
ϕ2 , and the
ultrasonic waves for range measurement
corresponding can meet is
frequency difference theΔrequirements
f1 ( Δf1 = f1 − fsimultaneously.
2 ). Similarly,
Δϕ2 is
The working principle of MFCW in [132] is shown in Figure 13. Suppose that the
the phase difference
frequencies between phase
of the transmitted ϕ1 andwaves
continuous ϕ3, and of the corresponding
different frequencies frequency
are f 1 , differ-
f 2 and
f 3 ( f 1 is> Δ
ence f 2f > Δff32) =respectively,
f1 − f3 ). the phase shifts of the received signal and the transmitted
2(
signal are ϕ1 , ϕ2 and ϕ3 respectively. ∆ϕ1 is the phase difference between phase ϕ1 and
ϕ2 , and Thethe calculation
corresponding formula of the distance
frequency difference is [23]
is ∆ f 1 (∆ f 1 = f 1 − f 2 ). Similarly, ∆ϕ2 is the
phase difference between phase é j  f ù c
and , and é jcorresponding
the ù frequency difference is
ê 2  f1 ú c + j1  c
ϕ 1 ϕ 3
∆ f 2 (∆ f 2 = f 1 − f 3 ). L = Int ê 1  2 ú +Int (17)
ê 2p  f ú  f ê 2p  f ú f 2p f
The calculation formula of ë the distance
1û ë
2 is [23] 2û 1 1

4 ϕ1 4 f 2 c 4 ϕ2 f 1 c
   
+ 1 · by c / Δf2 (m /(17)
ϕ c °) .
The first step, L= yielding
Int the· largest resolution
+Int scale· is determined
2π 4 f 1 4 f 2 2π 4 f 2 f 1 2π f 1
In the second step, yielding finer resolution is determined by c / f1 (m / °) . In the final
The first step, yielding the largest resolution scale is determined by c/∆ f 2 (m/◦ ).
step,
In thethe highest
second level
step, of resolution
yielding is determined
finer resolution by c / (360°
is determined byf1 c/) (m °) . ◦Taking
f 1 (/m/ ). In the FMCWfinal
step, the highest level of resolution is determined by c/ ( 360 ◦ · f )( m/ ◦ ) . Taking FMCW
of frequencies f1 = 40.0 kHz , f2 = 39.9kHz , f3 = 38.0kHz as an1 example, due to most
of frequencies f 1 = 40.0 kHz, f 2 = 39.9kHz, f 3 = 38.0kHz as an example, due to most
commercial ultrasonic transducers having a narrow bandwidth of 40±2 kHz [125], its high-
commercial ultrasonic transducers having a narrow bandwidth of 40 ± 2 kHz [125], its
est resolution is 0.0243 mm/degree.
highest resolution is 0.0243 mm/degree.

Figure 13. The

Figure 13. The working
workingmechanism
mechanismofofthe
theMFCW
MFCW algorithm
algorithm where
where L represented
L represented the overall
the overall dis-
distance
tance to be measured between transmitter and receiver.
to be measured between transmitter and receiver.

References [23,132] use the MFCW method, and the performances of the two ranging
systems are summarized in Table 4.

Table 4. Performance comparison of several MFCW based ranging systems.

 (1500 mm in Table 4) than the TFCW method (70~150 mm in Table 3)
accuracy of both the TFCW and the MFCW depends on the phase mea
and the maximum range depends on the frequency difference. How
Micromachines 2022, 13, 520 ment period becomes longer and the measurement rate is low due
16 of 33 to th
ing to transmit two or three different frequency signals successively.
References [23,132] use the MFCW method, and the performances of the two ranging
4.3. Signal
systems Modulation
are summarized Method
in Table 4.
4.3.1. Binary Frequency
Table 4. Performance ShiftMFCW
comparison of several Keying
based(BFSK)
ranging systems.

Binary frequency
Range shiftAccuracy
keying is similarSignalto the conventional freq
Frequency
method
[23] except1500for
mm its center or carrier40.0
0.05 mm frequency
kHz, 39.9 kHzis
andshifted
38.0 kHz by the b

which
[132] varies between logic 0 and logic 1.
<100 mm 0.0711 mm 497.0 kHz, 496.8 kHz and 487 kHz
100 mm~300 mm 1.8208 mm 492 kHz, 491.8 kHz and 490 kHz
BFSK signal can be expressed as [133]
In conclusion, the MFCW method can achieve high accuracy within a larger range
M


(1500 mm in Table 4) than the TFCW method (70~150 mm in Table 3). The measurement
u(ti )=A onΠthe
accuracy of both the TFCW and the MFCW depends
(ti )sin[2
j phase
π fbj (ti − accuracy
measurement
dj −1 )]
j =1
and the maximum range depends on the frequency difference. However, the measurement
period becomes longer and the measurement rate is low due to the transmitter needing to
transmit two or three
Where A isdifferent frequency signals
the amplitude of thesuccessively.
transmitted signal, bj
is the jth el
4.3. Signal Modulation Method
0,1 code sequence. When bj = 0 , fbj = f0 , and when b j = 1 , fbj = f1 .
4.3.1. Binary Frequency Shift Keying (BFSK)
Binary frequency
different shift keying
frequencies is similar
and the function Π j (t i )
to the conventionalisfrequency
defined modulation
as follows
method except for its center or carrier frequency is shifted by the binary input signal which
varies between logic 0 and logic 1.
BFSK signal can be expressed as [133] 1  for d ≤ t ≤ dj −1
Π j (ti ) = 
i j
M
u(ti ) = A ∑ Π j (ti ) sin[2π f bj
j =1
(t0i − d j−1 )] elsewhere (18)

where A is the amplitude of the transmitted signal, b j is the jth element of the binary 0,1
and
code sequence. When b j = 0, f bj = f 0 , and when b j = 1, f bj = f 1 . f 0 and f 1 are two different
frequencies and the function Π j (ti ) is defined as follows
j
1
Π j ( ti ) =

1 f or 
djd j=−1v≤ ti ≤ d j
0 elsewhere k =1 fb
j=1,2,3..., M
(19)
K

and j
d0 = 0 . v represents
d j = v ∑the number
1
j = 1, 2, 3of
. . .wavelengths
,M corresponding
(20)
f
k =1 Kb
value, which determines the duration of the total pulse d M .
d0 = 0. v represents the number of wavelengths corresponding to each bit of code
value, An
whichexample
determinesof
theaduration
transmitted BFSK
of the total pulse dsignal
M. is shown in Figure 14.
An example of a transmitted BFSK signal is shown in Figure 14.

Figure 14. Transmitted BFSK signals with a binary code of 11010 (d0 = d1 = d3 = 0 and the
Figure 14. Transmitted BFSK signals with a binary code of 11010 ( d0 = d1 =
corresponding frequency is f 0 , d2 = d4 = 1 and the corresponding frequency is f 1 ).
sponding frequency is f0 , d2 = d4 = 1 and the corresponding frequency is f

The transmitted signal can then be received and processed to ob

mation and calculate the distance, using methods of the phase measur
Micromachines 2022, 13, 520 17 of 33

The transmitted signal can then be received and processed to obtain the ToF infor-
mation and calculate the distance, using methods of the phase measurement method,
correlation algorithm or the combination of different methods.
First, the phase measurement method can be adopted. The method in [24] is repre-
sented as shown in Figure 15. For example, when the transmitted BFSK signal is

As sin(2π f 1 t) − t p <t<0
u(t) = (21)
As sin(2π f 2 t) 0 ≤ t<t p
where As is the amplitude of the transmitted signals, f 1 and f 2 are frequencies of transmitted
signals respectively, 2t p is the total pulse duration, without considering any distortion, the
received signal can be expressed as

Ar sin[2π f 1 (t − τ ) + ψ1 ]+n t) τ − t p <t<τ
r (t) = (22)
Ar sin[2π f 2 (t − τ ) + ψ2 ]+n t) τ ≤ t<t p + τ
where Ar is the amplitude of the received signals, τ is the time delay which is ToF, ψ1 and
ψ2 are the phase shifts respectively, n(t) is Gaussian noise. Extracted from Equation (22),
the phase of the two signals can be expressed as

θ1i = 2π f 1 ti − 2π f 1 τ + ψ1 + ω (ti ) τ − t p <ti <τ ,i = 1, . . . , m (23)

θ2j = 2π f 2 t j − 2π f 2 τ + ψ2 + ω (t j ) τ ≤ t<t p + τ , j = 1, . . . , n (24)

where ω (t) is phase noise. The phase change can be expressed as shown in Figure 15
through linear regression of θ1 and θ2 , and the intersection of the two lines tintersection can
be expressed as
b − b2
tintersection = 1 = τ+K (25)
a2 − a1
where a1 = 2π f 1 , b1 = −2π f 1 τ + ψ1 , a2 = 2π f 2 , b2 = −2π f 2 τ + ψ2 . Equation (25) can be
ψ −ψ
simplified as tintersection = τ + K, where K = 2π (1f − 2f ) is a constant, and tintersection we get
2 1
from Figure 15, so the ToF τ can be obtained.
Second, correlation calculation can be adopted as polarity correlation function due to
both of those signals are always a logic one or a logic zero, and is defined by [133]
1 N
N i∑
Cuv (τl ) = sgn[u(ti )]sgn[v(ti+l )] (26)
=1

with τl = lT, l = 0, 1, 2, . . . , L, where T is the sampling time interval. u(ti ) is the transmitted
signal, the returning echoes v(ti ) digitized by a comparator circuit and converted to a
binary representation 
1 f or v(ti ) > k
v k ( ti ) = (27)
−1 f or v(ti ) ≤ k
where k is the threshold set by the comparison circuit. The correlation signals are analyzed
for peak detection to obtain the ToF.
Third, the phase-shift detection method can be combined with the direct measurement
of ToF to complete ranging [134]. The target distance is expressed as d = (c·∆t)/2, where ∆t
is ToF, d is divided into the regions as shown in Figure 16, [(k − 1) Lr , kLr ] (k = 1, 2, 3, . . .), Lr
is the wavelength of frequency difference ∆ f , then d can be expressed as d = 1/2[(k − 1) +
(∆θ/2π )]·(c/∆ f ) since k is an integer. Therefore, the distance is

∆θ
 
1 c
d= Int(∆t·∆ f )+ · (28)
2 2π ∆ f

The performance achieved by different processing methods is summarized in Table 5.

 Where a1 = 2π f1 , b1 = −2π f1τ + ψ 1 , a2 = 2π f 2 , b2 = −2π f 2τ + ψ 2 . Equation (25
1ψ −ψfor v(ti )＞k
simplified as tintersection v (t
= τ + K , wherek K ) =
i =
1 2
is a constant, and tintersection
Micromachines 2022, 13, 520
2− f1 )v(ti )≤k
π1( f2 −for 18 of 33
from Figure 15, so the ToF τ can be obtained.
Where k is the threshold set by the comparison circuit. The correlation sig
analyzed for peak detection to obtain the ToF.
Third, the phase-shift detection method can be combined with the direct m
ment of ToF to complete ranging [134]. The target distance is expressed as d = (
where Δ t is ToF, d is divided into the regions as shown in Fig
[( k − 1) Lr , kLr ] ( k = 1, 2, 3,...) , Lr is the wavelength of frequency difference Δ f , th
be expressed as d = 1 / 2[( k − 1) + (Δθ / 2π )](c / Δf ) since k is an integer. Therefore
tance is

1 Δθ  c
d = Int(ΔtΔf )+ 
2  signal phase2πverse
Figure 15. A function diagram of received BFSK  Δftime if using phase
Figure 15. A function diagram of received BFSK
measurement method.
signal phase verse time if using phase meas
method.

