The fundamentals of

millimeter wave sensors

Cesar Iovescu
Radar Applications Manager

Sandeep Rao
Radar Systems Architect

Texas Instruments
Introduction
Millimeter wave (mmWave) is a special class of radar technology that uses short-
wavelength electromagnetic waves. Radar systems transmit electromagnetic wave
signals that objects in their path then reflect. By capturing the reflected signal, a radar
system can determine the range, velocity and angle of the objects.

mmWave radars transmit signals with a wavelength that is in the millimeter range. This
is considered a short wavelength in the electromagnetic spectrum and is one of the
advantages of this technology. Indeed, the size of system components such as the
antennas required to process mmWave signals is small. Another advantage of short
wavelengths is the high accuracy. An mmWave system operating at 76–81 GHz (with
a corresponding wavelength of about 4 mm), will have the ability to detect movements
that are as small as a fraction of a millimeter.

A complete mmWave radar system includes transmit (TX) and receive (RX) radio
frequency (RF) components; analog components such as clocking; and digital
components such as analog-to-digital converters (ADCs), microcontrollers (MCUs) and
digital signal processors (DSPs). Traditionally, these systems were implemented with
discrete components, which increased power consumption and overall system cost.
System design is challenging due the complexity and high frequencies.

Texas Instruments (TI) has solved these challenges and designed complementary
metal-oxide semiconductor (CMOS)-based mmWave radar devices that integrate TX-
RF and RX-RF analog components such as clocking, and digital components such
as the ADC, MCU and hardware accelerator. Some families in TI’s mmWave sensor
portfolio integrate a DSP for additional signal-processing capabilities.

TI devices implement a special class of mmWave technology called frequency-

modulated continuous wave (FMCW). As the name implies, FMCW radars transmit a
frequency-modulated signal continuously in order to measure range as well as angle
and velocity. This differs from traditional pulsed-radar systems, which transmit short
pulses periodically.

The fundamentals of millimeter wave sensors 2 May 2017

 Figure 2 shows the same chirp signal, with frequency as a function of time. The chirp is characterized by a
frequency (fc), bandwidth (B) and duration (Tc). In the example provided in Figure 2, fc = 77GHz, B = 4GHz

Figure 2 shows the same chirp signal, with frequency as a function of time. The chirp is characterized by a
frequency (fc), bandwidth (B) and duration (Tc). In the example provided in Figure 2, fc = 77GHz, B = 4GHz

Range measurement 2
TX ant. 1
The fundamental concept in radar systems is the
transmission of an electromagnetic signal that
objects reflect in its path. In the signal used in
3
FMCW radars, the frequency increases linearly Figure 2. Chirp signal, with frequency as a function of time.
Synth
with time. This type of signal is also called radar
a chirp. 4
An FMCW system transmits a chirp signal and captures the signals reflected by objects in its path. Fi
Figure 1 shows a representation represents
of a chirp asignal,
simplified block diagram
Figureof the main RF components of aasFMCW radar.ofThe radar operates as
RX ant.2. Chirp signal, with frequency a function
IF signal time.
with magnitude (amplitude) as a function of time. mixer
FIGURE 1 & 2An FMCW
• A synthesizer
radar system(synth) generates
transmits
Figure 3.aFMCW a radar
chirp chirp.
signal captures the signals reflected by objects in its path. F
anddiagram.
block
• The chirp is transmitted by a transmit antenna (TX ant).
represents a simplified block diagram of the main RF components of a FMCW radar. The radar operates a
A
• The reflection of the chirp by an object generates a reflected chirp captured by the receive antenn
• generates
A synthesizer (synth) generates a chirp.
•• AA“mixer”
synthesizer (synth)
combines the RX and TX a chirp.
signals to produce an intermediate frequency (IF) signal.
• The chirp is transmitted
• Theby achirp
transmit antenna (TXby
is transmitted ant).
a transmit antenna
• The reflection of the chirp by an object generates a reflected chirp captured by the receive anten
(TX ant).
• At “mixer” combines the RX and TX signals to produce an intermediate frequency (IF) signal.
• The reflection of the chirp by an object
generates a reflected chirp captured by the
Figure 1. Chirp signal, with amplitude as a function of time.
f receive antenna (RX ant).
81 GHz
• A “mixer” combines the RX and TX signals to
S
FIGURE 1 & 2
Figure 2 shows the same chirp signal, with
B = 4 GHz
produce an intermediate frequency (IF) signal.
frequency
A as a function of time. The chirp is A frequency mixer is an electronic component that
characterized by a start frequency (fc), bandwidth (B) combines two signals to create a new signal with a
fc = 77 GHz
and duration (Tc). The slope of the chirp (S) captures Figure 3. FMCW radar block diagram.
t new frequency.
T = 40 µs
the rate of change of frequency.
c In the example
A frequency mixer is an electronic component that combines two signals to create a 1new signal with a new
t For two sinusoidal inputs x1 and x2 (Equations 1
provided in Figure 2, fc = 77 GHz, B = 4 GHz, Figure 3. FMCW radar block diagram.
For two sinusoidal inputs x1and 2):
and x2 (Equations 1 and 2):
Tc = 40 µs and S = 100 MHz/µs.
A frequency mixer is an electronic component that combines two signals to create a new signal with a ne
𝑥𝑥1 = sin(𝜔𝜔1 𝑡𝑡 + 𝛷𝛷1 ) (1)(1)
f For two sinusoidal inputs x1 and x2 (Equations 1 and 2):
81 GHz 𝑥𝑥2 = sin(𝜔𝜔2 𝑡𝑡 + 𝛷𝛷2 ) (2)(2)
(1)
S
The output
The output xout has an instantaneous xout 𝑥𝑥has
frequency 1 = an
sin(𝜔𝜔1 𝑡𝑡 + 𝛷𝛷1 )
equalinstantaneous
to the difference frequency
of the instantaneous frequencies o
input sinusoids. The phase equal
B = 4 GHz of the output
to the xdifference
out𝑥𝑥is equal to
of 2the
2 = sin(𝜔𝜔 𝑡𝑡 +difference
the of the(2)
instantaneous
𝛷𝛷2 ) phases of the two input signa
3): frequencies of the two input sinusoids. The phase
The output xout has an instantaneous frequency equal to the difference of the instantaneous frequencies
fc = 77 GHz
input sinusoids. The phaseof ofthe
the output
𝑥𝑥𝑜𝑜𝑜𝑜𝑜𝑜 =xxsin[(𝜔𝜔
output outisisequal
out
equal 𝜔𝜔to
1 − to ) 𝑡𝑡the
2the + difference
(𝛷𝛷 1 − 𝛷𝛷of
difference 2 )] of phases
the the (3)of the two input sign
t
Tc = 40 µs 3): phases of the two input signals (Equation 3):
The operation of the frequency mixer can also be understood graphically by looking 1 at TX and RX chirp fre
Figure 2. Chirp signal, with frequency as a function of time.
representation as a function of time.𝑥𝑥𝑜𝑜𝑜𝑜𝑜𝑜 = sin[(𝜔𝜔1 − 𝜔𝜔2 ) 𝑡𝑡 + (𝛷𝛷1 − 𝛷𝛷2 )] (3) (3)

