SENSORS and

SIGNAL PROCESSING

Lecture notes
Ha Le Nhu Ngoc Thanh, Ph.D.

Mechatronics Engineering – Sensors and Signal Processing 1 © Ha Le Nhu Ngoc Thanh, Ph.D.
 Course Info.
1. Lecturer: Ha Le Nhu Ngoc Thanh, Ph.D.
▪ Email: [email protected]
▪ Department: Mechatronics Engineering (E1-302)
2. Course evaluation:
▪ Schedule: 15 weeks (2 credits)
▪ Assignment: TBA (small project)
▪ Midterm exam: Week 8th (TBA) – Exam (normal)
▪ Final exam: Week 16th (TBA) – Exam (50%)
3. Course relations: electrical-electronic engineering course, Physics 1,2,3, Applied math, digital system,
automatic control
4. Textbooks and references
▪ Main book:
[1]. Jacob Fraden, “Handbook of modern sensors; physics, design, and applications” 5th Edition, Springer,
[2]. Ramon, J.G. Webster, “Sensors and signal conditioning”, 2nd Edition, John Wiley & Sons, Inc.
[2]. David G. Alciatore, “Introduction to Mechatronics and Measurement Systems”, 5th Edition, Mc Graw Hill

Mechatronics Engineering – Sensors and Signal Processing 2 © Ha Le Nhu Ngoc Thanh, Ph.D.
 Lecture 1-6 (Week 1-6):
General Concept and
Characteristics of sensors
Main contents
1. Mechatronics and Measurement systems
2. General concepts and Terminology
3. Sensor classification
4. General Characteristics of sensors
5. Dynamics characteristic of measurement system

Mechatronics Engineering – Sensors and Signal Processing 3 © Ha Le Nhu Ngoc Thanh, Ph.D.
 MECHATRONICS SYSTEMS*
❖ The actuators: produce motion
❖ Sensors: detect the state of the system parameters, inputs, and outputs.
❖ Conditioning and interfacing: provide connections between control circuit & input/output
❖ Digital control: used to generate control signal
❖ User interface: manual inputs, graphical displays or visual feedback to the user.

Example: Air-conditioner
Mechatronics Engineering – Sensors and Signal Processing 4 © Ha Le Nhu Ngoc Thanh, Ph.D.
 MECHATRONICS SYSTEMS
❖ Overview Signal:

Mechatronics Engineering – Sensors and Signal Processing 5 © Ha Le Nhu Ngoc Thanh, Ph.D.
 MEASUREMENT SYSTEMS
❖ Measurement system: a fundamental part to gather information about the measured quantity and
to convert it to an electric signal. Including:
1. Transducer: is a sensing element that converts a physical/stimulus input into
voltage/current/charge(điện tích).
2. Signal processor: filtering, amplification, or other signal conditioning.
3. Recorder: instruments, computer, stores, display, monitoring

Example: digital thermometer

Mechatronics Engineering – Sensors and Signal Processing 6 © Ha Le Nhu Ngoc Thanh, Ph.D.
 MEASUREMENT SYSTEMS*
❖ Complex example measuring system: motor speed controller

Mechatronics Engineering – Sensors and Signal Processing 7 © Ha Le Nhu Ngoc Thanh, Ph.D.
 MEASUREMENT SYSTEM DIAGRAM

Mechatronics Engineering – Sensors and Signal Processing 8 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CONCEPTS AND TERMINOLOGY*
Definition of sensor and transducer (no uniform, various interpretation)
• A Sensor: is a device to detect events or changes in its environment, and then provide a corresponding output (digital/
analog).
• Another definition: A sensor is a device that, when exposed to a physical phenomenon (temperature displacement, angle,
velocity, etc.), produces a proportional (linear or nonlinear) output electrical signal.
• E.g.