Second, correlation calculation can be adopted as polarity correlation function

both of those signals are always a logic one or a logic zero, and is defined by [133]

1 N
Cuv (τ l ) =  sgn[u(ti )]sgn[v(ti + l )]
N i =1

with τ l = lT , l = 0,1, 2,..., L , where T is the sampling time interval. u(ti ) is th

mitted signal, the returning echoes v(ti ) digitized by a comparator circuit and con
Relation between
Figure 16.16. standard ToF method where ·∆t)/2 and
d = (cwhere d = (the
cΔcombined
t) / 2 andmethod
toFigure
a binary Relation between
representation standard ToF method the combined
where d = 1/2[(k − 1) + (∆θ/2π )]·(c/∆ f ).
where d = 1 / 2[(k − 1) + (Δθ / 2π )](c / Δf ) .
Table 5. Performance comparison of several BFSK based ranging systems.
The performance
Reference Rangeachieved by Accuracy
different processing methodsMethod
Signal Processing is summarized
5. [24] ~5 m 1 mm Phase measurement
[135] 0.5 m~4 m / Correlation algorithm
Table 5. Performance comparison of several BFSK based ranging systems.
[136] ~7.2 m 6.2 cm Correlation algorithm
Reference
[134] ~6Range
m Accuracy
0.05 mm ToFSignal processing
and phase shift metho
[24] ~5 m 1 mm Phase measurement
The BFSK
[135] method can improve
0.5 m~4 m the power of the
/ received signal, which is beneficial
Correlation algorithm
to the signal processing process. Compared with the correlation algorithm, the phase
[136] combined with
shift method ~7.2 ToF
m can obtain better
6.2 cm Correlation
measurement accuracy, which algorithm
can
[134]
reach 0.05 ~6 mmethod can be0.05
mm [134]. The BFSK usedmm to measure a longer ToF
rangeand
due phase
to the shift
outstanding energy characteristic.

4.3.2. Amplitude Modulation (AM) Method

AMThe BFSKismethod
method can improve
to add different amplitudethe power oftothe
information thereceived
ultrasonic signal, which is b
transmitted
to thetosignal
signal improveprocessing process.
the measurement Compared
accuracy, with the
measurement rate correlation algorithm, the ph
and other performances.
method combined
The first applicationwith ToF can obtain
is the multifrequency bettermodulation
amplitude measurementmethodaccuracy,
(MFAM) [25]which c
based on phase measurement. If the transmitter is excited by a continuous
0.05 mm [134]. The BFSK method can be used to measure a longer range due to sine wave, the
received echo signal can be expressed as
standing energy characteristic.
VR (t) = A sin(ωt + ϕ) (29)
4.3.2.AAmplitude
where Modulation
is the peak amplitude of the (AM) Method
echo signal, ω is the resonant angular frequency, ϕ is
phase shift which is proportional to the measures
AM method is to add different amplitude distance. Assume that to
information both theultrasonic
the carrier
tran
signal and the modulation signal are assumed to be sinusoidal signals with zero phase shift
signal to improve the measurement accuracy, measurement rate and other perfor
The first application
VT (t) =isAthe multifrequency
m [1 + m sin(ωm t)] Ac sinamplitude
( ωc ) modulation(30)
method
[25] based on phase measurement. If the transmitter is excited by a continuous si
the received echo signal can be expressed as
 both the carrier signal and the modulation signal are assumed to be sinusoidal
with zero phase shift
VT (t ) = Am [1 + m sin(ω m t )]Ac sin(ωc )
Micromachines 2022, 13, 520 19 of 33
Where A m and Ac are the peak amplitudes of the modulated signal and
signals, respectively; ω m
and ω are the angular frequencies of the modulate
c
where Am and Ac are the peak amplitudes of the modulated signal and carrier signals,
and carrier signals, respectively; m is the modulation
respectively; ωm and ωc are the angular frequencies index signal
of the modulated whichand defines
carrierthe rati
plituderespectively;
signals, between modulating signal and
m is the modulation carrier
index whichsignal.
definesThe waveform
the ratio and spectral
of amplitude
between modulating signal and carrier signal. The waveform
of the transmitted signal are shown in Figure 17 as an example.and spectral density of the
transmitted signal are shown in Figure 17 as an example.

Figure (a)(a)Waveform
Figure17.17. Waveform and (b)
andspectral density density
(b)spectral of an amplitude-modulated signal.
of an amplitude-modulated signal.
When the modulation index is less than 1, the envelope of the modulation waveform
When the
is proportional modulation
to the modulationindex
signal,is less
and thethan 1, the
detected envelope
echo of signal
modulation the modulation
is wa
is proportional to the modulation
0 signal, and the
0 detected echo modulation signa
VR (t) = Am [1 + m sin(ωm t + ∆ϕm )] Ac sin(ωc t + ∆ϕc ) (31)

where ∆ϕm and ∆ϕc are the V ( t ) = A [1

phase
R of +
'
shiftsm sin(ω m t + signal
mmodulated
the Δ ϕ m )] A sin(
'
and ω c tsignal,
carrier
c
+ Δ ϕ crespec-
)
tively. ∆ϕm and ∆ϕc are proportional to the measured distance L. When using phase-shift
Where
calculations Δϕ m andsignals,
of modulated Δϕc the aredistance
the phase
L is shifts of the modulated signal and car
nal, respectively. Δϕ m and Δϕc are∆ϕproportional
m ·c to the measured distance L. W
L= (32)
2π f m signals, the distance L is
ing phase-shift calculations of modulated
When L/λm <1 the phase difference ∆ϕm provides a unique measurement distance L
and the maximum measuring distance is λm which is dependent Δϕm c on f m .
L =
By adopting MFAM, the measurement range can2reach π fm the level of 10 m with the
measurement accuracy in millimeters [137,138]. Since the distance information is obtained
according
When L λ m ＜1 the phase difference Δϕ m provides of
to the envelope of the received signal, so the performance the measurement
a unique measurement
system will not be affected by the loss of the initial part of the received signal due to it could
L and
be the maximum
too weak to detect. Themeasuring distanceofisAMλmultrasonic
sinusoidal envelope whichsignals
is dependent on f m .
can be extracted
real-time using the sliding discrete Fourier transform (SDFT) algorithm. In addition, for
By adopting MFAM, the measurement range can reach the level of 10 met
25 Hz modulation, the present method shows a range of 13.72 m and an update rate of
the measurement
3200/s for static objects. accuracy in millimeters
For moving objects, the[137,138].
update rateSince thevaries
directly distance
with informatio
the
modulating frequency. The transient update rate for 25 Hz modulation is about 8/s in [138], of th
tained according to the envelope of the received signal, so the performance
urement
so system
it is suitable willand
for static notslow-moving
be affected by the
targets loss of the initial part of the receive
measurement.
The performance parameters that can be
due to it could be too weak to detect. The sinusoidal achieved by the ranging systems
envelope of based on
AM ultrasonic
AM method are summarized in Table 6.
can Thebe extracted real-time using the sliding discrete Fourier transform (SDFT) alg
second application is the combination of amplitude modulation and phase modu-
In addition,
lation method (AM for 25 Hzand
+ PM) modulation, the from
obtains the ToF present method
the peak of theshows
receivedasignals.
range Theof 13.72 m
update
peak rate
of the of 3200/s
received signalfor static isobjects.
envelope sharper For moving
through objects,
amplitude the update
modulation rate directl
and phase
inverted
with the(AMPI) of the transmitted
modulating frequency. wave,Theastransient
shown in Figure
update 18.rate
This for
is because
25 Hzseveral
modulation
periods of signal with opposite phases are added after the conventional excitation wave to
8/s in [138], so it is suitable for static and slow-moving targets measurement.
reduce the trailing part of the signal that is not expected to appear, and when the coefficient
 [137] MFAM ~1.5 m 2 mm
~13.72 m
[138] MFAM
~3 m 1.343 mm
Micromachines 2022, 13, 520 [139] AM+PM 0.1 m~0.5 m 0.02
20 mm
of 33
[140] AM+PM 0.05 m~0.5 m 0.2 mm
0.05 m~1 m 0.06 mm
[141] AM+PM
~2 m signal) is larger, the
k = A2 /A1 (A1 and A2 are the amplitudes of the AMPI driving 0.15effect
mm
is more obvious. Its sharp envelope improves the measurement accuracy of the peak time.

Table 6. Performance comparison of several AM based ranging systems.

The second application is the combination of amplitude modulation and phase mod
ulation method (AM+PM)
Reference and Method
Modulation obtains the ToFRange
from the peak of the received signals. Th
Accuracy
peak of[137]
the received signalMFAM
envelope is sharper~1.5through
m amplitude modulation
2 mm and phas
inverted (AMPI) of the transmitted wave, as shown ~13.72 min Figure 18. This is because sever
periods[138] MFAM
of signal with opposite phases are added~3 mafter the conventional
1.343 mmexcitation wav
to reduce
[139]the trailing partAM
of +the
PMsignal that is
0.1not expected
m~0.5 m to appear, and when the coe
0.02 mm
k= A2 A1 ( A1 andAM
ficient [140] A 2 + are
PM the amplitudes of the
0.05 m~0.5 m AMPI driving signal) is larger, th
0.2 mm

effect is[141]
more obvious. ItsAM
sharp 0.05 m~1the
envelope improves m measurement
0.06accuracy
mm of the pea
+ PM
~2 m 0.15 mm
time.

Figure
Figure 18.18. Ultrasonic
Ultrasonic waveform
waveform (a) (a) Conventional
Conventional driving
driving signalsignal and
and (b) its(b) its received
received waveform. (
waveform.
AMPI driving signal and (d) its received waveform [139].
(c) AMPI driving signal and (d) its received waveform [139].

Another
Another specific modulation
specific modulationmethod
methodof AM + PM is used
of AM+PM in in
is used [140], to to
[140], reduce
reduce thethe mea
measurement error of ToF extraction caused by inertia delay and amplitude attenuation
urement error of ToF extraction caused by inertia delay and amplitude attenuation of th
of the echo signal. The method is represented in this paper as shown in Figure 19. The
echo signal. The method is represented in this paper as shown in Figure 19. The transmi
transmitted signal ST consists of warm up waves and measurement waves. The first low-
ted signal
amplitude ST consists
square waves ofarewarm upwarm
used to waves upand measurement
transducers, which canwaves. The first
eliminate low-amplitud
the inertia
square waves are used to warm up transducers, which can eliminate
delay caused by the piezoelectric effect of the ultrasonic transducer. The high pulse the inertia
is dela
caused
the phaseby the piezoelectric
modulation effect of the
and measurement ultrasonic
pulse. transducer.pulse
The measurement The high ◦
pulse
is 180 out is
ofthe phas
modulation
phase with theand measurement
warm-up pulse. The
pulse. Therefore, the measurement
receiving signalpulse is 180°
can easily out ofthe
identify phase wit
position of the measured pulse. Here, the received signal R is displayed in the form
the warm-up pulse. Therefore, the receiving signal Tcan easily identify the position of th of the
transformed
measured square
pulse.wave.
Here,Combined withsignal
the received the clock,
RT is ToF TF canin
thedisplayed bethe
obtained,
form ofwhich
the can
transforme
be expressed as
square wave. Combined with the clock, the  ToF T can be obtained, which can be e
F
θST
pressed as TF = N + Tperiod (33)
2π
where N represents the integral part of the clock cycle, θST is phase shift and Tperiod is the
clock cycle.
By adopting AM + PM, the measurement accuracy of peak time of received signal can
be improved due to the reduction of inertia delay. The performance of the ranging system
in [139–141] is summarized in Table 6. It can be seen that the measurement accuracy can
be very high in the short distance range, up to 0.02 mm, smaller than 0.3% wavelength,
improving the ranging accuracy effectively. Therefore, MFAM is more suitable for the
Micromachines 2022, 13, x FOR PEER REVIEW 22 of 35

Micromachines 2022, 13, 520  θ  21 of 33

TF =  N + ST  Tperiod (33)
 2π 

Where N represents the integral part of the clock cycle, θ is phase shift and Tperiod
requirement of high update rate and larger range, while AM +STPM is more suitable for the
is the clock cycle.
requirement of high accuracy and smaller range.

Figure
Figure 19.
19. Examples
Examples of
of transmitted
transmitted and
and received
received signals
signals and ToF acquisition
and ToF acquisition method.
method.