The operation
The operation of the frequency of the
mixer can also be frequency
understood mixer can also
graphically be at TX and RX chirp fr
by looking
representation as a function of time. graphically by looking at TX and RX
understood
An FMCW radar system transmits a chirp signal and
captures the signals reflected by objects in its path. chirp frequency representation as a function of time.
Figure 3 represents a simplified block diagram of The upper diagram in Figure 4 on the following
the main RF components of an FMCW radar. The page shows TX and RX chirps as a function of time
radar operates as follows: for a single object detected. Notice that the RX chirp
is a time-delay version of the TX chirp.

The fundamentals of millimeter wave sensors 3 May 2017

 The initial phase of the IF signal (Φ0) is the difference between the phase of the TX chirp and the ph
The mixer output signal asThe a magnitude
mixer outputfunction
signalofastime is a sine wave,
a magnitude since
function of ittime
has isa aconstant
sine wave,frequ
si
at the time instant corresponding to the start of the IF signal (i.e., the time instant represented by th
line
The in Figure
initial 4). (Equation
phase 5): initial
of the IF signal
The (Φ0) phase
is the difference between
of the IF signal (Φ0) isthe
thephase of thebetween
difference TX chirp the and the
phase p
gram in Figure 4 shows TX and RX chirps as a function of time for a single object detected.
at the time instant corresponding Notice
to the that the
startcorresponding
of the IF signalto (i.e.,
at the time instant thethe time
start instant
of the represented
IF signal (i.e., theby
timt
me-delay version of the TX chirp. 𝜙𝜙0 = 2𝜋𝜋𝑓𝑓𝑐𝑐 𝜏𝜏 (5)
line in Figure 4). (Equationline
5): in Figure 4). (Equation 5):
The time delay (t) can be mathematically derived as Mathematically, it can be further derived into
(τ) can be mathematically derived as Equation 4: Mathematically, it can be further derived into Equation 6:
Equation 4: Equation 6: 𝜙𝜙0 = 2𝜋𝜋𝑓𝑓𝑐𝑐 𝜏𝜏 (5)𝜙𝜙0 = 2𝜋𝜋𝑓𝑓𝑐𝑐 𝜏𝜏 (5
2𝑑𝑑 4𝜋𝜋𝑑𝑑
𝜏𝜏 = (5) (4) it can be further
Mathematically, derived into =further
𝜙𝜙0be
Equation (6)*
(6)
𝑐𝑐 Mathematically, it can 𝜆𝜆 6: derived into Equation 6:

where object
distance to the detected d is the distance
and c is the to the of
speed detected
light. objectfor an objectInatsummary,
In summary, a distance for
d from the
𝜙𝜙0 radar,
an object= at athe
4𝜋𝜋𝑑𝑑 IF signal
distance d will
from be a=sine
(6)𝜙𝜙 4𝜋𝜋𝑑𝑑wave (Equation 7
(6
𝜆𝜆 0 𝜆𝜆
and c is the speed of light. the radar, the IF signal will be a sine wave
requency representation as a function of time of the IF signal at the output of the frequency mixer, 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴(2𝜋𝜋𝑓𝑓𝑜𝑜 𝑡𝑡 + 𝜙𝜙0 ) (7)
In summary, for an object In at summary,
a distancefor d from
an the radar,
object the IF signal
at a distance willthe
d from be radar,
a sine wave
the IF(Equation
signal will
To obtain the frequency representation as a function (Equation 7), then:
wo lines presented in the upper section of Figure 4. The distance between the two lines is fixed, which
𝑆𝑆2𝑑𝑑 4𝜋𝜋𝑑𝑑
e IF signal consists of atime
toneofwith
the aIFconstant
signal atfrequency.where
Figure𝑓𝑓04=frequency
shows and 𝜙𝜙0 =frequency .
the output of the 𝑐𝑐 that this 𝜆𝜆 is Sτ. The IF
𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴(2𝜋𝜋𝑓𝑓 𝑜𝑜 𝑡𝑡 + 𝜙𝜙0 ) (7)**(7) 𝑜𝑜 𝑡𝑡 + 𝜙𝜙0 )
𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴(2𝜋𝜋𝑓𝑓
nly in the time interval where both the TX chirp and the RX chirp
mixer, subtract the two lines pre­sented in the 𝑆𝑆2𝑑𝑑 overlap
upper (i.e., the interval between the
The
whereassumption
𝑓𝑓0 = so
andfar𝜙𝜙is that the
4𝜋𝜋𝑑𝑑
. radar
𝑓𝑓0 = has and
𝑆𝑆2𝑑𝑑 detected only 4𝜋𝜋𝑑𝑑 one object. Let’s analyze a case when ther
lines in Figure 4) . 0 = where
where 𝜙𝜙0 = .
section of Figure 4. The distancedetected.
betweenFigurethe𝑐𝑐two 𝜆𝜆 𝑐𝑐 𝜆𝜆
per diagram in Figure 4 shows TX and RX chirps as a function of time for5ashows singlethree
objectdifferent
detected. RXNotice
chirps that
receivedthe from different objects. Each chirp is dela
lines is
p is a time-delay version offixed,
the TXwhich
chirp.means that the TheIFassumption
signal consists The
so far is that
Thethe assumption
radar hasso
assumption so farisisthat
detected
far that
only the
theoneradar has
object.
radar has detected
Let’s
detected analyze
onlyaone
caseobject.
when Let’
ther
of a tone with a constant frequency. FigureFigure
detected. 4 shows 5 showsonly three onedifferent
object. RX chirps
Let’s received
analyze a from
case different
when there objects.
detected. Figure 5 shows three different RX chirps received from differen Each chirp is del
e delay (τ) can be mathematically derived as Equation 4:
that this frequency is St. The IF signal is valid only in are several objects detected. Figure 5 shows three
the time interval where both the
𝜏𝜏 = TX chirp and
2𝑑𝑑
(5)the RX different RX chirps received from different objects.
𝑐𝑐
chirp overlap (i.e., the interval between the vertical Each chirp is delayed by a different amount of time
d is the distance to dotted
the detected
lines inobject
Figureand4).c is the speed of light. proportional to the distance to that object. The
different
in the frequency representation as a function of time of the IF signal at the output of theRX chirps translate
frequency mixer, to multiple IF tones,
f TX chirp
t the two lines presented in the upper section of Figure 4. The distance between each withlines
the two a constant
is fixed,frequency.
which
RX chirp
that the IF signal consists of a tone with a constant frequency. Figure 4 shows that this frequency is Sτ. The IF Reflected signal
S f
valid only in the time interval where both the TX chirp and the RX chirp overlap (i.e., the interval TX
between
chirp the from multiple
objects
dotted lines in Figure 4) .
 t
Figure 4. IF frequency
T is constant.
c