Thermocouple sensor Encoder sensor

Mechatronics Engineering – Sensors and Signal Processing 10 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CONCEPTS AND TERMINOLOGY*
• Transducer (integrated sensor): is a device that converts one form of energy (or signal) into another, with preserving
information. It consists of a sensor/actuator long with signal conditioning circuits.
Parts of Transducer
A transducer consists of the following two important parts:
• Sensing element
• Signal conditioning unit

(sensing element)

Mechatronics Engineering – Sensors and Signal Processing 11 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CONCEPTS AND TERMINOLOGY*
• Transducer and Transmitter

Example: measure pressure (0-100psi)

Transducer convert pressure to 0-3mV
Transmitter convert its to standard
electrical signal (1-5V or 4-20mA)

Mechatronics Engineering – Sensors and Signal Processing 12 © Ha Le Nhu Ngoc Thanh, Ph.D.
 SENSORS CLASSIFICATIONS*
1. Smart transducer (smart integrated sensor)
Integrate microcontroller to Self-diagnosis, self-identification, and self-adaptation, optimization of its
operation in terms of accuracy, speed, and energy consumption.
- There is a radio communication module and a battery for powering the sensor without requiring a
physical connection to the network.
- to be able to receive data from the outside, either in the form of commands that change its
configuration, either in the form of data to synchronize.

Mechatronics Engineering – Sensors and Signal Processing 13 © Ha Le Nhu Ngoc Thanh, Ph.D.
 SENSORS CLASSIFICATIONS
2. Analogue Signal and Digital signal
Analogue sensors: produce a continuous output
signal which is generally proportional to the
quantity being measured.

For example:
Using the thermocouple to measure the
temperature of a liquid, the output value
continuously responds to temperature changes
as the liquid is heated up or cooled down.

Example: Pressure transmitter to measure the

pressure in the pipe.

Mechatronics Engineering – Sensors and Signal Processing 16 © Ha Le Nhu Ngoc Thanh, Ph.D.
 SENSORS CLASSIFICATIONS
Digital sensors: produce a discrete digital output signals. Digital sensors produce a binary output
signal in the form of a logic “1” or a logic “0”, (ON, OFFF). This means that the signal may be
outputted as a single “bit”,
For example: To measure the speed of a motor, we use an Encoder sensor.

Mechatronics Engineering – Sensors and Signal Processing 17 © Ha Le Nhu Ngoc Thanh, Ph.D.
 SENSORS CLASSIFICATIONS*
3. signal conditioning perspective: The sensors can be divided into passive and active sensor/
transducer

e.g. microphones, thermocouples, photodiodes,

e.g. ultrasonic sensor, lidar sensor, photoresistor, and piezoelectric sensors,…
encoder,… - Disadvantage that the output signals are typically
weak and often require amplification using an active
electronic circuit.

Mechatronics Engineering – Sensors and Signal Processing 18 © Ha Le Nhu Ngoc Thanh, Ph.D.
 EXAMPLE
• Sensor and Actuators applied in some common engineering systems

Mechatronics Engineering – Sensors and Signal Processing 19 © Ha Le Nhu Ngoc Thanh, Ph.D.
 EXAMPLE
• Sensor and Actuators applied in UAV

Barometer sensor, GPS sensor

Mechatronics Engineering – Sensors and Signal Processing 20 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
• The choice of a sensor or actuator for an specific application depends on the requirements of
that application and the characteristics of the sensors/actuators that are provided by the
manufacture.
• There are some specifications that apply to most sensors are discussed:

1. Transfer function
2. Full-Scale Input (FS – full span) 10. Saturation
3. Range. 11. Repeatability
4. Full-Scale Output (FSO – Full span output)
12. Dead zone
5. Sensitivity
6. Offset 13. Resolution
7. Accuracy 14. Precision
8. Hysteresis 15. Calibration
9. Nonlinearity
16. Noise
17. Dynamics characteristics