4.3.3.By adopting
Coded Signal AM+PM,
ExcitationtheMethod
measurement accuracy of peak time of received signal can
be improved due to the reduction
Signal coding is to add characteristics of inertia delay. The performance
to transmitted signals,of the ranging
including system
frequency,
in
phase, pulse position, pulse width, etc. It has recorded benefits to increase the SNRcan
[139–141] is summarized in Table 6. It can be seen that the measurement accuracy of
be
thevery high signal,
received in the short distance
to reduce range,between
crosstalk up to 0.02mm,multiplesmaller
sensors than 0.3%
and wavelength,
to improve the
improving
measurement therate
ranging accuracy effectively.
in long-distance measurement. Therefore, MFAM is more suitable for the re-
quirement of high
At present, theupdate
coding rate and larger
sequence used inrange, while system
the ranging AM+PM is moreChirp
includes suitable forGold
[142], the
requirement of high accuracy and smaller range.
codes [143], Kasami codes [144], Golay codes and Loosey Synchronous codes [145], all of
which improve the quality of the received echo signal. Encoding the transmitted signal and
4.3.3.
matchingCoded Signalat
filtering Excitation Method
the receiving end can improve the performance of noise immunity
due to the good
Signal coding auto-correlation property ofto
is to add characteristics thetransmitted
codes, thussignals,
improving the measurement
including frequency,
resolution
phase, pulseand measurement
position, accuracy.
pulse width, etc. It has recorded benefits to increase the SNR of the
In the
received case where
signal, to reduce there are multiple
crosstalk between sensors in thesensors
multiple ranging andsystem, the crosstalk
to improve the meas-be-
tween sensors will cause confusion
urement rate in long-distance measurement. and damage to the system performance. Minimizing
the crosstalk
At present, effects
the iscoding
a key problem
sequenceinused the field
in the ofranging
ultrasonic measurement
system includes [146].
Chirp There-
[142],
fore, itcodes
Gold is necessary to add codes
[143], Kasami distinguishing
[144], Golaycharacteristics to the signals
codes and Loosey sent by codes
Synchronous each sensor.
[145],
Theofsignals
all whichcan be modulated
improve by adopting
the quality e.g., Barker
of the received echocodes
signal.[147], Golay the
Encoding codes [148,149],
transmitted
M-Sequence
signal [150], pulse
and matching position
filtering modulation
at the receiving [151,152] and pulsethe
end can improve width modulation
performance [153],
of noise
even a combination of the multiple methods above. For example,
immunity due to the good auto-correlation property of the codes, thus improving the M-sequence (which
is one of the pseudorandom
measurement sequences generated
resolution and measurement accuracy.from a linear feedback shift register
(LFSR))combined with Chirp signals
In the case where there are multiple sensorsis shown in Figure
in the20. An M-sequence
ranging system, the is crosstalk
a pseudoran-be-
dom sequence
tween sensors willof binary
cause words composed
confusion and damage of “1” to the“−
and nth-order M-sequence
1”. Anperformance.
system Minimizing is
generated from an n-bit LFSR and its length is 2 n − 1 words. The binary words determine
the crosstalk effects is a key problem in the field of ultrasonic measurement [146]. There-
the phase
fore, and the Chirp
it is necessary to addsignals determinecharacteristics
distinguishing the frequencytoofthe thesignals
transmitted
sent by signals. In the
each sensor.
figure, code values of different positions in the same sequence
The signals can be modulated by adopting e.g. Barker codes [147], Golay codes [148,149], are selected for encoding.
The key to coding
M-Sequence [150], is to find
pulse codesmodulation
position with good autocorrelation
[151,152] and pulse and bad width cross-correlation,
modulation[153], so
that each code has its own characteristics and the receiver can identify whether it is the
even a combination of the multiple methods above. For example, M-sequence (which is
signal sent by the corresponding transmitter.
one of the pseudorandom sequences generated from a linear feedback shift register
Third, the measurement rate of ultrasonic ranging is low in long-distance measurement
(LFSR))combined with Chirp signals is shown in Figure 20. An M-sequence is a pseu-
due to the slow propagation speed of ultrasonic in the air, for instance, the measurement
dorandom sequence of binary words composed of“ 1” and“ −1”. An nth-order M-se-
rate is only 17 Hz when the target distance is 10 m. Therefore, it is necessary to increase the
quence
measurementis generated
rate tofrom
makeanit n-bit LFSR
suitable forand is 2n − 1 words. The binary words
its lengthmeasurement.
high-speed
determine the phasetoand
One approach the Chirp
increase signals determine
the measurement rate isthe frequency
through of the transmitted
the application of pulse
position modulation (PPM). The time interval between each pulse varies, linearly increasing
in [154] shown in Figure 21. The pulse position characteristics will be transferred to the
received signals and reflected by measuring ToF of each pulse to obtain distance.
 signals. In the figure, code values of different positions in the same seq
for encoding. The key to coding is to find codes with good autocorrela
Micromachines 2022, 13, 520
correlation, so that each code has its own characteristics and 22 of 33
the rec
whether it is the signal sent by the corresponding transmitter.

Figure 20. Two LFM signal examples coded by the same M-sequence but start from dif
quence indexes.

Third, the measurement rate of ultrasonic ranging is low in long-distance m

ment due to the slow propagation speed of ultrasonic in the air, for instance, the m
ment rate is only 17 Hz when the target distance is 10 m. Therefore, it is nece
increase the measurement rate to make it suitable for high-speed measurement
One approach to increase the measurement rate is through the application
position modulation (PPM). The time interval between each pulse varies, linearly
ing in [154]
Two
Figure 20.20.
shown
LFMLFM
in examples
Figure 21.
signal signal
The pulse position characteristics
coded by thecoded
same M-sequence but start
will be transf
from different but sta
Figure Two examples by the same M-sequence
the received
sequence indexes.signals and reflected by measuring ToF of each pulse to obtain dista
quence indexes.

Third, the measurement rate of ultrasonic ranging is low in long-

ment due to the slow propagation speed of ultrasonic in the air, for inst
ment rate is only 17 Hz when the target distance is 10 m. Therefore
increase the measurement rate to make it suitable for high-speed meas
One approach to increase the measurement rate is through the a
position modulation (PPM). The time interval between each pulse varie
Figure 21. Pulse position modulation signals.
Figure 21.
ing in [154] Pulse positionin
shown modulation
Figure 21. signals.
The pulse position characteristics wi
Reference [154] reported that they set the pulse time interval to increase from 3 ms
the received
Reference
to 11 ms,
signals
[154]
with a step
and
reported
of 1 ms.
reflected
that they
The average
by measuring
set the
theoretical
ToF
pulse timerate
measurement
of each
interval
pulse
to143
can reach increase
Hz
to o
from
due
11ms,to the
withaverage
a stepinterval
of 1ms. of The
7 ms.average
In fact, theoretical
according tomeasurement
the experiment rateresults,
canthereach 14
measurement rate of 4 m target can be increased by 2.7 times [154]. However, when the
to the average interval of 7 ms. In fact, according to the experiment results, the m
distance increases, there is a big difference between the actual measurement rate and the
ment rate
theoretical of 4m
value, whichtarget
is duecan be influence
to the increased by 2.7
of noise times
signal [154].
when However,
the distance when the
increases.
increases, there is a big difference between the actual measurement
Chaotic pulse position modulation (CPPM) [155] is another approach that is derived rate and the th
from
value,PPM. The CPPM
which is duemethodto the was used to of
influence increase
noisethe measurement
signal when the rate in [156]. increases.
distance The
method is represented in this paper as shown in Figure 22. The chaotic information of
Chaotic pulse position modulation (CPPM) [155] is another approach that is
the analogue voltage is transferred into the intervals between digital pulses. The received
from PPM.
signals The CPPM
are converted into pulsemethod
signalswas usedintoFigure
as shown increase thethen
22d and measurement
cross-correlated rate in [1
method
with is represented
single-bitted transmitted in this papersignals
reference as shown in Figure
as shown 22. 22a
in Figure Thetochaotic
obtain informatio
the
analogue voltage is transferred into the intervals between digital pulses. The rece
measured distance of each pulse.
nalsAaremeasurement
converted rateinto
greater thansignals
pulse 100 Hz canas be obtained
shown in up to the22d
Figure distance
andrange
thenofcross-co
9 m, reported
Figure 21. in [156],
Pulse which effectively
position modulationimproves the measurement rate compared to the
signals.
with single-bitted transmitted reference signals as shown in Figure 22a to obtain th
traditional pulse-echo method. The performance of this method is better than PPM. This
ured
also distance
shows of correlation
that this each pulse. method can extend the ranging distance to 5–10 m whilst
Reference
still obtaining a better[154]
result reported that they
of the measurement rate. set the pulse time interval to in
In conclusion, ToF is the
11ms, with a step of 1ms. The averagebasic method, usually using ATM and correlation
theoretical methods forrate ca
measurement
data processing. TFCW and MFCW adopt several signals of similar frequency for ranging,
to
andthedataaverage
processinginterval
is based onof 7 ms.
phase In fact,
detection according
to improve to the
the accuracy experiment re
in a short-range.
ment
BFSK and rate
AMof 4m target
methods modulatecan be increased
transmitted by 2.7
signals which can times
improve[154]. However,
the quality of
received signals to improve the measurement range or measurement accuracy. The signal
increases,
coding method there is aidentifiable
is to add big difference
features between the actualsignals,
to different transmitted measurement
which can rate
value,
solve thewhich
problem is due to thecrosstalk
of multi-sensor influence of noise
and improve signal when
the measurement rate inthethe distance
case i
of long distances.
Chaotic pulse position modulation (CPPM) [155] is another appro
from PPM. The CPPM method was used to increase the measuremen
method is represented in this paper as shown in Figure 22. The chaotic
analogue voltage is transferred into the intervals between digital pulse
 Micromachines
Micromachines 2022,
2022, 13,13,
520x FOR PEER REVIEW 24ofof3335
23

Figure22.
Figure 22.Signal
Signalprocessing
processingprocedure.
procedure.(a)
(a)Reference
Referencesignal;
signal;(b)
(b)Transmitted
Transmittedsignal
signal(V);
(V);(c)(c)Received
Received
signal(V);
signal (V);(d)
(d)Single-bit
Single-bitsignal.
signal. .