put signal as a magnitude function of

f time is a sine wave, since it has a constant frequency.
IF signal
se of the IF signal (Φ0) is the difference
S  between the phase of the TX chirp and the phase of the RX chirp
t

tant corresponding to the start of the IF signal (i.e., the time instant represented by
f the left vertical dotted
 t
). (Equation 5):
Figure 4. IF frequency is constant.
𝜙𝜙0 = 2𝜋𝜋𝑓𝑓𝑐𝑐 𝜏𝜏 (5) t
Figure 5. Multiple IF tones for multiple-object detection.
y, it can be furtherThe
derived into
mixer Equation
output 6: as a magnitude function
signal
of time is a sine wave,
4𝜋𝜋𝑑𝑑 since it has a constant This IF signal consisting of multiple tones must
𝜙𝜙0 = 𝜆𝜆 (6)
frequency. be processed using a Fourier transform in order
or an object at a distance d from the radar, to separate the different tones. Fourier transform
The initial phase of the the
Figure IF IF
4.
IF signal
signal (F0will
frequency) isisbe
theadifference
sine wave (Equation
constant.
7), then:

between the phase𝑜𝑜 𝑡𝑡of+the processing will result in a frequency spectrum that
𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴(2𝜋𝜋𝑓𝑓 𝜙𝜙0TX
) chirp and the (7) phase
er output signal as a magnitude function of time is a sine wave, since it has a constant frequency.
has separate peaks for the different tones each
of the RX chirp at the time instant corresponding
𝑑𝑑 4𝜋𝜋𝑑𝑑
ialand 𝜙𝜙0 of
phase = the IF
𝜆𝜆
. signal
to the(Φ ) isof
start0 thethedifference
IF signalbetween
(i.e., thethe
timephase of the TX chirp and the phase of the RX chirp
instant * This equation is an approximation and valid only if Slope and
me instant corresponding to theby
represented start
theof the
left IF signal
vertical (i.e., the
dotted linetime
in instant represented by the
distance are left vertical
sufficiently dottedit is still true that the phase
small. However,
n so far is that the radar has detected only one object. Let’s analyze a case when oftherethe IF are several
signal respondsobjects
linearly to a small change in the distance
igure 4). (EquationFigure 4).
5):
re 5 shows three different RX (Equation 5):
chirps received from different objects. Each chirp is(i.e.,
delayed by a different
Δf=4πΔd/l).
𝜙𝜙0 = 2𝜋𝜋𝑓𝑓𝑐𝑐 𝜏𝜏 (5)(5) ** In this introductory white paper we ignore the dependence of
the frequency of the IF signal on the velocity of the object. This
matically, it can be further derived into Equation 6: is usually a small effect in fast-FMCW radars, and further can be
easily corrected for once the Doppler-FFT has been processed.
4𝜋𝜋𝑑𝑑
𝜙𝜙0 = (6)
𝜆𝜆

mary, for an object The

at afundamentals
distance d of from thewave
millimeter radar, 4
the IF signal will be a sine wave
sensors (Equation 7), then: May 2017

𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴(2𝜋𝜋𝑓𝑓𝑜𝑜 𝑡𝑡 + 𝜙𝜙0 ) (7)

constant frequency.