Mechatronics Engineering – Sensors and Signal Processing 21 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
1. Transfer function (TF) of sensor:
- Transfer function specifies the functional relationship between the input quantity and the
electrical output signal. This relationship is generally described by a graph or mathematic
equation.
- In freq. domain, a transfer function describes true value (ideal value in ideal working
environment )
Y (s )
- TF characteristic can be: G (s )
* Linear R(s )
* Nonlinear
G (s )
R(s ) Y (s )
R(s ) G (s ) Y (s ) Sensor
stationary

Y (s ) G (s )R(s )
(stimulus)
1
y(t ) L G (s )R(s ) f r (t )
E S
E f (s )

Mechatronics Engineering – Sensors and Signal Processing 22 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
1. Transfer function (TF) of sensor:
- In time domain:

E f (s )
s : stimulus input (need to measure) E
E : electrical signal output
- Stimulus s is unknown
- The value of E that becomes known during
measurement is a number (voltage, current, digital
count, etc.) that represents stimulus s.
- A job of the designer is to make that representation
as close as possible to the true value of stimulus s.
(stimulus) s
s f 1(E )

Mechatronics Engineering – Sensors and Signal Processing 23 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*

Example 1: transfer function of a relative E

humidity sensor.
Dash line: manufacturing tolerance
Thick line: typical value (measured value)
Straight line: ideal value.
Known via
measured
value of
s f 1(E ) sensor

V k. Relative Humidity %
1
Relative Humidity % V unknown
k
need to
determine s

Mechatronics Engineering – Sensors and Signal Processing 24 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
Example2 : the sensor that measures mass flow of fluid .

Sensor output can be in volts or in

volt ADC
digital count received from the ADC,
as shown on the y-axis of for a 10-bit
ADC converter.

n f (s ), n is ADC value

How to find real flow?

* Using Interpolating method to f(s)

s

Mechatronics Engineering – Sensors and Signal Processing 25 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
Example2 : the sensor that measures mass flow of fluid .

After the output count n is measured,

it has to be translated back to the
flow rate by use of the inverse transfer
function.

1
s f (n ) F (n )

ADC

volt

Mechatronics Engineering – Sensors and Signal Processing 26 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
1.1. Functional Approximations:
Approximation is a selection of a suitable mathematical expression that can fit the experimental
data as close as possible (curve fitting )

i. The ideal model is linear transfer function:

(linear approximation)

E A Bs B : slope sensitivity

Mechatronics Engineering – Sensors and Signal Processing 27 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
Example:
clc; clear; close all;
% data
x = linspace(0.01,3,100)'; % s=x
y = log10(x) + 0.1*randn(100,1); % E=y
% approximation
f = fittype('A + B*x') %function E
E 0.62 0.434s approximate = fit(x,y,f)
E (0) 0.62 data plot(approximate,x,y)

In many practical cases, It is just

difficult or impossible to test a
sensor at a zero input.

For example, a temperature sensor

(Kelvin scale) cannot be tested at
the absolute zero (- 273.15 C)

Mechatronics Engineering – Sensors and Signal Processing 28 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
1.1. Functional Approximations:
Thus in general cases: If the sensor response is E0 for some known input stimulus s0,

E E0 B s s0 Example PT100:

E E0
s s0
B
Note that: three constants shall be known
for computing the stimulus s: sensitivity B
E0
and coordinates s0 and E0 of the reference
point.

However: in real, very few sensors are truly

s0
linear.

Mechatronics Engineering – Sensors and Signal Processing 29 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
1.1. Functional Approximations:
Thus in general cases: If the sensor response is E0 for some known input stimulus s0,

E E0 B s s0

B 0.434
select : s 0 , E 0 0.297, 0.4901
E 0.4901 0.434 s 0.297

E 0.4901
s 0.297
0.434

Mechatronics Engineering – Sensors and Signal Processing 30 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
Example: a semiconductor temperature sensor know TF is linear
Let proposes an experimental method to determine TF of temperature sensor in below
Figure
by experimental method:
o
T( C ) sensor Measure:
t t1 t2
n n1 n 2
(ADC) n1 A B t t1
n2 A B t t2
n2 n1
Sensitivity(slope): B , select A n1
t2 t1
n n1
n n1 B t t1 t t1
B
How to find TF if TF is nonlinear function?