A measurement
5. Ranging Error Analysis rate greater than 100 Hz can be obtained up to the distance range of
and Compensation
9 meters, reported in [156],
The source of ranging error is mainly which effectively improves
divided into two theparts,
measurement
one is therate
systemcompared
error
to the traditional pulse-echo method. The performance of this
such as the acquisition error of ToF, the other is the external environmental error such method is better than PPM.
as
This also shows that this correlation
sound velocity variation due to temperature, humidity. method can extend the ranging distance to 5–10 me-
ters whilst still obtaining a better result of the measurement rate.
5.1. System AcquisitionToF
In conclusion, Error andbasic
is the Compensation
method, usually using ATM and correlation methods
for data processing. of
The acquisition TFCW and MFCW
ultrasonic ToF isadopt
relatedseveral
to thesignals
qualityofofsimilar frequency
the echo signal and for rang-
the
ing, and data processing is based on phase detection to
processing method. Under the premise of the same processing method, the better theimprove the accuracy in a short-
range.of
quality BFSK and AM
the echo methods
signal, the moremodulate
accurate transmitted signals
the acquisition of which
ToF. can improve the qual-
ity of
First, the error is caused by the noise of the ultrasonic echo measurement
received signals to improve the measurement range or accuracy.
signal. Signal filtering The
such
assignal
Kalman coding method
filtering [157]is can
to add be identifiable
used to remove features to differentsignals,
the irrelevant transmitted signals,
and then which
improve
canSNR
the solve of the
theproblem
echo signal. of multi-sensor
The coded signalcrosstalk and improve
excitation method theinmeasurement
Section 4.3.3 ratecan in the
also
case of long
improve distances. of noise immunity. In addition, increasing the transmitting signal
the performance
intensity can improve the strength of the received signal, thus reducing the error of ToF
5. Rangingcaused
acquisition Error Analysis
by noise. and Compensation
Second,
The source the range error error
of ranging will increase
is mainly with distance
divided intodue
twotoparts,
the significant
one is the attenuation
system error
ofsuch
echoassignal amplitudeerror
the acquisition in the ofair.
ToF, Therefore,
the other aisgain compensation
the external module [80,158]
environmental error such canas
besound
added inside the receiving circuit to
velocity variation due to temperature, humidity.amplify the echo amplitude and compensate for
the attenuation of ultrasonic energy. The compensation principle in [80] is shown in
Micromachines 2022, 13, x FOR PEER Figure
REVIEW 23, which makesError the andechoCompensation
signal intensity of different distances consistent and
5.1. System Acquisition
reduces the acquisition error of ToF caused by subsequent signal processing due to different
The acquisition of ultrasonic ToF is related to the quality of the echo signal and the
signal intensities.
processing method. Under the premise of the same processing method, the better the qual-
ity of the echo signal, the more accurate the acquisition of ToF.
First, the error is caused by the noise of the ultrasonic echo signal. Signal filtering
such as Kalman filtering [157] can be used to remove the irrelevant signals, and then im-
prove the SNR of the echo signal. The coded signal excitation method in Section 4.3.3 can
also improve the performance of noise immunity. In addition, increasing the transmitting
signal intensity can improve the strength of the received signal, thus reducing the error of
ToF acquisition caused by noise.
Figure Second,
Gain
23.23. the range error
compensation (a) will
Soundincrease
intensitywith distance
varies due to(b)
withvaries
distance; theThe
significant
relationshipattenuation
between
Figure
of echo
Gain
signal
compensation
amplitude in the
(a)
air.
Sound intensity
Therefore, a gain
with distance;
compensation module
(b)[80,158]
The relationship
can
received signal intensity and distance; (c) Compensation coefficient; (d) The relationship between
received
be added signal
inside intensity
the receiving and distance;
circuit to (c) Compensation
amplify the echo coefficient;
amplitude and (d) The
compensate relationship
for
received signal intensity and distance after compensation.
received signal intensity
the attenuation of ultrasonic andenergy.
distance Theafter compensation.
compensation principle in [80] is shown in Fig-
ure Finally,
23, which themakes
error theis caused by the
echo signal measurement
intensity angle
of different within consistent
distances the detectable range
and reduces
ofthe
theacquisition
beam spread
Finally, the error angle
error
of ToF When
iscaused
θ. caused the
byby position of the
the measurement
subsequent target and
signal processing the
angle transducer
duewithin present
the detectable
to different signal
the beam spread angle θ . When the position of the target and the transducer
different angles
intensities. which are shown in Figure 24, the ranging errors are different. The greater
the angle is, the greater the ranging error will be [159]. The waveform of the ultrasonic
different angles which are shown in Figure 24, the ranging errors are different. The
the angle is, the greater the ranging error will be [159]. The waveform of the ul
wave away from the center line of the main beam, shown in Figure 3, will be d
which makes the accurate echo time to be more difficult to identify. The detectio
 Finally, the error is caused by the measurement angle within the detectable
the beam spread angle θ . When the position of the target and the transducer
Micromachines 2022, 13, 520
different angles which are shown in Figure 24, the ranging errors are different. 24 of 33
The
the angle is, the greater the ranging error will be [159]. The waveform of the ul
wave away from the center line of the main beam, shown in Figure 3, will be d
which
wave makes
away fromthe accurate
the center line echo
of thetime
main to be more
beam, showndifficult
in Figure to identify.
3, will The detectio
be distorted
which
shouldmakes thebe
either accurate echo time
controlled, to beerror
or the morecan
difficult to identify. Theby
be compensated detection angle
a compensation al
should either be controlled, or the error can be compensated by a compensation
which can be established through pre-measurement with sensors located at diffe algorithm
which can be established through pre-measurement with sensors located at different angles.
gles. Reference [160] reported an improvement of the accuracy from 12.24 mm to
Reference [160] reported an improvement of the accuracy from 12.24 mm to 3.4 mm within
1within
m at 60◦1angle
meter at 60°with
position angle position
angle with angle compensation.
compensation.

Figure Angular
Figure24.24. range
Angular of target
range object detection.
of target object detection.
5.2. Environmental Error and Compensation
5.2. Environmental Error and Compensation
The sound velocity is easily affected by environmental factors, including temperature,
Thepressure,
humidity, sound velocity
etc. As a is easily affected
nonlinear bytemperature,
function of environmental factors,
pressure, including
humidity and temp
CO concentration,
humidity,
2 sound velocity can be expressed as [161]
pressure, etc. As a nonlinear function of temperature, pressure, humid
CO2 concentration, sound velocity
c = f (tcan
C , p, be
xw , expressed
xc ) as [161] (34)

where tc is Celsius temperature, p is the air pressure, xw the water vapor mole fraction, and
xc is the carbon dioxide mole fraction.
c = f (tC , p, xw , xc )
The temperature has the greatest influence on sound velocity, which can be expressed
by thewhere
followingt formula [91] temperature, p is the air pressure, x the water
is Celsius vap
C r w
tC
fraction, and xc is the carbon c = dioxide +
331.45 1mole fraction.
273.15
(35)

whereThe
331.45temperature hasvelocity
m/s is the sound the greatest
in dry airinfluence
at 0 ◦ C, tC ison
thesound velocity,
temperature which
in degree can
pressedAs
Celsius. bytemperature
the following formula
increases, [91] velocity increases. Equation (35) adopts
the sound
Taylor series expansion and can be simplified as
tC
c =+ 331.45
c = 331.45 0.607tC 1+ (36)
273.15
For every 1 ◦ C increase in temperature in air, the speed of sound increases by 0.607 m/s.
The influence
Where of humidity
331.45 m/s is onthethesound
sound velocity
velocityis in
inferior
dry to
airthat
at of0 temperature and temper
°C, tC is the
can be expressed by the following formula [91]
degree Celsius. As temperature increases,  the
 sound velocity increases. Equat
2 RT 2pB
adopts Taylor series expansion c = and
γ can
1 +be simplified as (37)
M RT
where c is sound velocity, T is the temperature c =on3an
31.45 + 0.607
absolute t C γ, p, R, M, B represent
scale,
the specific heat ratio, pressure, the universal gas constant, molecular mass and second
virial coefficient respectively. The speed of sound is faster in moist air than in dry air. The
variation of the speed of sound with respect to temperature and relative humidity is shown
in Figure 25.
  
M RT 

where c is sound velocity, T is the temperature on an absolute scale， γ , p, R, M, B

represent the specific heat ratio, pressure, the universal gas constant, molecular mass and
Micromachines 2022, 13, 520
second virial coefficient respectively. The speed of sound is faster in moist air than25inofdry
33
air. The variation of the speed of sound with respect to temperature and relative humidity
is shown in Figure 25.

Figure25.
Figure 25.Variation
Variationofofthe
thespeed
speedofofsound
soundtototemperature
temperatureand
andrelative
relativehumidity
humidity[162].
[162].

Thevariations
The variationsof ofair
airpressure
pressureand andCO CO2 2concentration
concentrationhave havesmall
smallinfluences
influenceson onthethe
speedof
speed of sound
sound which can can bebe negligible
negligiblewhen
whencompared
comparedtotothe theeffects
effects caused
causedbyby temper-
tem-
perature
ature andand humidity.
humidity. In theIn extreme
the extreme humidity(100%condition,
humidity(100%RH) RH) condition,
there isthere
a worstis ascenario
worst
scenario for the pressure’s
for the pressure’s effect oneffect
soundon sound
speed as speed as pressure’s
pressure’s influenceinfluence
increases increases with
with humidity,
humidity, and where the maximum relative variation of sound speed
and where the maximum relative variation of sound speed caused by air pressure within caused by air pressure
within [50,110] ◦ C and CO concentration at normal
[50,110] kPa iskPa
0.67%is 0.67%
at theattemperature
the temperature
of 30 of°C30and CO2 concentration
2 at normal level
level
(Xc =(Xc = 0.000383)
0.000383) [161].[161]. The influence
The influence of COof CO2 concentration
2 concentration on sound on speed
soundisspeed
even is even
smaller,
smaller, aboutrelative
about 0.06% 0.06% relative
variation variation
of sound of speed
soundin speed 19 ◦ C
19 °Cinwith with amole
a 0.2% 0.2%fraction
mole fraction
change
change
in CO2 in CO2 concentration
concentration [161,163].[161,163].
In
In practice, the ranging systemonly
practice, the ranging system onlycompensates
compensatesfor forerrors
errorscaused
causedby bytemperature
temperature
and
and humidity. The temperature sensor module and humidity sensor modulecan
humidity. The temperature sensor module and humidity sensor module canbe besetset
inside the ranging system to measure the value of environmental parameters
inside the ranging system to measure the value of environmental parameters in real-time in real-time
and
andcompensate
compensatethe thesound
soundvelocity
velocityvalue
valueaccordingly.
accordingly. The The sound
sound velocity
velocity can
canalso
alsobe be
measured
measuredininaareal-time
real-timefashion
fashionalongside
alongsidethe themain
mainsystem,
system,withwithaatarget
targetsetsetup
upatataaknown
known
exact
exactdistance.
distance.TheTheToFToFinformation
informationextracted
extractedfromfromthethetarget
targetcancanbebeusedusedtotocalculate
calculatethe the
sound velocity under current conditions [164]. The advantage is that
sound velocity under current conditions [164]. The advantage is that it can compensate it can compensate for
the
forsound velocity
the sound errors
velocity caused
errors by a by
caused variety of environmental
a variety of environmental factors simultaneously.
factors simultaneously. The
disadvantage is that the space occupied by the ultrasonic ranging device
The disadvantage is that the space occupied by the ultrasonic ranging device will increase. will increase.
In
Inaddition,
addition,multiple
multiplesensors
sensorscan canbebesetsetfor
formeasurement
measurement[165] [165]or orthe
thetime
timeof offlight
flight
ratio
ratio of different transducers can be used to obtain the positioning parameter [166].The
of different transducers can be used to obtain the positioning parameter [166]. The
final measurement result can also be obtained through multiple measurements by a single
final measurement result can also be obtained through multiple measurements by a single
sensor [167,168] to enhance the measurement reliability. For example, M-sequence can also
sensor [167,168] to enhance the measurement reliability. For example, M-sequence can
be used to improve the ToF estimation accuracy as it measures the ToF of each pulse of the
echo signal, and [168] reported that through averaging the 50 measures the error is reduced
to one-tenth of one measure.

6. Conclusions and Outlook

The present status of ultrasonic ranging is reviewed, evaluated and discussed in
this paper.
First, the characteristics of ultrasonic transducers are classified, and their relationship
to the ranging application is analyzed. To increase the energy transfer efficiency of the
transducer, the working frequency of the transducer is selected as close as possible to the
series resonant frequency, and the matching circuit is designed according to its impedance
characteristics. The range of ultrasonic ranging depends on the directivity and propagation
attenuation characteristics of the transducer. The higher the electromechanical coupling
coefficient is, the lower the power consumption of the ranging system will be.
Micromachines 2022, 13, 520 26 of 33

Second, the principle of ultrasonic ranging, system composition and evaluation param-
eters are summarized. Ultrasonic ranging is generally measured by the pulse-echo method.
The ranging system includes transducers, transmitting circuits, receiving circuits, a micro-
processor, compensation modules, etc. The performance of the ranging system is evaluated
according to the measurement range, measurement accuracy and measurement rate.
Third, the method of ultrasonic ranging and its signal processing are classified, and the
performance of the measurement system is analyzed. The measurement results of the ToF
method can be obtained by ATM, envelope fitting and correlation method. Among them,
ATM is the simplest, with the fastest processing speed and a wider range of applications.
The envelope fitting method has lower processing speed and higher theoretical measure-
ment accuracy but is limited by the envelope model. The processing speed of the correlation
method is average, but the overall performance is better. TFCW and MFCW methods use
multi-frequency measurement signals based on phase detection to obtain results. Com-
pared with the ToF method, the measurement range is smaller, and the measurement period
is longer, but the measurement accuracy is higher. BFSK and AM methods modulate the
transmitted signals which can improve the quality of received signals, thus improving the
measurement range or measurement accuracy. The BFSK method controls the frequency
of transmitted signals through binary coding and the accuracy of the method can reach
0.05 mm within 5 m. The AM method modulates the amplitude of the transmitted signals
and the accuracy can reach 0.02 mm within 0.5 m. The signal coding method can suppress
crosstalk between multiple sensors and improve the measurement rate of long-distance
(10 m) measurement.
Finally, the ranging error and compensation methods are summarized. The ranging
error includes the error of ToF estimation and the error of sound velocity. The error of
ToF is compensated for by signal filtering, gain compensation of ultrasonic propagation
attenuation and compensation algorithm of different measurement angles. The sound
velocity error can be compensated for by setting an environmental parameter sensor
or a known exact distance compensation module to improve the performance of the
ranging system.
The application of ultrasonic ranging is increasingly developing due to its unique
characteristics such as low hardware requirements, not being sensitive to ambient light and
electromagnetic interference, and the low cost of transducers. With the continuous progress
of ultrasonic ranging technology in transducers, measurement system configuration, mea-
surement methods and signal processing, error compensation methods, we predict that the
performance of ultrasonic ranging will present the following development trends:
Adoption of MUTs. MUTs will be more widely used in ultrasonic ranging systems with
their advantages, such as small size, low power consumption, low cost, mass production
and integration with other electronics.
Optimization of the transducer performance. At present, the electromechanical coupling
coefficient of PMUT is relatively low. The increase in the electromechanical coupling
coefficient will reduce the power consumption of the system and improve the measurement
range. In addition, the improvement of the transducer bandwidth will improve the working
frequency range of the transducer and make its signal modulation more flexible, thus
improving the overall performance of the ranging system.
Improvement of the measurement range. Performance improvements include the mini-
mum detectable range and the maximum detectable range. At present, the blind area of
ultrasonic ranging is large, which is at the level of centimeters. Optimization of devices
and hardware circuits to reduce the blind area whilst extending the measurement range
at the far end is the direction of future development. In addition, multi-sensor arrays and
optimization of the transducer mentioned above will increase the maximum detectable
distance and direction angle.
Development of the processing algorithm. The future development direction is multi-
algorithm fusion processing of ultrasonic received signals so that the measurement accuracy
and processing time can reach a balance with a higher comprehensive level.
Micromachines 2022, 13, 520 27 of 33