peak denoting the presence of an object at a Velocity measurement with two chirps
specific distance.
Velocity Measurement
In order to measure velocity, an FMCW radar
Range resolution Velocity Measurement
transmits two chirps separated by Tc. Each reflected
In this section, let’s use chirp
phasor
is notation
processed(distance,
throughangle)
FFT for a complex number.
Range resolution is the ability to distinguish
In this section, let’s use phasor notation (distance, angle) fortoa detect
complexthe
number.
ty Measurement
between two or more objects. When Velocity Measurement
two objects range
withofTwo
the object
Chirps (range-FFT). The range-FFT
Velocity Measurementcorresponding with Two Chirps to each chirp will have peaks in
move closer, at some point, a radar system will
tion, let’s use phasor notation (distance, angle) for a complex Velocity
In order toMeasurement
number. measure velocity, the asame a FMCW radar
location, transmits
but withtwo a two chirpsphase.
different separated
The by Tc Each reflected ch
no longer be able to distinguish them In order to measure velocity,
as separate FMCW radar transmits chirps separated by Tc Each reflected chir
Figure through FFT to IF
5. Multiple detect
tonesthe forrange of the object
multiple-object (range-FFT). The range-FFT corresponding to each ch
detection.
Measurement with Two Figure
objects. Chirps
Fourier5.transformMultiple IFtheory tonesIn through
for
states FFT you
multiple-object
this that
section, tolet’s
detect
can themeasured
detection.
use range of the phaseobjectdifference
(range-FFT). corresponds
The tonumber.
range-FFT acorresponding
motion to each chir
the same location, butphasor
with anotation
different(distance,
phase. The angle)
measured for a complex
phase difference corresponds to a m
gure 5. Multiple IFincrease tones for
This IF signal multiple-object
the consistingby
resolution detection.
of multiple
increasing the tonessame
the amust location,
length but within a
thedifferent
object phase.
of vTc. The measured
beofprocessed using a Fourier transform in order to separate the different phase difference corresponds to a mot
oF measure
signal consisting
velocity,ofamultipleFMCW Figure radar 5.
tones Multiple
must beIFprocessed
transmits twotones for
chirps vTc.
multiple-object
using
separated Fourier
by T detection.
transform
Each reflectedin order chirpto separate
is processed the different
tones. Fourier transform processing Velocity
vTc.will result Measurement c
in a frequency with Two Chirps
spectrum that has separate peaks for the different tones
eFT. tones
Fourier must
to detect transform
the the
be processed
range IF
of signal.
processing using
the will
object a Fourier
result intransform
(range-FFT). a frequency The in order
spectrum
range-FFT to corresponding
separate
that hasthe different
separateto each peaks chirpfor thehave
will different
peakstones in
signal consisting of multiple each peak tonesdenotingmust be theprocessed
presence using of an aobject Fourier at transform
a specific distance. in order to separate the different
ng willdenoting
resultbutinwith a frequency spectrum thatmeasured
athas separate In order
peaks to measure velocity, to a motion in the object of chirps separated by Tc Each reflected chir
a FMCW radar transmits two
peak
location,
Fourier transform processing
the presence
aTo different
increase ofphase.
an
the
will
object
The
length
result
a the
in aoffrequency
specific IF phasedistance.
signal,
spectrum theforthat
difference thecorresponds
different tones
has separate peaks for the different tones The range-FFT corresponding to each chir
of an object at a specific Range distance.
Resolution through FFT to detect the range of the object (range-FFT).
eak denoting
e Resolution the bandwidth
presence of anmustobject also at be
a increased
specific distance. proportionally. An
the same location, but with a different phase. The measured phase difference corresponds to a mot
increased-length IF signal
Range resolution is the ability to distinguish results in an IF spectrum
vTc. between two or more objects. Tc When two objects move closer, at some
eResolution
resolution is the ability with to distinguish
twoa separate between
peaks. two or more objects. When two objects move closer, at some
point, radar system will no longer be able to distinguish them as separate objects. Fourier transform theory states that
,istinguish
a radar system between willtwo or more
no longer beobjects.
able to When distinguish two objects
them asmove separatecloser, at some
objects. Fourier transform theory states that
esolution is the ability youtocan
Fourier increase between
distinguish
transform the
theory resolution
alsotwostates orbymore increasing an theWhen
objects. length two of objects
the IF signal. move closer, at some
anr be able tothe
increase distinguish
resolution them as separate
by increasing theobjects.
length of thethat
Fourier transform
IF signal. theory states that
radar system will no longer be able to distinguish them as separate objects. Fourier transform theory states that
increasing the length Toof
observation the IF signal.
increase window
the length (T)ofcan theresolve
IF signal, frequency
the bandwidth must also be increased proportionally. An increased-length IF
increase
crease thethe resolution
length of the IFbysignal, increasing the lengthmust
the bandwidth of thealso IF signal.
be increased proportionally. An increased-length IF
components
signal resultsthat in anare separatedwith
IF spectrum by more two separatethan peaks.
nal, the bandwidth must also
results in an IF spectrum with two separate peaks. be increased proportionally. An increased-length IF
ease the length of the 1/THz. IF signal,
This the means bandwidth
that two must also betones
IF signal increasedcan proportionally. Figure 6. An increased-length IF
h two separate peaks. Fourier transform theory also states that an observation window (T)Two-chirp
can resolvevelocityfrequency
measurement. components that are
esults in an IFtheory
er transform spectrum also with
states two that separate
an peaks.
observation
be resolved in frequency as long as the frequency window (T) can resolve frequency components that are
separated by more than 1/THz. This means that two IF signal tones can be Figure resolved 6. in
Two-chirp
frequency velocity
as longmeasurement.
as the
es
atedthatbyan moreobservation
thandifference
1/THz. window (T) canthat
Thissatisfies
means resolve
the twofrequency
relationshipIF signal given components
tones that arein frequency asFigure
can be resolved
in long as 6. the
Two-chirp velocity measurement.
transform theory also frequency
states that difference
an observation satisfieswindow the relationship (T) can resolvegiven infrequency
Equationphase 8:
components that are
is means
ency that two
difference IF signal
satisfies thetones can be resolved
relationship given in Equation in frequency The8:phase as long as the is The
difference defined as Equation difference 10:is derived from Equation 6 as