Mechatronics Engineering – Sensors and Signal Processing 31 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
1.1. Functional Approximations:
A nonlinear transfer function can be approximated by a nonlinear mathematical function.
ii. The logarithmic approximation function:
Example:
clc; clear; close all;
% data
x = linspace(0.01,3,100)';
y = log10(x) + 0.1*randn(100,1);
% approximation
f = fittype('A + B*log(x)') %function E
approximate = fit(x,y,f)
% plot(x,y,'.')
% hold
plot(approximate,x,y)

E 0.011
s e 0.44

Mechatronics Engineering – Sensors and Signal Processing 32 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
1.1. Functional Approximations:
A nonlinear transfer function can be approximated by a nonlinear mathematical function.
iii. The exponential approximation function:

clc; clear; close all;

% data
x = linspace(0.01,3,100)';
y = exp(x) + 0.5*randn(100,1);
% approximation
f = fittype('A*exp(k*x)'); %function E
approximate = fit(x,y,f)
% plot(x,y,'.')
% hold
plot(approximate,x,y)

E
1
s ln 1.06
0.978

Mechatronics Engineering – Sensors and Signal Processing 33 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
1.1. Functional Approximations:
A nonlinear transfer function can be approximated by a nonlinear mathematical function.
iv. The power approximation function:
clc; clear; close all;
% data
x = linspace(0.01,3,100)'; % s=x
y = log10(x) + 0.1*randn(100,1); % E=y
% approximation
f = fittype('A + B*x^k') %function E
approximate = fit(x,y,f)
% plot(x,y,'.')
% hold
plot(approximate,x,y)

1
E 166.1 0.002592
s
166.1

Mechatronics Engineering – Sensors and Signal Processing 34 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
Example (find the TF of IR distance sensor)

Using MATLAB for approximating

How to find TF if TF is arbitrary

nonlinear function ?
Klimenda, F.; Cizek, R.; Pisarik, M.; Sterba, J. Stopping the Mobile Robotic Vehicle at a Defined Distance from the Obstacle by Means of an
Infrared Distance Sensor. Sensors 2021, 21, 5959. https:// doi.org/10.3390/s21175959

Mechatronics Engineering – Sensors and Signal Processing 35 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
v. Computation from Linear Piecewise approximation (ước lượng tuyến tính từng phần):
n
f (s )
Accept
error

s
si sx si 1
Problem:
Given signal n f (s )
n : ADC
s : Physical input
User use data in si , si 1
,
let measure data in sx
Mechatronics Engineering – Sensors and Signal Processing 36 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
Example of Linear Piecewise approximation:
Problem:
A thermistor temperature sensor with a pull-up resistor R1=10[kΩ] connected to a ADC 12bit
(N0) used to measure temperature from 0 to +60 C.
Compute the unknown temperature when
receive counts n x =1505 by experiment method
where n1 1863, n2 1078

Mechatronics Engineering – Sensors and Signal Processing 37 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
Example of Linear Piecewise approximation:
Solution: Method to determine ADC at 20C and 40C: The thermistor sequentially is immersed into a
liquid bath at 20(oC) and 40(oC).The ADC counts are from output: n1=1863 and n2=1078
nx n1 1505 1863
tx t1 t2 t1 20 40 20 29.12 oC
n2 n1 1078 1863
How to select two points for Linear Piecewise approximation to reduce
error?