Applications in high dynamic and complex multi-target measurement. At present, the

number of detectable targets is limited and ultrasonic ranging is mostly used in the static or
low dynamic measurement of simple targets, but in the future, high dynamic and complex
multi-target measurement problems will be solved through better algorithms.
Improvement of ultrasonic measurement rate. At present, there are few studies in this
area. The method to improve the measurement rate is to encode the ultrasonic signal and
realize the continuous transmission of the ultrasonic signal to overcome the limitation of
the slow propagation speed of the ultrasonic signal in the long-distance measurement. On
the other hand, the research of new coding technology will also make the comprehensive
performance of the coded ultrasonic more superior, improve the measurement rate, to be
more suitable for high-speed measurement requirements.
In short, the performance of ultrasonic ranging technology will be further improved,
and the application range will be wider in the further.

Author Contributions: Conceptualization, Z.Q. (Zurong Qiu), Y.L. and Z.Q. (Zhen Qiu); formal
analysis, Y.L. and Z.Q. (Zhen Qiu); investigation, Y.L.; writing—original draft preparation, Y.L.;
writing—review and editing, Z.Q. (Zhen Qiu); supervision, Z.Q. (Zurong Qiu); project administration,
Z.Q. (Zurong Qiu); funding acquisition, Z.Q. (Zurong Qiu). All authors have read and agreed to the
published version of the manuscript.
Funding: This research was funded by National Natural Science Foundation of China grant number
52175514.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Ashhar, K.; Noor-A-Rahim, M.; Khyam, M.O.; Soh, C.B. A Narrowband Ultrasonic Ranging Method for Multiple Moving Sensor
Nodes. IEEE Sens. J. 2019, 19, 6289–6297. [CrossRef]
2. Wang, J.; Kong, X.; Cheng, G.; Chen, Y. Research on Detection Range Prediction for Oversea Wide-aperture Towed Sonar. In
Proceedings of the 2021 OES China Ocean Acoustics (COA 2021), Harbin, China, 14–17 July 2021. [CrossRef]
3. Lee, W.; Roh, Y. Ultrasonic transducers for medical diagnostic imaging. Biomed. Eng. Lett. 2017, 7, 91–97. [CrossRef] [PubMed]
4. Sheng, J.; Zhao, Y.; Qiao, Y.; Ling, J.; Fu, J.; Yang, Y.; Ren, T. The manufacture and characterization of a novel ultrasonic transducer
for medical imaging. In Proceedings of the 5th IEEE Electron Devices Technology and Manufacturing Conference (EDTM),
Chengdu, China, 8–11 April 2021. [CrossRef]
5. Guangzhen, X.; Volker, W.; Ping, Y. Review of field characterization techniques for high intensity therapeutic ultrasound.
Metrologia 2021, 58, 22001. [CrossRef]
6. Park, S.H.; Choi, S.; Jhang, K.Y. Porosity Evaluation of Additively Manufactured Components Using Deep Learning-based
Ultrasonic Nondestructive Testing. Int. J. Precis. Eng. Manuf. Green Technol. 2021, 9, 395–407. [CrossRef]
7. Pan, Q.; Pan, R.; Shao, C.; Chang, M.; Xu, X. Research Review of Principles and Methods for Ultrasonic Measurement of Axial
Stress in Bolts. Chin. J. Mech. Eng. Engl. Ed. 2020, 33, 11. [CrossRef]
8. Kumar, S.J.; Kamaraj, A.; Sundaram, K.C.; Shobana, G.; Kirubakaran, G. A comprehensive review on accuracy in ultrasonic flow
measurement using reconfigurable systems and deep learning approaches. AIP Adv. 2020, 10, 105221. [CrossRef]
9. Fang, Z.; Su, R.; Hu, L.; Fu, X. A simple and easy-implemented time-of-flight determination method for liquid ultrasonic flow
meters based on ultrasonic signal onset detection and multiple-zero-crossing technique. Measurement 2021, 168, 108398. [CrossRef]
10. Fu, D.; Zhao, Z. Moving Object Tracking Method Based on Ultrasonic Automatic Detection Algorithm. In Proceedings of the 6th
International Conference on Information Science and Technology (ICIST), Dalian, China, 6–8 May 2016. [CrossRef]
11. Allevato, G.; Hinrichs, J.; Rutsch, M.; Adler, J.P.; Jager, A.; Pesavento, M.; Kupnik, M. Real-Time 3-D Imaging Using an Air-Coupled
Ultrasonic Phased-Array. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2021, 68, 796–806. [CrossRef]
12. Carotenuto, R.; Merenda, M.; Iero, D.; Della Corte, F.G. Mobile Synchronization Recovery for Ultrasonic Indoor Positioning.
Sensors 2020, 20, 702. [CrossRef]
13. Patkar, A.R.; Tasgaonkar, P.P. Object Recognition Using Horizontal Array of Ultrasonic Sensors. In Proceedings of the IEEE
International Conference on Communication and Signal Processing (ICCSP), Melmaruvathur, Tamilnadu, India, 6–8 April 2016.
[CrossRef]
14. Xia, M.; Xiu, C.; Yang, D.; Wang, L. Performance Enhancement of Pedestrian Navigation Systems Based on Low-Cost Foot-
Mounted MEMS-IMU/Ultrasonic Sensor. Sensors 2019, 19, 364. [CrossRef]
15. Ge, R.; Aa, G. Ultrasonic ranging in air. Meas. Tech. 1969, 12, 1675–1678.
16. Hoffstatter, G. Using the polaroid ultrasonic ranging system. Rob. Age 1984, 6, 35–37.
17. Jaffe, D.L. Polaroid ultrasonic ranging sensors in robotic applications. Rob. Age 1985, 7, 23–25, 27–30.
Micromachines 2022, 13, 520 28 of 33

18. Borenstein, J.; Koren, Y. Obstacle avoidance with ultrasonic with ultrasonic sensors. IEEE J. Rob. Autom. 1988, 4, 213–218.
[CrossRef]
19. Fiorillo, A.S.; Allotta, B.; Dario, P.; Francesconi, R. Ultrasonic range sensor array for a robotic fingertip. Sens. Actuators 1989, 17,
103–106. [CrossRef]
20. Yang, M.; Hill, S.L.; Gray, J.O. Design of ultrasonic linear array system for multi-object identification. IEEE Int. Conf. Intell. Rob.
Syst. 1992, 3, 1625–1632. [CrossRef]
21. Tanaka, S.; Jiang, J.; Takesue, T. A model-based adaptive algorithm for determination of time-of-flight in ultrasonic measurement.
In Proceedings of the 1997 36th SICE Annual Conference, Tokushima, Japan, 29–31 July 1997. [CrossRef]
22. Tardajos, G.; Gonzalez Gaitano, G.; Montero De Espinosa, F.R. Accurate, sensitive, and fully automatic method to measure sound
velocity and attenuation. Rev. Sci. Instrum. 1994, 65, 2933–2938. [CrossRef]
23. Huang, C.F.; Young, M.S.; Li, Y.C. Multiple-frequency continuous wave ultrasonic system for accurate distance measurement.
Rev. Sci. Instrum. 1999, 70, 1452–1458. [CrossRef]
24. Webster, D. A pulsed ultrasonic distance measurement system based upon phase digitizing. IEEE Trans. Instrum. Meas. 1994, 43,
578–582. [CrossRef]
25. Yang, M.; Hill, S.L.; Bury, B.; Gray, J.O. A multifrequency AM-based ultrasonic system for accuracy distance measurement. IEEE
Trans. Instrum. Meas. 1994, 43, 861–866. [CrossRef]
26. Barshan, B. Fast processing techniques for accurate ultrasonic range measurements. Meas. Sci. Technol. 2000, 11, 45–50. [CrossRef]
27. Carullo, A.; Parvis, M. An ultrasonic sensor for distance measurement in automotive applications. IEEE Sens. J. 2001, 1, 143–147.
[CrossRef]
28. Al-Smadi, A.M.; Al-Ksasbeh, W.; Ababneh, M.; Al-Nsairat, M. Intelligent Automobile Collision Avoidance and Safety System. In
Proceedings of the 17th International Multi-Conference on Systems, Signals and Devices (SSD), Sfax, Tunisia, 20–23 July 2020.
[CrossRef]
29. Jiménez, F.; Naranjo, J.; Gómez, O.; Anaya, J. Vehicle Tracking for an Evasive Manoeuvres Assistant Using Low-Cost Ultrasonic
Sensors. Sensors 2014, 14, 22689–22705. [CrossRef] [PubMed]
30. Kim, M.; Park, J.; Choi, S. Road Type Identification Ahead of the Tire Using D-CNN and Reflected Ultrasonic Signals. Int. J.
Automot. Technol. 2021, 22, 47–54. [CrossRef]
31. Singh, J.; Dhuheir, M.; Refaey, A.; Erbad, A.; Mohamed, A.; Guizani, M. Navigation and Obstacle Avoidance System in Unknown
Environment. In Proceedings of the IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), London, ON,
Canada, 30 August–2 September 2020. [CrossRef]
32. Anis, H.; Fadhillah, A.H.I.; Darma, S.; Soekirno, S. Automatic Quadcopter Control Avoiding Obstacle Using Camera with
Integrated Ultrasonic Sensor. In Proceedings of the International Conference on Theoretical and Applied Physics (ICTAP),
Yogyakarta, Indonesia, 6–8 September 2017. [CrossRef]
33. Jianwei, Z.; Jianhua, F.; Shouzhong, W.; Kun, W.; Chengxiang, L.; Tao, H. Obstacle Avoidance of Multi-Sensor Intelligent Robot
Based on Road Sign Detection. Sensors 2021, 21, 6777. [CrossRef]
34. Hamanaka, M.; Nakano, F. Surface-Condition Detection System of Drone-Landing Space using Ultrasonic Waves and Deep Learn-
ing. In Proceedings of the International Conference on Unmanned Aircraft Systems (ICUAS), Athens, Greece, 1–4 September 2020.
[CrossRef]
35. Hsu, C.; Chen, H.; Lai, C. An Improved Ultrasonic-Based Localization Using Reflection Method. In Proceedings of the Interna-
tional Asia Conference on Informatics in Control, Automation and Robotics, Bangkok, Thailand, 1 February–2 September 2009.
[CrossRef]
36. Kim, S.J.; Kim, B.K. Dynamic localization based on EKF for indoor mobile robots using discontinuous ultrasonic distance
measurements. In Proceedings of the International Conference on Control, Automation and Systems (ICCAS), Gyeonggi do,
South Korea, 27–30 October 2010. [CrossRef]
37. Ijaz, F.; Yang, H.K.; Ahmad, A.W.; Lee, C. Indoor Positioning: A Review of Indoor Ultrasonic Positioning systems. In Proceed-
ings of the 15th International Conference on Advanced Communications Technology (ICACT), PyeongChang, Korea, 27–30
January 2013.
38. Wang, R.; Chen, L.; Wang, J.; Zhang, P.; Tan, Q.; Pan, D. Research on autonomous navigation of mobile robot based on multi
ultrasonic sensor fusion. In Proceedings of the IEEE 4th Information Technology and Mechatronics Engineering Conference
(ITOEC), Chongqing, China, 14–16 December 2018. [CrossRef]
39. Ju, X.T.; Gu, L.C. Long-Distance Ultrasonic Ranging System Oriented to Tower Crane Anti-Collision. In Proceedings of the 2nd
International Conference on Machine Design and Manufacturing Engineering (ICMDME), Jeju Island, Korea, 1–2 May 2013.
[CrossRef]
40. Przybyla, R.J.; Shelton, S.E.; Guedes, A.; Izyumin, I.I.; Kline, M.H.; Horsley, D.A.; Boser, B.E. In-Air Rangefinding with an AlN
Piezoelectric Micromachined Ultrasound Transducer. IEEE Sens. J. 2011, 11, 2690–2697. [CrossRef]
41. Saad, M.; Bleakley, C.J.; Nigram, V.; Kettle, P. Ultrasonic hand gesture recognition for mobile devices. J. Multimodal User Interfaces
2018, 12, 31–39. [CrossRef]
42. Wang, Z.; Hou, Y.; Jiang, K.; Dou, W.; Zhang, C.; Huang, Z.; Guo, Y. Hand Gesture Recognition Based on Active Ultrasonic
Sensing of Smartphone: A Survey. IEEE Access 2019, 7, 111897–111922. [CrossRef]
Micromachines 2022, 13, 520 29 of 33