ed by more than 1/THz. Equation 8: This means that two IF signal The tones phase can be resolved
difference is in frequency
defined as as long
Equation as
10: the
relationship given in Equation 8: 1 Equation 10:
ncy difference satisfies the relationship given Δ𝑓𝑓 > (8)
Δ𝑓𝑓 in > Equation
velocity8:
1 4𝜋𝜋𝜋𝜋𝑇𝑇𝑐𝑐
Figure 6. Two-chirp measurement. (8) 𝑇𝑇𝑐𝑐 ∆𝛷𝛷 = 4𝜋𝜋𝜋𝜋𝑇𝑇𝜆𝜆𝑐𝑐 (10)
1 𝑇𝑇𝑐𝑐 ∆𝛷𝛷 = (10)
(10)
Δ𝑓𝑓 > (8) 1 𝜆𝜆
where
eedifference is defined as TTc cisisthe
Equation the10: observation
Δ𝑓𝑓 > interval.
𝑇𝑇𝑐𝑐 (8)
Tc is the observation where interval. observation 𝑇𝑇𝑐𝑐 interval.You can derive the velocity using Equation Figure11: 6. Two-chirp velocity measurement.
You can derive the velocity You canEquation
using derive the 11:velocity using Equation 11:
al. 𝑆𝑆2Δ𝑑𝑑 4𝜋𝜋𝜋𝜋𝑇𝑇𝑐𝑐 𝑐𝑐 𝑐𝑐
Since
Since Δ𝑓𝑓 = Equation
, equation
𝑐𝑐 ∆𝛷𝛷 = as
8 can
(8) The be
𝑐𝑐be expressed
expressed
𝑐𝑐 (10) asasΔ𝑑𝑑 > = as(since B = 10:
STc ).
TΔ𝑓𝑓
c is =the𝑆𝑆2Δ𝑑𝑑
observation interval. =phase difference
B = STcis). defined Equation 𝜆𝜆Δ𝛷𝛷
, equation (8) can be expressed 𝜆𝜆 Δ𝑑𝑑 > (since 2𝑆𝑆𝑇𝑇𝑐𝑐 2𝐵𝐵 𝑣𝑣 =𝜆𝜆Δ𝛷𝛷 (11)
(11)
𝑐𝑐 2𝑆𝑆𝑇𝑇𝑐𝑐 2𝐵𝐵 4𝜋𝜋𝑇𝑇
n be expressed as Δ𝑑𝑑 >
𝑐𝑐 𝑐𝑐
= (since B =BST ). c). 𝑐𝑐 𝑣𝑣 = 𝑐𝑐 (11)
𝑆𝑆2Δ𝑑𝑑 (since = c ST 𝑐𝑐 4𝜋𝜋𝑇𝑇
4𝜋𝜋𝜋𝜋𝑇𝑇
𝑐𝑐 𝑐𝑐
erive
𝑓𝑓 = the velocity, equation usingThe range
(8)Equation resolution
2𝑆𝑆𝑇𝑇𝑐𝑐be2𝐵𝐵
can 11:
expressed (dasResΔ𝑑𝑑 ) depends
> =only(since on theB bandwidth
= ST ). swept the
Since by the chirp measurement
(Equation
∆𝛷𝛷 = 9): is based on (10)a
ange resolution
𝑐𝑐 (dRes) depends only on the bandwidth 2𝑆𝑆𝑇𝑇swept 2𝐵𝐵 by
𝑐𝑐 Since thethe chirpc (Equation
velocity measurement 9): isvelocity
based on a phase 𝜆𝜆difference, there will be ambiguity. The meas
ds only on the bandwidth The range swept resolution
by the chirp (d ) depends
(Equation Since
9): the
only on the velocity measurement
𝑐𝑐 phase difference, there will be ambiguity. The be ambiguity. The measur
is based on a phase difference, there will 𝜆𝜆
𝜆𝜆Δ𝛷𝛷
Res
ge resolution (dRes)bandwidth depends only on 𝑣𝑣
the =bandwidth You
𝑐𝑐 swept unambiguous
canby (11)
derive
the chirp only
𝑑𝑑the = if |∆𝛷𝛷|<
velocity
(Equation
𝑅𝑅𝑅𝑅𝑅𝑅 using
9): π. Using (9)
Equation 11: 11 above, one can mathematically derive 𝑣𝑣 <𝜆𝜆 4𝑇𝑇𝑐𝑐.
equation
swept by the𝑅𝑅𝑅𝑅𝑅𝑅 𝑑𝑑
4𝜋𝜋𝑇𝑇 =
𝑐𝑐chirp (Equation 9):unambiguous (9) only if 2𝐵𝐵
|∆𝛷𝛷|< π. Using equation 11 above, one can mathematically derive 𝑣𝑣 < .
𝑐𝑐 2𝐵𝐵 measurement is unambiguous only if |DF|< p. Using 4𝑇𝑇𝑐𝑐
𝑑𝑑𝑅𝑅𝑅𝑅𝑅𝑅 = (9) 𝜆𝜆Δ𝛷𝛷
velocity measurement Thus
2𝐵𝐵 an FMCW radar with
𝑑𝑑of𝑅𝑅𝑅𝑅𝑅𝑅 a chirp
𝑐𝑐
= GHz will bandwidth
Equation willa(9) of
12(9) a few
provides GHz will
the have
maximum
Equation a range
11 resolution
relative
above, speed
𝑣𝑣one
= in
(v
can the ) order
measured
mathematically of cm's
by (e.g.
two
(11) a chirp
chirps
derive spaced T apart.
an FMCW radar with a is based
chirp on a phase
bandwidth adifference,
few there have be ambiguity.
12range resolution Theinmeasurement
the order is (e.g.
of cm's max
a𝑐𝑐 chirp c
bandwidth of 4GHz translates to Equation provides the maximum relative speed 4𝜋𝜋𝑇𝑇 (vmax ) measured by two chirps spaced Tc apart. Hi
2𝐵𝐵
a shorter
range resolution
transmission 3.75cm) times between
𝜆𝜆 chirps.
bandwidth
ous only
width of atranslates
if |∆𝛷𝛷|<
of 4GHz fewπ.GHz Using will ahave
toequationrange a range
11 resolution
above,
resolution one can
3.75cm) inshorter
the order
mathematically of cm's (e.g.
derive a chirp
𝑣𝑣 < .
FMCW radar withThus a chirp anbandwidth
FMCW radar of a with few GHz a chirp willSincebandwidth
havethe atransmission
range
velocityofresolution timesinbetween
ameasurement 4𝑇𝑇𝑐𝑐order
the chirps.
is based of oncm's (e.g. a difference,
a phase chirp there will be ambiguity. The measur
a range resolution 3.75cm) 𝜆𝜆
dth of 4GHz translatesfew GHz
to a will
range have a
resolutionrange resolution
3.75cm) in the order Equation 12 provides v =
the maximum (12) 𝜆𝜆
12 provides the maximum relative speed (vmax) measured unambiguous
by two chirps only spaced
if |∆𝛷𝛷|< Tc π. Using
apart. equation
Higher vmax 11 max
vmaxabove,
requires 𝜆𝜆
= 4𝑇𝑇one 𝑐𝑐 canrelative speed
mathematically
(12) derive 𝑣𝑣 < 4𝑇𝑇 .
4𝑇𝑇𝑐𝑐 𝑐𝑐
ansmission times between of centimeters chirps. (e.g., a chirp bandwidth of 4 GHz (vmax) measured by two chirps spaced Tc apart.
Velocity
Equation cm). 12
Measurement
provides the maximum with Multiple Objects at)the Same Range
translates to a range resolution 3.75 Velocity Measurement Higher
with Multiple vmaxrelative
requires speed
Objects (v
shorteratmax
the measured
transmission
Same Rangebytimes
two chirps spaced Tc apart. Hi
𝜆𝜆
vmax = shorter (12)
transmission times betweenchirps.
between chirps.
4𝑇𝑇𝑐𝑐 The two-chirp velocity measurement method does not work if multiple moving objects with differ
Velocity measurement The two-chirp velocity measurement method does not work if multiple moving objects with differen
the time of measurement, both at the same distance𝜆𝜆 from the radar. Since these objects are at th
Measurement with Multiple Objects at the Samethe Rangetime of measurement, both at the same distance vmax = from the radar. Since (12)these objects are at the s
(12)
In this section, let’s use phasor notation will generate
(distance, reflective chirps with identical IF frequencies. 4𝑇𝑇𝑐𝑐 As a consequence, the range-FFT will r
will generate reflective chirps with identical IF frequencies. As a consequence, the range-FFT will res
hirp velocity measurement angle) formethod doesnumber. not work if multiple which moving represents objectsthe combined
with different signal from all are
velocities of these equi-range objects. A simple phase compa
a complex Velocity
which represents Measurement the combined with Multiple
signal from Objects
all of theseatatthe Same Range
equi-range objects. A simple phase comparis
of measurement, both at the same distance from the radar. not work.
Since these objects are at the same distance, they
not work.
ate reflective chirps with identical IF frequencies. As aThe two-chirp velocity
consequence, the range-FFT measurement will result method
in single does not work if multiple moving objects with differen
peak,
resents the combined signal from all of these equi-range objects. A simple phase comparison technique willfrom the radar. Since these objects are at the s
the time of measurement, both at the same distance
will generate reflective chirps with identical IF frequencies. As a consequence, the range-FFT will res
The fundamentals of millimeter wave sensors 5 May 2017
which represents the combined signal from all of these equi-range objects. A simple phase comparis
not work.
 Velocity Measurement
Figure 8. The range-FFT of the reflected chirp frame results in N phasors.
In this section,
Velocity measurement withlet’s use phasor
multiple notation
objects at (distance, angle)FFT,
A second for acalled
complex number. is performed on
Doppler-FFT,
A second FFT, called Doppler-FFT, is performed on the N phasors to resolve the two objects, as shown
Velocity Measurement with Two Chirps the N phasors to resolve the two objects, as shown
the same range