Based on the using area,

we can select the suitable
two points for piecewise
approximation as beside
picture

Using area

Using area

Mechatronics Engineering – Sensors and Signal Processing 38 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
vi. The spline Interpolating method: Let estimate the temperature when ADC
obtained 1018
By the spline interpolating method?
Solve:

Mechatronics Engineering – Sensors and Signal Processing 39 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
vii. Newton interpolating approximation

Estimate the distance of sensor given

obtained output 1.4808 (V)

Mechatronics Engineering – Sensors and Signal Processing 40 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
2. Full-Scale Input (Full span - FS):
- Full scale input (or FS) represents the largest possible input value, which can be measured by
sensor/transducer/transmitter without changing accuracy.
3. Range: from lowest to highest values of the stimulus
4. Full-Scale Output (FSO – Full span output):
- Difference between the electrical output signals measured with maximum input and minimum
input stimulus applied.
E electrical signal
Full span s max s min
Range : from s min to s max E max
FSO E max E min
- FS may be equal in two sensors, but ranges are not same E min
s stimulus
s min s max

Mechatronics Engineering – Sensors and Signal Processing 41 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
Example: pressure sensor
- FS is the range between the minimum pressure Pmin and the maximum pressure Pmax that can be
measured by the sensor or FS= Pmax - Pmin
- FSO is the difference between the output voltages Vmin and Vmax corresponding to Pmin and Pmax ,
respectively, or FSO= Vmax - Vmin
- For an actual transfer function, FSO should include all deviations from the ideal transfer function

Mechatronics Engineering – Sensors and Signal Processing 42 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
Example:
• A sensors is designed for: -30 °C to +80 °C to output 2.5V to 1.2V
Determine range, span (FSI), FSO?

Mechatronics Engineering – Sensors and Signal Processing 43 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
5. Offset :
- When the physical value is zero but a sensor’s output value is not zero, then the sensor has an
offset. Offset is also called zero or null offset, zero or null drift, offset error, bias. Figure 1.3a
- An offset can also be described by the difference between the sensor’s actual output value and a
specified ideal output value (Figure 1.3b)

Mechatronics Engineering – Sensors and Signal Processing 44 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
6. Sensitivity:
- Sensitivity, the slope of a sensor’s transfer function, is defined as the ratio of a small change in the
sensor’s output to a small change in its input.

- Sensor is more sensitive when the same input but the output is larger

Mechatronics Engineering – Sensors and Signal Processing 45 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
6. Sensitivity:
- For a nonlinear transfer function, sensitivity is not a fixed number
- A nonlinear transfer function exhibits different sensitivities at different points in intervals of stimuli.
- Sensitivity is defined as a first derivative of the transfer function:
dE (si ) Ei
bi (si )
ds si
1.25V

0.2V

More sensitive Less sensitive

30 cm 30cm

Mechatronics Engineering – Sensors and Signal Processing 46 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
6. Sensitivity:
- Example:

Mechatronics Engineering – Sensors and Signal Processing 47 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
6. Sensitivity:

Mechatronics Engineering – Sensors and Signal Processing 48 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
6. Sensitivity:
- Sensitivity error or sensitivity drift is the actual sensitivity deviation from the ideal sensitivity,
usually expressed as a percentage.

- Example 1: HMC1021 magneto-resistive temperature sensor

E electrical signal

E max

kreal kideal kideal

%sensitivity error 100% kreal
kideal E min
s stimulus
s min si s max

Mechatronics Engineering – Sensors and Signal Processing 49 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
6. Sensitivity:

Mechatronics Engineering – Sensors and Signal Processing 50 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
7. Accuracy (Error):
Very important characteristics.
- Errors: is the difference between the result of
the measurement value and the true value of
the stimulus input
Error = measured value – true value
The real Transfer function must be fall within
the limit specified accuracy area.
- Error computed in output. At same error at
stimulus x, error: ∆E = Y’ – Y
- Error computed in input. error at output y’,
error: ∆s = Z’ – Z
- The real functions deviate from the ideal by ±δ,
where 𝛿 ≤ ±∆

Mechatronics Engineering – Sensors and Signal Processing 51 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
7. Accuracy (Error): Error rating may be represented by:
✓ Directly in terms of measured value (∆): It is used when error is independent on the input
signal magnitude.
- For example for a temperature sensor: it can be stated: the accuracy 0.15C.
- Flowmeter sensor: accuracy 5 m3/s in the range from below 100 m3/s, and 10 m3/s in the range over 100
m3/s.