43. Ling, K.; Dai, H.; Liu, Y.; Liu, A.X.; Wang, W.; Gu, Q. UltraGesture: Fine-Grained Gesture Sensing and Recognition. IEEE T. Mobile
Comput. 2020. [CrossRef]
44. Wang, T.; Lee, C. Zero-Bending Piezoelectric Micromachined Ultrasonic Transducer (pMUT) with Enhanced Transmitting
Performance. J. Microelectromech. Syst. 2015, 24, 2083–2091. [CrossRef]
45. Yano, T.; Tone, M.; Fukumoto, A. Range Finding and Surface Characterization Using High-Frequency Air Transducers. IEEE
Trans. Ultrason. Ferroelectr. Freq. Control 1987, 34, 232–236. [CrossRef] [PubMed]
46. Toda, M. New type of matching layer for air-coupled ultrasonic transducers. IEEE Trans. Ultrason. Ferr. 2002, 49, 972–979.
[CrossRef] [PubMed]
47. Alvarez-Arenas, T. Acoustic impedance matching of piezoelectric transducers to the air. IEEE Trans. Ultrason. Ferr. 2004, 51,
624–633. [CrossRef]
48. Kelly, S.; Hayward, G.; Alvarez-Arenas, T. Characterization and assessment of an integrated matching layer for air-coupled
ultrasonic applications. IEEE Trans. Ultrason. Ferr. 2004, 51, 1314–1323. [CrossRef]
49. Alvarez-Arenas, T. Air-Coupled Piezoelectric Transducers with Active Polypropylene Foam Matching Layers. Sensors 2013, 13,
5996–6013. [CrossRef]
50. Ealo, J.L.; Seco, F.; Prieto, C.; Jimenez, A.R.; Roa, J.; Koutsou, A.; Guevara, J. Customizable Field Airborne Ultrasonic Transducers
based on Electromechanical Film. In Proceedings of the IEEE Ultrasonics Symposium, Beijing, China, 2–5 November 2008.
[CrossRef]
51. Dausch, D.E.; Castellucci, J.B.; Chou, D.R.; von Ramm, O.T. Theory and operation of 2-D array piezoelectric micromachined
ultrasound transducers. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2008, 55, 2484–2492. [CrossRef]
52. Brenner, K.; Ergun, A.; Firouzi, K.; Rasmussen, M.; Stedman, Q.; Khuri Yakub, B. Advances in Capacitive Micromachined
Ultrasonic Transducers. Micromachines 2019, 10, 152. [CrossRef]
53. Qiu, Y.; Gigliotti, J.V.; Wallace, M.; Griggio, F.; Demore, C.E.M.; Cochran, S.; Trolier-Mckinstry, S. Piezoelectric Micromachined
Ultrasound Transducer (PMUT) Arrays for Integrated Sensing, Actuation and Imaging. Sensors 2015, 15, 8020–8041. [CrossRef]
54. Yaralioglu, G.G.; Ergun, A.S.; Bayram, B.; Haeggstrom, E.; Khuri-Yakub, B.T. Calculation and measurement of electromechanical
coupling coefficient of capacitive micromachined ultrasonic transducers. EEE Trans. Ultrason. Ferroelectr. Freq. Control 2003, 50,
449–456. [CrossRef]
55. Oralkan, O.; Ergun, A.S.; Johnson, J.A.; Karaman, M.; Demirci, U.; Kaviani, K.; Lee, T.H.; Khuri-Yakub, B.T. Capacitive
micromachined ultrasonic transducers: Next-generation arrays for acoustic imaging? IEEE Trans. Ultrason. Ferroelectr. Freq.
Control 2002, 49, 1596–1610. [CrossRef]
56. Wygant, I.; Kupnik, M.; Windsor, J.; Wright, W.; Khuri-Yakub, B.T. 50 kHz capacitive micromachined ultrasonic transducers for
generation of highly directional sound with parametric arrays. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2009, 56, 193–203.
[CrossRef] [PubMed]
57. Paul, M.; Nicolas, L.; Jacek, B.; Abdoighaffar, B.; Sandrine, G.; Brahim, B.; Sylvain, P.; Alain, B.; Nava, S. Piezoelectric Microma-
chined Ultrasonic Transducers Based on PZT Thin Films. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2005, 52, 2276–2288.
58. Robichaud, A.; Deslandes, D.; Cicek, P.; Nabki, F. A System in Package Based on a Piezoelectric Micromachined Ultrasonic
Transducer Matrix for Ranging Applications. Sensors 2021, 21, 2590. [CrossRef] [PubMed]
59. Kaajakari, V.; Mattila, T.; Oja, A.; Seppa, H. Nonlinear limits for single-crystal silicon microresonators. J. Microelectromech. Syst.
2004, 13, 715–724. [CrossRef]
60. Przybyla, R.J.; Shelton, S.E.; Guedes, A.; Krigel, R.; Horsley, D.A.; Boser, B.E. In-air ultrasonic rangefinding and angle estimation
using an array of aln micromachined transducers. In Proceedings of the Solid-State Sensors, Actuators and Microsystems
Workshop, Hilton Head, SC, USA, 3–7 June 2012. [CrossRef]
61. Jung, J.; Lee, W.; Kang, W.; Shin, E.; Ryu, J.; Choi, H. Review of piezoelectric micromachined ultrasonic transducers and their
applications. J. Micromech. Microeng. 2017, 27, 113001. [CrossRef]
62. Przybyla, R.J.; Tang, H.; Guedes, A.; Shelton, S.E.; Horsley, D.A.; Boser, B.E. 3D Ultrasonic Rangefinder on a Chip. IEEE J. Solid-St.
Circ. 2015, 50, 320–334. [CrossRef]
63. Przybyla, R.; Flynn, A.; Jain, V.; Shelton, S.; Guedes, A.; Izyumin, I.; Horsley, D.; Boser, B. A micromechanical ultrasonic distance
sensor with >1 m range. In Proceedings of the 16th International Solid-State Sensors, Actuators and Microsystems Conference,
Beijing, China, 5–9 June 2011. [CrossRef]
64. Li, H.; Lv, J.; Li, D.; Xiong, C.; Zhang, Y.; Yu, Y. MEMS-on-fiber ultrasonic sensor with two resonant frequencies for partial
discharges detection. Opt. Express 2020, 28, 18431–18439. [CrossRef]
65. Rocchi, A.; Santecchia, E.; Ciciulla, F.; Mengucci, P.; Barucca, G. Characterization and Optimization of Level Measurement by an
Ultrasonic Sensor System. IEEE Sens. J. 2019, 19, 3077–3084. [CrossRef]
66. Zhang, J.; Long, Z.; Ma, W.; Hu, G.; Li, Y. Electromechanical Dynamics Model of Ultrasonic Transducer in Ultrasonic Machining
Based on Equivalent Circuit Approach. Sensors 2019, 19, 1405. [CrossRef]
67. Zhou, H.; Huang, S.H.; Li, W. Electrical Impedance Matching between Piezoelectric Transducer and Power Amplifier. IEEE Sens.
J. 2020, 20, 14273–14281. [CrossRef]
68. Garcia-Rodriguez, M.; Garcia-Alvarez, J.; Yañez, Y.; Garcia-Hernandez, M.J.; Salazar, J.; Turo, A.; Chavez, J.A. Low cost matching
network for ultrasonic transducers. Phys. Procedia 2010, 3, 1025–1031. [CrossRef]
Micromachines 2022, 13, 520 30 of 33