The two-chirp velocity measurement method does in Figure 9.

In order to measure velocity, a FMCW radar transmits two chirps separated by Tc Each reflected chirp is processed
not work if multiple moving objects with different
through FFT to detect the range of the object (range-FFT). The range-FFT corresponding to each chirp will have peaks in
velocities are the
at the time
same of measurement,
location, both phase. The measured phase difference corresponds to a motion in the object of
but with a different
at the same distance
vTc. from the radar. Since these ω1 ω2

objects are at the same distance, they will generate Figure 9. Doppler-FFT separates the two objects.

reflective chirps with identical IF frequencies. As

a consequence, the range-FFT will result in single w1 and w2 correspond to the phase difference
Figure 9. Doppler-FFT separates the two objects.
peak, which represents the combined signal from between consecutive chirps for the respective
ω1simple
all of these equi-range objects. A and ω2phase
correspond to the phase (Equation
objects difference 13):
between consecutive chirps for the respective objects
comparison technique will not work. 𝜆𝜆𝜔𝜔1 𝜆𝜆𝜔𝜔2
(13)
𝑣𝑣1 = 4𝜋𝜋𝑇𝑇𝑐𝑐
, 𝑣𝑣2 = 4𝜋𝜋𝑇𝑇𝑐𝑐
(13)
In this case, in order to measure the speed, the
Velocity
radar system must transmit more Resolution
than two chirps. It Velocity resolution
transmits a set of N equally spaced chirps.
You have This set
already seen thatThe
twotheory
discrete offrequencies,
discrete Fourier transforms
ω1 and ω2, can beteaches
resolved if Δω= ω2 - ω1 > 2π/N
of chirps is called a chirp frame. Figure 7 shows the us that two discrete frequencies, w1 and w2, can be
Figure 6. Two-chirp velocity measurement. 4𝜋𝜋𝜋𝜋𝜋𝜋𝑇𝑇𝑐𝑐
Since
frequency as a function of time for Δω is frame.
a chirp also defined byresolved
the following
if Dwequation
= w2 – w (equation 10) one can mathematicall
∆𝜔𝜔>=2p/N radians/sample.
1 𝜆𝜆
resolution(v
The phase difference as) Equation
is defined res if the frame
10:period Tf = NTc (Equation 14):
1 2 3 N Since Dw is also defined by the following equation
4𝜋𝜋𝜋𝜋𝑇𝑇𝑐𝑐
∆𝛷𝛷 = (Equation 10),
(10) one can mathematically
𝜆𝜆
derive the velocity resolution (vres) if the frame period
Tc You can derive the velocity using Equation 11:
Tf = NTc (Equation 14):
Figure 7. Chirp frame. 𝜆𝜆Δ𝛷𝛷 𝜆𝜆
𝑣𝑣 =
4𝜋𝜋𝑇𝑇𝑐𝑐
𝑣𝑣 > (11)
𝑣𝑣𝑟𝑟𝑅𝑅𝑅𝑅 = (14)
(14)
2𝑇𝑇𝑓𝑓
The processing technique is described below using
the example ofSince
twothe velocity
objects measurementfrom is
The velocity
equidistant based onofa The
resolution
the phasevelocity
the radar resolution
difference, there
is inversely of the
will radar is inversely
be ambiguity.
proportional to the The measurement
frame time (Tf). is
𝜆𝜆
radar but withunambiguous only if v|∆𝛷𝛷|<
different velocities π. Using equation proportional
11 above, one tocan frame time (Tf).derive 𝑣𝑣 < .
themathematically
1 and v2.
Angle Detection 4𝑇𝑇𝑐𝑐

Range-FFT processes theprovides

Equation 12 reflectedtheset of chirps,
maximum Angle
relative speed (vmax)detection
measured by two chirps spaced Tc apart. Higher vmax requires
resulting in a set of Ntransmission
shorter Angle
identically located Estimation
peaks, but
times between chirps. Angle estimation
each with a different phase incorporating
An FMCWthe phase
radar system can estimate 𝜆𝜆 the angle of a reflected signal with the horizontal plane, as sho
An
vmaxFMCW
= radar system(12)
can estimate the angle of a
contributions from both these objects (the individual
angle is also called the angle of arrival
4𝑇𝑇𝑐𝑐 (AOA).
reflected signal with the horizontal plane, as shown
phase contributions from each of these objects
Velocity Measurement with Multiple Objects at the10.
in Figure Same
ThisRange
angle is also called the angle of
being represented by the red and blue phasors in
The two-chirp velocity measurement method does arrivalnot
(AoA).
work if multiple moving objects with different velocities are at
Figure 8).
the time of measurement, both at the same distance from the radar. Sinceobject these objects are at the same distance, they
will generate reflective chirps with identical
v1
Key
IF frequencies. As a consequence, the range-FFT will result in single peak,
θ
which represents the combined signal from all of these equi-range objects. A simple phase comparison technique will
radar
not work. v2
radar
Figure 8. The range-FFT of the reflected chirp frame results in
N phasors. Figure 10. Angle of arrival.Figure 10. Angle of arrival.

Angular estimation is based on the observation that a small change in the distance of an object resul
in the peak of the range-FFT or Doppler-FFT.This result is used to perform angular estimation, using
antennas as shown in Figure 11. The differential distance from the object to each of the antennas re
change in the FFT peak. The phase change enables to estimate the AoA.
The fundamentals of millimeter wave sensors 6 May 2017
stimation
ngle Detection
W radar system can estimate the angle of a reflected signal with the horizontal plane, as shown in Figure 10. This
also called the angle of arrival (AOA).
gle Estimation
Figure 12. AoA estimation is more accurate for small values.
FMCW radar system can estimate the angle of a reflected signal with the horizontalFigure plane,12.
as shown in Figure is
AoA estimation 10.more
This accurate for small values.
gle is alsoofcalled Angular estimation is based on the observation that Maximum angular field of view
ar Field Viewthe angle of arrival (AOA).
a small change in the distance ofMaximum Angular Field of View
an object results The maximum angular field of view of the radar is
ular field of view ofinthe radar ischange
a phase definedinbythe
thepeak
maximum
of theAoA radar can estimate. See Figure
that theor
range-FFT
The maximum angular field defined
of viewby
ofthe
the maximum AoA that
radar is defined themaximum
by the radar canAoA that the radar can e
Doppler-FFT. This result is used to 13.perform angular estimate. See Figure 13.
estimation, using at least two RX antennas as
shown in Figure 11. The differential distance from -θmax θmax
Figure 10. Angle of arrival.
the object to each of the antennas results in a phase
estimation is based on theinobservation
change that aThe
the FFT peak. small change
phase in theenables
change distance of an object results in a phase change
eak of the range-FFT or Doppler-FFT.This result is used to perform
you to estimate the AoA.Figure 10. Angle of arrival. angular estimation, using at least two radar RX
as as shown in Figure 11. The differential distance from the object to each of theFigure antennas resultsangular
13. Maximum in a phase
field of view.
in theestimation
gular FFT peak. The phase
is based onchange enables tothat
the observation estimate
a smallthe AoA. in the distance of an object results in a phase change
change
he peak of the range-FFT or Doppler-FFT.This result is used to perform angular Unambiguous
estimation, using at least two RX
measurement of angle requires
Figure11.
13.The
Maximum angular d+Δdof view.
field
ennas as shown in Figure differential
d distance from the object to each of the antennas results in a phase
|Dw| < 180°. Figure Using 13.
Equation
Maximum 16, angular
this corresponds to
field of view.
nge in the FFT peak. The phase changeo enables to estimate the AoA. 2𝜋𝜋𝜋𝜋𝑅𝑅𝜋𝜋𝜋𝜋(𝜃𝜃)
surement of angle requires |∆𝜔𝜔|<180 . Using equation 16, this corresponds to < π.
Unambiguous measurement of 𝜆𝜆 angle requires |∆𝜔𝜔|<180o. Using equation 16, this corresponds to 2𝜋𝜋