69. Rathod, V.T. A Review of Electric Impedance Matching Techniques for Piezoelectric Sensors, Actuators and Transducers.
Electronics 2019, 8, 169. [CrossRef]
70. Ling, J.; Chen, Y.; Chen, Y.; Wang, D.; Zhao, Y.; Pang, Y.; Yang, Y.; Ren, T. Design and Characterization of High-Density Ultrasonic
Transducer Array. IEEE Sens. J. 2018, 18, 2285–2290. [CrossRef]
71. Arman, H.; Dimitre, L.; Deane, D.G.; Azadeh, H.; Darren, I.; Marc, T.; Martin, S. Three-dimensional micro electromechanical
system piezoelectric ultrasound transducer. Appl. Phys. Lett. 2013, 101, 253101. [CrossRef]
72. Kang, L.; Feeney, A.; Dixon, S. The High Frequency Flexural Ultrasonic Transducer for Transmitting and Receiving Ultrasound in
Air. IEEE Sens. J. 2020, 20, 7653–7660. [CrossRef]
73. Wang, H.; Wang, X.; He, C.; Xue, C. Investigation and analysis of the influence of excitation signal on radiation characteristics of
capacitive micromachined ultrasonic transducer. Microsyst. Technol. 2018, 24, 2999–3018. [CrossRef]
74. Pop, F.; Herrera, B.; Cassella, C.; Rinaldi, M. Modeling and Optimization of Directly Modulated Piezoelectric Micromachined
Ultrasonic Transducers. Sensors 2021, 21, 157. [CrossRef]
75. Krishnakumar, R.; Ramesh, R. Enhancing the Performance Characteristics of Piezoelectric Transducers by Broadband Tuning.
IEEE Sens. J. 2021, 21, 21667–21674. [CrossRef]
76. Blackstock, D.T. Fundamentals of Physical Acoustics; John Wiley and Sons: Hoboken, NJ, USA, 2000.
77. Tiwari, K.; Raisutis, R.; Mazeika, L.; Samaitis, V. 2D Analytical Model for the Directivity Prediction of Ultrasonic Contact Type
Transducers in the Generation of Guided Waves. Sensors 2018, 18, 987. [CrossRef]
78. Okamoto, K.; Okubo, K. Development of omnidirectional audible sound source using facing ultrasonic transducer arrays driven
at different frequencies. Jpn. J. Appl. Phys. 2021, 60, SDDD17. [CrossRef]
79. Wang, H.; Wang, X.; He, C.; Xue, C. Design and Performance Analysis of Capacitive Micromachined Ultrasonic Transducer Linear
Array. Micromachines 2014, 5, 420–431. [CrossRef]
80. Xiang, R.; Shi, Z. Design of Millimeter Range High Precision Ultrasonic Distance Measurement System. In Proceedings of
the International Conference on Computer Systems, Electronics and Control (ICCSEC), Dalian, China, 25–27 December 2017.
[CrossRef]
81. Fu, X.C.; Zhou, L.; Wen, G.J. Ultrasonic Ranging System Based on Single Chip Microprocessor. Appl. Mech. Mater. 2013, 441,
360–363. [CrossRef]
82. Zhao, X.; Qian, P.; Lu, N.; Li, Y. Design and Experimental Study of High Precision Ultrasonic Ranging System. In Proceedings of the
5th International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS), Naha, Japan, 18–20 November 2020.
[CrossRef]
83. Xiao, Z.H.; Wu, S.Y.; An, Q.Y. Design of Ultrasonic Distance Measurement System Based on Microcontroller. Appl. Mech. Mater.
2013, 333–335, 296–299. [CrossRef]
84. Tai, H.; Zhang, H. The hardware research of ultrasonic ranging system based on variable emission wavelength. In Proceedings
of the IEEE 4th Advanced Information Technology, Electronic and Automation Control Conference (IAEAC), Chengdu, China,
20–22 December 2019. [CrossRef]
85. Cabral, E.A.V.; Valdez, I. Airborne ultrasonic sensor node for distance measurement. In Proceedings of the 12th IEEE Sensors
Conference, Baltimore, MD, USA, 3–6 November 2013. [CrossRef]
86. Zhi, S.; Yi, Y.; Wang, Z. Design of Distance Measuring and Reversing System. In Proceedings of the IEEE 3rd Advanced
Information Technology, Electronic and Automation Control Conference (IAEAC), Chongqing, China, 12–14 October 2018.
[CrossRef]
87. Lu, H.; Li, Z.; Gao, P. Design of Ultrasonic Ranging System Based on Cross-correlation Method. In Proceedings of the 39th
Chinese Control Conference (CCC), Shenyang, China, 27–29 July 2020. [CrossRef]
88. Jiang, Y.; Yuan, M. The study of improving ultrasonic ranging accuracy based on the double closed-loop control technology. In
Proceedings of the Chinese Automation Congress (CAC), Jinan, China, 20–22 October 2017. [CrossRef]
89. Gao, Y.; Jia, L.N.; Wang, B.; Liu, L.H.; Huang, L.M. High Precision Ultrasonic Ranging System for Mobile Robot Navigation. Appl.
Mech. Mater. 2012, 249–250, 1139–1143. [CrossRef]
90. Wu, J.; Zhu, J.; Yang, L.; Shen, M.; Xue, B.; Liu, Z. A highly accurate ultrasonic ranging method based on onset extraction and
phase shift detection. Measurement 2014, 47, 433–441. [CrossRef]
91. Sahoo, A.K.; Udgata, S.K. A Novel ANN-Based Adaptive Ultrasonic Measurement System for Accurate Water Level Monitoring.
IEEE Trans. Instrum. Meas. 2019, 69, 3359–3369. [CrossRef]
92. Yue, H.; Zhang, X.; Shi, Z. Simulation and Implementation of an Ultrasonic Ranging Experiment System. In Proceedings of the
IEEE 2nd International Conference on Computer Science and Educational Informatization (CSEI), Xinxiang, China, 17 July 2020.
[CrossRef]
93. Wenhuan, H. Software implementation of a wireless ultrasonic ranging system. In Proceedings of the 12th IEEE International
Conference on Electronic Measurement & Instruments (ICEMI), Qingdao, China, 16–18 July 2015. [CrossRef]
94. Zhang, L.; Zhao, L. Research of ultrasonic distance measurement system based on DSP. In Proceedings of the International
Conference on Computer Science and Service System (CSSS), Nanjing, China, 27–29 June 2011. [CrossRef]
95. Li, Q.; Huang, Z.; Zhu, Z. Design of High-precision Ultrasonic Ranging System for Mobile Robot. In Proceedings of the 2021 IEEE
4th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC), Chongqing,
China, 18–20 June 2021. [CrossRef]
Micromachines 2022, 13, 520 31 of 33

96. Tai, H.L.; Zhang, H.; Yang, S. Study on software part of ultrasonic ranging system based on variable emission wavelength. In
Proceedings of the 6th International Conference on Computer Science and Network Technology (ICCSNT), Dalian, China, 21–22
October 2017. [CrossRef]
97. Yuan, S.; Mao, H.; Guo, S.; Zhang, F.; Yao, X. Research on the DSmT Based Model of the Ultrasonic Sensor Detection for
Indoor Environment Contour. In Proceedings of the 35th Chinese Control Conference (CCC), Chengdu, China, 27–29 July 2016.
[CrossRef]
98. Rozen, O.; Block, S.T.; Mo, X.; Bland, W.; Hurst, P.; Tsai, J.M.; Daneman, M.; Amirtharajah, R.; Horsley, D.A. Monolithic MEMS-
CMOS ultrasonic rangefinder based on dual-electrode PMUTs. In Proceedings of the IEEE 29th International Conference on
Micro Electro Mechanical Systems, Shanghai, China, 24–28 January 2016; pp. 3336–3339. [CrossRef]
99. Gluck, T.; Kravchik, M.; Chocron, S.; Elovici, Y.; Shabtai, A. Spoofing Attack on Ultrasonic Distance Sensors Using a Continuous
Signal. Sensors 2020, 20, 6157. [CrossRef]
100. Li, X.; Wu, R.; Sheplak, M.; Li, J. Multifrequency CW-based time-delay estimation for proximity ultrasonic sensors. IEE P-Radar
Son. Nav. 2002, 149, 53–59. [CrossRef]
101. Lagler, D.; Anzinger, S.; Pfann, E.; Fusco, A.; Bretthauer, C.; Huemer, M. A Single Ultrasonic Transducer Fast and Robust
Short-Range Distance Measurement Method. In Proceedings of the IEEE International Ultrasonics Symposium (IUS), Glasgow,
England, 6–9 October 2019. [CrossRef]
102. Tian, Y.; Sun, X.Y.; Chen, J. Design Principle and Error Analysis of a New Large-Range Ultrasonic Position and Orientation System
Based on TDOA. In Proceedings of the International Conference on Advanced Materials and Engineering Materials (ICAMEM),
Shenyang, China, 22–24 November 2012. [CrossRef]
103. Zivkovic, D.S.; Markovic, B.R.; Rakic, D.; Tadic, S. Design considerations and performance of low-cost ultrasonic ranging system.
In Proceedings of the 10th Workshop on Positioning, Navigation and Communication (WPNC), Dresden, Germany, 18 June 2013.
[CrossRef]
104. Patole, S.M.; Torlak, M.; Wang, D.; Ali, M. Automotive radars: A review of signal processing techniques. IEEE Signal Proc. Mag.
2017, 34, 22–35. [CrossRef]
105. Gan, T.H.; Hutchins, D.A.; Green, R.J. A swept frequency multiplication technique for air-coupled ultrasonic NDE. IEEE Trans.
Ultrason. Ferr. 2004, 51, 1271–1279. [CrossRef]
106. Suzuki, K.; Endo, M.; Ishikawa, M.; Nishino, H. Air-coupled ultrasonic vertical reflection method using pulse compression and
various window functions: Feasibility study. Jpn. J. Appl. Phys. 2019, 58, SGGB09. [CrossRef]
107. Baksheeva, I.V.; Khomenko, A.A. Evaluation of the Ultrasonic Medical Scanners Range Resolution in Real Biological Tissue and
Ways of Its Improvement. In Proceedings of the Wave Electronics and its Application in Information and Telecommunication
Systems (WECONF), St. Petersburg, Russia, 3–7 June 2019. [CrossRef]
108. Devaud, F.; Haward, G.; Soraghan, J.J. The use of chirp overlapping properties for improved target resolution in an ultrasonic
ranging system. In Proceedings of the IEEE Ultrasonics Symposium, Montreal, QC, Canada, 23–27 August 2004. [CrossRef]
109. Lu, L.; Ye, W.; Tao, J. The Research of Amplitude Threshold Method in Ultrasound- based Indoor Distance-Measurement System.
In Proceedings of the 4th International Conference on Electronic Information Technology and Computer Engineering (EITCE),
Virtual, Online, China, 6–8 November 2020. [CrossRef]
110. Chiu, Y.; Wang, C.; Gong, D.; Li, N.; Ma, S.; Jin, Y. A Novel Ultrasonic TOF Ranging System Using AlN Based PMUTs.
Micromachines 2021, 12, 284. [CrossRef] [PubMed]
111. Cai, G.; Zhou, X.; Yi, Y.; Zhou, H.; Li, D.; Zhang, J.; Huang, H.; Mu, X. An Enhanced-Differential PMUT for Ultra-Long Distance
Measurement in Air. In Proceedings of the 34th IEEE International Conference on Micro Electro Mechanical Systems (MEMS),
Virtual, Gainesville, FL, USA, 25–29 January 2021. [CrossRef]
112. Naba, A.; Khoironi, M.F.; Santjojo, D.D.H. Low Cost but Accurate Ultrasonic Distance Measurement Using Combined Method of
Threshold Correlation. In Proceedings of the 14th International Conference on Quality in Research (QiR), Lombok, Indonesia,
10–13 August 2015. [CrossRef]
113. Li, W.; Chen, Q.; Wu, J. Double threshold ultrasonic distance measurement technique and its application. Rev. Sci. Instrum. 2014,
85, 44905. [CrossRef] [PubMed]
114. Choe, I.; Lee, K.; Choy, I.; Cho, W. Ultrasonic Distance Measurement Method by Using the Envelope Model of Received Signal
Based on System Dynamic Model of Ultrasonic Transducers. J. Electr. Eng. Technol. 2018, 13, 981–988. [CrossRef]
115. Khyam, M.O.; Ge, S.S.; Li, X.; Pickering, M.R. Highly Accurate Time-of-Flight Measurement Technique Based on Phase-Correlation
for Ultrasonic Ranging. IEEE Sens. J. 2017, 17, 434–443. [CrossRef]
116. Carotenuto, R.; Merenda, M.; Iero, D.; Della Corte, F.G. Simulating Signal Aberration and Ranging Error for Ultrasonic Indoor
Positioning. Sensors 2020, 20, 3548. [CrossRef]
117. Carotenuto, R.; Pezzimenti, F.; Corte, F.G.D.; Iero, D.; Merenda, M. Acoustic Simulation for Performance Evaluation of Ultrasonic
Ranging Systems. Electronics 2021, 10, 1298. [CrossRef]
118. Przybyla, R.; Izyumin, I.; Kline, M.; Boser, B.; Shelton, S. An ultrasonic rangefinder based on an AlN piezoelectric micromachined
ultrasound transducer. In Proceedings of the IEEE Sensors Conference, Kona, HI, USA, 1–4 November 2010. [CrossRef]
119. Carotenuto, R.; Merenda, M.; Iero, D.; Corte, F.G.D. Ranging RFID Tags with Ultrasound. IEEE Sens. J. 2018, 18, 2967–2975.
[CrossRef]
Micromachines 2022, 13, 520 32 of 33

120. Queirós, R.; Corrêa Alegria, F.; Silva Girão, P.; Cruz Serra, A. A multi-frequency method for ultrasonic ranging. Ultrasonics 2015,
63, 86–93. [CrossRef]
121. Saad, M.M.; Bleakley, C.J.; Dobson, S. Robust High-Accuracy Ultrasonic Range Measurement System. IEEE Trans. Instrum. Meas.
2011, 60, 3334–3341. [CrossRef]
122. Kredba, J.; Holada, M. Precision ultrasonic range sensor using one piezoelectric transducer with impedance matching and digital
signal processing. In Proceedings of the IEEE International Workshop of Electronics, Control, Measurement, Signals and their
Application to Mechatronics (ECMSM), Donostia-San Sebastian, Spain, 24–26 May 2017. [CrossRef]
123. Krenik, M.; Li, X.; Akin, B. Improved TOF Determination Algorithms for Robust Ultrasonic Positioning of Smart Tools. In
Proceedings of the 40th Annual Conference of the IEEE-Industrial-Electronics-Society (IECON), Dallas, TX, USA, 29 October–1
November 2014. [CrossRef]
124. Assous, S.; Hopper, C.; Lovell, M.; Gunn, D.; Jackson, P.; Rees, J. Short pulse multi-frequency phase-based time delay estimation.
J. Acoust. Soc. Am. 2010, 127, 309–315. [CrossRef] [PubMed]
125. Huang, K.; Huang, Y. Multiple-frequency ultrasonic distance measurement using direct digital frequency synthesizers. Sens.
Actuators A-Phys. 2009, 149, 42–50. [CrossRef]
126. Kimura, T.; Wadaka, S.; Misu, K.; Nagatsuka, T.; Tajime, T.; Koike, M. A high resolution ultrasonic range measurement method
using double frequencies and phase detection. In Proceedings of the 1995 IEEE Ultrasonics Symposium, Seattle, WA, USA, 7–10
November 1995. [CrossRef]
127. Chongchamsai, M.; Sinchai, S.; Wardkein, P.; Boonjun, S. Distance Measurement Technique Using Phase Difference of Two
Reflected Ultrasonic Frequencies. In Proceedings of the 3rd International Conference on Computer and Communication Systems
(ICCCS), Nagoya, Japan, 27–30 April 2018. [CrossRef]
128. Lee, K.; Huang, C.; Huang, S.; Huang, K.; Young, M. A High-Resolution Ultrasonic Distance Measurement System Using Vernier
Caliper Phase Meter. IEEE Trans. Instrum. Meas. 2012, 61, 2924–2931. [CrossRef]
129. Kuratli, C.; Huang, Q. A CMOS Ultrasound Range-Finder Microsystem. In Proceedings of the International Solid-State Circuits
Conference, San Francisco, CA, USA, 7–9 February 2000. [CrossRef]
130. Jiang, S.; Yang, C.; Huang, R.; Fang, C.; Yeh, T. An Innovative Ultrasonic Time-of-Flight Measurement Method Using Peak Time
Sequences of Different Frequencies: Part I. IEEE Trans. Instrum. Meas. 2011, 60, 735–744. [CrossRef]
131. Yang, C.; Jiang, S.; Lin, D.; Lu, F.; Wu, Y.; Yeh, T. An Innovative Ultrasonic Time-of-Flight Measurement Method Using Peak Time
Sequences of Different Frequencies—Part II: Implementation. IEEE Trans. Instrum. Meas. 2011, 60, 745–757. [CrossRef]
132. Chen, X.; Xu, J.; Chen, H.; Ding, H.; Xie, J. High-Accuracy Ultrasonic Rangefinders via pMUTs Arrays Using Multi-Frequency
Continuous Waves. J. Microelectromech. Syst. 2019, 28, 634–642. [CrossRef]
133. Nakahira, K.; Okuma, S.; Kodama, T.; Furuhashi, T. The use of binary coded frequency shift keyed signals for multiple user sonar
ranging. In Proceedings of the IEEE International Conference on Networking, Sensing and Control, Taipei, Taiwan, China, 21–23
March 2004. [CrossRef]
134. Huang, S.S.; Huang, C.F.; Huang, K.N.; Young, M.S. A high accuracy ultrasonic distance measurement system using binary
frequency shift-keyed signal and phase detection. Rev. Sci. Instrum. 2002, 73, 3671–3677. [CrossRef]
135. Nakahira, K.; Kodama, T.; Furuhashi, T.; Okuma, S. A self-adapting sonar ranging system based on digital polarity correlators.
Meas. Sci. Technol. 2004, 15, 347–352. [CrossRef]
136. Zhenjing, Y.; Li, H.; Yanan, L. Improvement of Measurement Range via Chaotic Binary Frequency Shift Keying Excitation
Sequences for Multichannel Ultrasonic Ranging System. Int. J. Control Autom. 2016, 9, 189–200. [CrossRef]
137. Hua, H.; Wang, Y.; Yan, D. A low-cost dynamic range-finding device based on amplitude-modulated continuous ultrasonic wave.
IEEE Trans. Instrum. Meas. 2002, 51, 362–367. [CrossRef]
138. Sumathi, P.; Janakiraman, P.A. SDFT-Based Ultrasonic Range Finder Using AM Continuous Wave and Online Parameter
Estimation. IEEE Trans. Instrum. Meas. 2010, 59, 1994–2004. [CrossRef]
139. Sasaki, K.; Tsuritani, H.; Tsukamoto, Y.; Iwatsubo, S. Air-coupled ultrasonic time-of-flight measurement system using amplitude-
modulated and phase inverted driving signal for accurate distance measurements. IEICE Electron. Expr. 2009, 6, 1516–1521.
[CrossRef]
140. Huang, Y.P.; Wang, J.S.; Huang, K.N.; Ho, C.T.; Huang, J.D.; Young, M.S. Envelope pulsed ultrasonic distance measurement
system based upon amplitude modulation and phase modulation. Rev. Sci. Instrum. 2007, 78, 65103. [CrossRef]
141. Huang, J.; Lee, C.; Yeh, C.; Wu, W.; Lin, C. High-Precision Ultrasonic Ranging System Platform Based on Peak-Detected
Self-Interference Technique. IEEE Trans. Instrum. Meas. 2011, 60, 3775–3780. [CrossRef]
142. Carotenuto, R. A range estimation system using coded ultrasound. Sens. Actuators A-Phys. 2016, 238, 104–111. [CrossRef]
143. Dou, Z.; Karnaushenko, D.; Schmidt, O.G.; Karnaushenko, D. A High Spatiotemporal Resolution Ultrasonic Ranging Technique
with Multiplexing Capability. IEEE Trans. Instrum. Meas. 2021, 70, 1–12. [CrossRef]
144. Alvarez, F.J.; Aguilera, T.; Fernandez, J.A.; Moreno, J.A.; Gordillo, A. Analysis of the performance of an Ultrasonic Local
Positioning System based on the emission of Kasami codes. In Proceedings of the International Conference on Indoor Positioning
and Indoor Navigation (IPIN), Zurich, Switzerland, 15–17 September 2010. [CrossRef]
145. Pérez, M.C.; Ureña, J.; Hernández, A.; Jiménez, A.; Ruíz, D.; álvarez, F.J.; De Marziani, C. Performance Comparison of Different
Codes in an Ultrasonic Positioning System using DS-CDMA. In Proceedings of the 6th IEEE International Symposium on
Intelligent Signal Processing, Budapest, Hungary, 26–28 August 2009. [CrossRef]
Micromachines 2022, 13, 520 33 of 33

146. Meng, Q.; Yao, F.; Wu, Y. Review of Crosstalk Elimination Methods for Ultrasonic Range Systems in Mobile Robots. In Proceedings
of the IEEE/RSJ International Conference on Intelligent Robots and Systems, Beijing, China, 9–13 October 2006. [CrossRef]
147. Ureña, J.; Mazo, M.; García, J.J.; Hernández, Á.; Bueno, E. Correlation detector based on a FPGA for ultrasonic sensors. Microprocess.
Microsyst. 1999, 23, 25–33. [CrossRef]
148. Hernandez, A.; Urena, J.; Garcia, J.J.; Mazo, M.; Hernanz, D.; Derutin, J.P.; Serot, J. Ultrasonic ranging sensor using simultaneous
emissions from different transducers. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2004, 51, 1660–1669. [CrossRef]
149. Hernández, Á.; Ureña, J.; Hernanz, D.; García, J.J.; Mazo, M.; Dérutin, J.; Serot, J.; Palazuelos, S.E. Real-time implementation of an
efficient Golay correlator (EGC) applied to ultrasonic sensorial systems. Microprocess. Microsyst. 2003, 27, 397–406. [CrossRef]
150. Yamanaka, K.; Hirata, S.; Hachiya, H. Evaluation of correlation property of linear-frequency-modulated signals coded by
maximum-length sequences. Jpn. J. Appl. Phys. 2016, 55, 7. [CrossRef]
151. Lufinka, O. Multiple-point ultrasonic distance measurement and communication with simulations. In Proceedings of the 24th
Telecommunications Forum (TELFOR), Belgrade, Serbia, 22–23 November 2016. [CrossRef]
152. Meng, Q.; Lan, S.; Yao, Z.; Li, G. Real-Time Noncrosstalk Sonar System by Short Optimized Pulse-Position Modulation Sequences.
IEEE Trans. Instrum. Meas. 2009, 58, 3442–3449. [CrossRef]
153. Yao, Z.; Meng, Q.; Li, G.; Lin, P. Non-Crosstalk Real-Time Ultrasonic Range System with Optimized Chaotic Pulse Position-Width
Modulation Excitation. In Proceedings of the IEEE Ultrasonics Symposium, Beijing, China, 2–5 November 2008. [CrossRef]
154. Shin, S.; Kim, M.; Choi, S.B. Improving Efficiency of Ultrasonic Distance Sensors using Pulse Interval Modulation. In Proceedings
of the 15th IEEE Sensors Conference, Orlando, FL, USA, 30 October–2 November 2016. [CrossRef]
155. Fortuna, L.; Frasca, M.; Rizzo, A. Chaotic pulse position modulation to improve the efficiency of sonar sensors. IEEE Trans.
Instrum. Meas. 2003, 52, 1809–1814. [CrossRef]
156. Shin, S.; Kim, M.; Choi, S.B. Ultrasonic Distance Measurement Method with Crosstalk Rejection at High Measurement Rate. IEEE
Trans. Instrum. Meas. 2019, 68, 972–979. [CrossRef]
157. Bischoff, O.; Wang, X.; Heidmann, N.; Laur, R.; Paul, S. Implementation of an ultrasonic distance measuring system with kalman
filtering in wireless sensor networks for transport logistics. In Proceedings of the 24th Eurosensors International Conference,
Linz, Austria, 5–8 September 2010. [CrossRef]
158. Yang, W.; Qing-Hao, M.; Jia-Lin, H.; Pu, L.; Ming, Z. Proportional-Integral-Differential-Based Automatic Gain Control Circuit for
Ultrasonic Ranging Systems. In Proceedings of the 5th International Conference on Measuring Technology and Mechatronics
Automation (ICMTMA), Hong Kong, China, 16–17 January 2013. [CrossRef]
159. Singh, N.A.; Borschbach, M. Effect of external factors on accuracy of distance measurement using ultrasonic sensors. In
Proceedings of the 1st IEEE International Conference on Signals and Systems (ICSigSys), Bali, Indonesia, 16–18 May 2017.
[CrossRef]
160. Yu, S.Q.; Feng, T.; Me, X.X. A High Precision Ultrasonic Ranging Method under Misalignmen ToF Transducer Pairs. J. Xiamen
Univ. Nat. Sci. 2006, 45, 513–517.
161. Nicolau, V.; Aiordachioaie, D.; Andrei, M. Fuzzy system for sound speed estimation in outdoor ultrasonic distance measurements.
In Proceedings of the 4th International Symposium on Computational Intelligence and Intelligent Informatics, Luxor, Egypt,
21–25 October 2009. [CrossRef]
162. Van Schaik, W.; Grooten, M.; Wernaart, T.; van der Geld, C. High Accuracy Acoustic Relative Humidity Measurement inDuct
Flow with Air. Sensors 2010, 10, 7421–7433. [CrossRef]
163. Löfqvist, T.; Sokas, K.; Delsing, J. Speed of sound measurements in gas-mixtures at varying composition using an ultrasonic gas
flow meter with silicon based transducers. In Proceedings of the 11th IMEKO TC9 Conference on Flow Measurement, Groningen,
The Netherlands, 12 May 2003.
164. Wen, Z.Z.; Li, F.N.; Xia, Z.B. High-Precision Ultrasonic Ranging System Design and Research. In Proceedings of the International
Academic Conference on Numbers, Intelligence, Manufacturing Technology and Machinery Automation (MAMT), Wuhan,
China, 24–25 December 2011. [CrossRef]
165. Khoenkaw, P.; Pramokchon, P. A software based method for improving accuracy of ultrasonic range finder module. In Pro-
ceedings of the 2nd Joint International Conference on Digital Arts, Media and Technology (ICDAMT), Chiang Mai, Thailand,
1–4 March 2017. [CrossRef]
166. Shen, M.; Wang, Y.; Jiang, Y.; Ji, H.; Wang, B.; Huang, Z. A New Positioning Method Based on Multiple Ultrasonic Sensors for
Autonomous Mobile Robot. Sensors 2020, 20, 17. [CrossRef]
167. Majchrzak, J.; Michalski, M.; Wiczynski, G. Distance Estimation with a Long-Range Ultrasonic Sensor System. IEEE Sens. J. 2009,
9, 767–773. [CrossRef]
168. Chaparro, L.X.; Contreras, C.R.; Meneses, J.E.; Martins Costa, M.F.P.C. Operating principle of a high resolution ultrasonic ranging
system based in a phase processing. In Proceedings of the 8th Ibero American Optics Meeting/11th Latin American Meeting on
Optics, Lasers, and Applications, Porto, Portugal, 22–26 July 2013. [CrossRef]