l apart can service is:Equation 17 shows that the maximum field of view
TX RX
that the maximum field of view that two antennas spaced
antenna antennas
Equation 18 shows that the maximum
that field of spaced
two antennas view thatl apart
two antennas spaced
can service is: l apart can service is:
Figure 11. Two antennas
−1 are
𝜆𝜆 required to estimate AoA.
𝜃𝜃𝑚𝑚𝑚𝑚𝑚𝑚 = 𝐴𝐴𝐴𝐴𝐴𝐴 ( ) (18) 𝜆𝜆
2𝜋𝜋
Figure 11. Two antennas are required to estimate AoA. 𝜃𝜃𝑚𝑚𝑚𝑚𝑚𝑚 = 𝐴𝐴𝐴𝐴𝐴𝐴−1 ( ) (17)
(18)
2𝜋𝜋
In this configuration, the phase change is derived
the two antennas of 𝑙𝑙 = 𝜆𝜆/2 results in the largest angular field of view ±90 degrees.
A spacing between the two antennas of l = l/2
onfiguration, the phase change is derived
mathematically 15: A spacing
mathematically
as Equation between
as Equation 15:the two antennas of 𝑙𝑙 = 𝜆𝜆/2 results in the largest angular field of view ±90 degr
results in the largest angular field of view ± 90°.
ments mmWave Sensor FigureSolution
11. Two antennas are required to estimate AoA.
∆𝛷𝛷 =
2𝜋𝜋∆𝑑𝑑
Texas Instruments
𝜆𝜆
(15)
(15) mmWave Sensor Solution
his configuration,
FMCW theto
sensor is able phase changethe
determine is derived mathematically
range, velocity and angleasofEquation
Texas Instruments mmWave
15:
nearby objects by using a
er the assumptionUnder
of a planarthe assumption
wavefront basic of ageometry
planarAswavefront
shows
you canthat basic
see, Δ𝑑𝑑 = 𝑙𝑙𝐴𝐴𝐴𝐴𝐴𝐴(𝜃𝜃),
an FMCW sensorwhere l is to
is able
sensor solution thedetermine
distance the range, velocity and angle of nearby object
analog and digital electronic components.
n the antennas Thisgeometry
enables to derivethat
shows the Ddangle
∆𝛷𝛷= = valuecombination
from
2𝜋𝜋∆𝑑𝑑
lsin(q), a measured
where l is the RF,∆𝜙𝜙
of(15) withand
analog Equation
digital16:
electronic components.
𝜆𝜆 As you can see, an FMCW sensor is able to
diagram of the different components.
distance between the antennas. Thus the angle of
𝜆𝜆𝜋𝜋𝜋𝜋 Figure 14 is a block diagram of the different components.
Under the assumption of a planar 𝜃𝜃 = 𝐴𝐴𝐴𝐴𝐴𝐴−1 (basic
wavefront ) geometry shows(16) that determine the range, velocity and angle of nearby
arrival (q), can be computed from the measured
2𝜋𝜋𝜋𝜋 DFΔ𝑑𝑑 = 𝑙𝑙𝐴𝐴𝐴𝐴𝐴𝐴(𝜃𝜃), where l is the distance
ween the antennas This enables to derive the angle value from a measured ∆𝜙𝜙objects by using
with Equation 16:a combination of RF, analog and
at ∆𝛷𝛷 depends onwith Equation
𝐴𝐴𝐴𝐴𝐴𝐴(𝜃𝜃). This is 16:
called a nonlinear dependency. 𝐴𝐴𝐴𝐴𝐴𝐴(𝜃𝜃) is approximated with a linear function
digital electronic components.
en 𝜃𝜃 has a small value: 𝐴𝐴𝐴𝐴𝐴𝐴(𝜃𝜃) ~ 𝜃𝜃. 𝜆𝜆𝜋𝜋𝜋𝜋
𝜃𝜃 = 𝐴𝐴𝐴𝐴𝐴𝐴−1 ( 2𝜋𝜋𝜋𝜋 ) (16) (16)
Figure 14 is a block diagram of the
ult, the estimation accuracy depends on AoA and is more accurate when 𝜃𝜃 has a small value.as shown in Figure
te that ∆𝛷𝛷 depends Note on 𝐴𝐴𝐴𝐴𝐴𝐴(𝜃𝜃).
that DFThis is calledona nonlinear
depends sin(q). This is called a 𝐴𝐴𝐴𝐴𝐴𝐴(𝜃𝜃) isdifferent components.
dependency. approximated with a linear function
y when 𝜃𝜃 has a small value: 𝐴𝐴𝐴𝐴𝐴𝐴(𝜃𝜃)
nonlinear ~ 𝜃𝜃.
dependency. sin(q) is approximated with a TI has brought innovation to the field of FMCW
linear function only when q has a small value:
a result, the estimation accuracy depends on AoA and is more accurate when 𝜃𝜃sensing by integrating
has a small value.as showna DSP, MCU and the TX
in Figure
sin(q) ~ q. RF, RX RF, analog and digital components into a
As a result, the estimation accuracy depends on RFCMOS single chip.
AoA and is more accurate when q has a small value.
as shown in Figure 12. TX ant.

Angle estimation is most accurate at θ close to zero

Synth

LP FFT
Estimation accuracy degrades as θ approaches 90o ADC Signal Processing
RX ant. IF signal Filter
}
}
}

RF Analog Digital
radar
Figure 14. RF, analog and digital components of an FMCW sensor.
Figure 12. AoA estimation is more accurate for small values.

The fundamentals of millimeter wave sensors 7 May 2017

TI’s RFCMOS mmWave sensors differentiate for radar signal processing, while IWR1642
themselves from traditional SiGe-based solutions mmWave sensors use a DSP to perform the
by enabling flexibility and programmability in the required processing. A DSP provides more flexibility
mmWave RF front-end and the MCU/HWA/DSP and allows for software integration of other higher-
processing back-end. Whereas a SiGe-based level algorithms such as tracking and classification.
solution can only store a limited number of chirps These single-chip devices provide simple access
and requires real-time intervention to update chirps to high-accuracy object data including range,
and chirp profiles during an actual frame, TI’s velocity and angle that enables advanced sensing
mmWave sensor solutions are able to store 512 in rising applications that demand performance and
chirps with four profiles before a frame starts. This efficiency such as smart infrastructure, Industry 4.0
capability allows TI’s mmWave sensors to be easily in factory and building automation products and
configured with multiple configurations to maximize autonomous drones.
the amount of useful data extracted from a scene. Texas Instruments has introduced a complete
Individual chirps and the processing back-end can development environment for engineers working on
be tailored “on-the-fly” for real-time application industrial and automotive mmWave sensor products
needs such as higher range, higher velocities, higher which include:
resolution, or specific processing algorithms.
• Hardware Evaluation Modules for the
The TI mmWave sensor portfolio for automotive AWR1x and IWR1x mmWave sensors
scales from a high-performance front-end radar—
• mmWave software development kit (SDK)
AWR1243 sensor to single-chip radar—AWR1443
which includes RTOS, drivers, signal-processing
sensor and AWR1642 sensor. Designers can
libraries, mmWave API, mmWaveLink and
address advanced driver assistance systems
security (available separately).
(ADAS) and autonomous driving safety regulations—
• mmWave Studio off-line tools for algorithm
including ISO 26262, which enables Automotive
development and analysis which includes data
Safety Integrity Level (ASIL)-B— with the AWR
capture, visualizer and system estimator.
mmWave portfolio.
To learn more about mmWave products, tools and
The TI mmWave sensor portfolio for industrial
software please visit www.ti.com/mmwave and start
includes two single-chip devices; IWR1443
your design today.
mmWave sensors integrate a hardware accelerator

Important Notice: The products and services of Texas Instruments Incorporated and its subsidiaries described herein are sold subject to TI’s standard
terms and conditions of sale. Customers are advised to obtain the most current and complete information about TI products and services before placing
orders. TI assumes no l­iability for applications assistance, customer’s applications or product designs, software performance, or infringement of patents.
The publication of information regarding any other company’s products or services does not constitute TI’s approval, warranty or endorsement thereof.

The platform bar is a trademark of Texas Instruments. All other trademarks are the property of their respective owners.

© 2017 Texas Instruments Incorporated SPYY005

 IMPORTANT NOTICE FOR TI DESIGN INFORMATION AND RESOURCES

Texas Instruments Incorporated (‘TI”) technical, application or other design advice, services or information, including, but not limited to,
reference designs and materials relating to evaluation modules, (collectively, “TI Resources”) are intended to assist designers who are
developing applications that incorporate TI products; by downloading, accessing or using any particular TI Resource in any way, you
(individually or, if you are acting on behalf of a company, your company) agree to use it solely for this purpose and subject to the terms of
this Notice.
TI’s provision of TI Resources does not expand or otherwise alter TI’s applicable published warranties or warranty disclaimers for TI
products, and no additional obligations or liabilities arise from TI providing such TI Resources. TI reserves the right to make corrections,
enhancements, improvements and other changes to its TI Resources.
You understand and agree that you remain responsible for using your independent analysis, evaluation and judgment in designing your
applications and that you have full and exclusive responsibility to assure the safety of your applications and compliance of your applications
(and of all TI products used in or for your applications) with all applicable regulations, laws and other applicable requirements. You
represent that, with respect to your applications, you have all the necessary expertise to create and implement safeguards that (1)
anticipate dangerous consequences of failures, (2) monitor failures and their consequences, and (3) lessen the likelihood of failures that
might cause harm and take appropriate actions. You agree that prior to using or distributing any applications that include TI products, you
will thoroughly test such applications and the functionality of such TI products as used in such applications. TI has not conducted any
testing other than that specifically described in the published documentation for a particular TI Resource.
You are authorized to use, copy and modify any individual TI Resource only in connection with the development of applications that include
the TI product(s) identified in such TI Resource. NO OTHER LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE TO
ANY OTHER TI INTELLECTUAL PROPERTY RIGHT, AND NO LICENSE TO ANY TECHNOLOGY OR INTELLECTUAL PROPERTY
RIGHT OF TI OR ANY THIRD PARTY IS GRANTED HEREIN, including but not limited to any patent right, copyright, mask work right, or
other intellectual property right relating to any combination, machine, or process in which TI products or services are used. Information
regarding or referencing third-party products or services does not constitute a license to use such products or services, or a warranty or
endorsement thereof. Use of TI Resources may require a license from a third party under the patents or other intellectual property of the
third party, or a license from TI under the patents or other intellectual property of TI.
TI RESOURCES ARE PROVIDED “AS IS” AND WITH ALL FAULTS. TI DISCLAIMS ALL OTHER WARRANTIES OR
REPRESENTATIONS, EXPRESS OR IMPLIED, REGARDING TI RESOURCES OR USE THEREOF, INCLUDING BUT NOT LIMITED TO
ACCURACY OR COMPLETENESS, TITLE, ANY EPIDEMIC FAILURE WARRANTY AND ANY IMPLIED WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NON-INFRINGEMENT OF ANY THIRD PARTY INTELLECTUAL
PROPERTY RIGHTS.
TI SHALL NOT BE LIABLE FOR AND SHALL NOT DEFEND OR INDEMNIFY YOU AGAINST ANY CLAIM, INCLUDING BUT NOT
LIMITED TO ANY INFRINGEMENT CLAIM THAT RELATES TO OR IS BASED ON ANY COMBINATION OF PRODUCTS EVEN IF
DESCRIBED IN TI RESOURCES OR OTHERWISE. IN NO EVENT SHALL TI BE LIABLE FOR ANY ACTUAL, DIRECT, SPECIAL,
COLLATERAL, INDIRECT, PUNITIVE, INCIDENTAL, CONSEQUENTIAL OR EXEMPLARY DAMAGES IN CONNECTION WITH OR
ARISING OUT OF TI RESOURCES OR USE THEREOF, AND REGARDLESS OF WHETHER TI HAS BEEN ADVISED OF THE
POSSIBILITY OF SUCH DAMAGES.
You agree to fully indemnify TI and its representatives against any damages, costs, losses, and/or liabilities arising out of your non-
compliance with the terms and provisions of this Notice.
This Notice applies to TI Resources. Additional terms apply to the use and purchase of certain types of materials, TI products and services.
These include; without limitation, TI’s standard terms for semiconductor products http://www.ti.com/sc/docs/stdterms.htm), evaluation
modules, and samples (http://www.ti.com/sc/docs/sampterms.htm).

Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265
Copyright © 2017, Texas Instruments Incorporated