✓ In % FSO: It is useful for highly nonlinear transfer function. %Error (FSO )

max E
.100%
E max E min
max s
%Error (FS ) .100%
✓ In % FS input: It is useful for a linear transfer function. s max s min

✓ Absolute & Relative Errors: How the RESULTS close to

the TRUE VALUE of the measured quantity.

Mechatronics Engineering – Sensors and Signal Processing 52 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
7. Accuracy (Error):
Example 1: in datasheet of a temperature sensor gives
- Range of measurement: -30 to 80(oC)
- Output: 2.8V to 1.5V
- Error input: ±0.5(oC), and output: ±0.059V
Let describes error in:
a) Directly
b) %FSI
c) %FSO

Temperature measurement (C)

Mechatronics Engineering – Sensors and Signal Processing 53 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
7. Accuracy (Error):
Example 1: in datasheet as Figure
Find absolute error

Mechatronics Engineering – Sensors and Signal Processing 54 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
7. Accuracy (Error):
Example 2: error in real measurement
Given a distance sensor with TF: 𝐸 = 𝑠 , where, E is output voltage [mV], s is distance [mm]. In
experiment, at s=10mm, the output is E = 10.5 mV, Find error in absolute and relative Error .

Mechatronics Engineering – Sensors and Signal Processing 55 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
7. Accuracy (Error):
If there many factors affecting on the accuracy of a sensor including temperature fluctuation,
linearity, hysteresis, repeatability, stability, zero offset, A/D (analog-to-digital) conversion error, and
display resolution,…,The overall accuracy of a sensor can be expressed by

Overall accuracy e12 e22

where e1, e2, represents each ontributing component of accuracy.

Mechatronics Engineering – Sensors and Signal Processing 56 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
8. Hysteresis:
• The maximum difference in terms of the measured
quantity (input) between increasing and decreasing
process is defined as hysteresis error meaning the decreasing
maximum deviation in value between the increasing and
decreasing cycle measured at the same input point.
• (Example: temperature, pressure, velocity…)
e0
Explain:
• If we want to measure an input x when input is increasing,
received output is e0
• However, when input is decreasing to output is e0 , then
input is y (# x) → xy is hysteresis. increasing
• If system have no hysteresis, increasing =decreasing

Max ( h ) y x
Hysteresis error in % = 100%
FS

Mechatronics Engineering – Sensors and Signal Processing 57 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
8. Hysteresis:
• Example:
FS 100 PSI
Pincrease 50
Pdecrease 49.8 V1
Vout V1
Find hysteresis error in%

Mechatronics Engineering – Sensors and Signal Processing 58 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
8. Hysteresis:

Measured quantity

Mechatronics Engineering – Sensors and Signal Processing 59 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
9. Nonlinearity:
• A nonlinearity is a maximum deviation Best-fit approximation
(Least Square Method)
(L) of a real transfer function and its
best-fit approximation straight line.

approximation straight line

Max L
Nonlinearity % 100%
FS

Max L2
If we use Line 1: Nonlinearity % 100%
FS
Max L1
If we use Line 2: Nonlinearity % 100%
FS

Mechatronics Engineering – Sensors and Signal Processing 60 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
9. Nonlinearity: Best-fit straight line
• A nonlinearity is a maximum deviation (L) of a real transfer approximation
function and its best-fit approximation straight line. (Least Square Method)

• Nonlinearity may be approximated by straight lines.

• Nonlinearity is straight line of two terminal points (Line no.1)
• Using a linear best fit (least squares method) through the
points of the curve (Line no.2)
• Improve accuracy by narrow the input stimuli (Line No.2)

Mechatronics Engineering – Sensors and Signal Processing 61 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
9. Nonlinearity:
• What is different between accuracy, hysteresis, and nonlinearity ?

BSL: best straight line ideal

1 hysteresis
1
2 Nonlinearity
3
Accuracy

(1) NLH: nonlinearity and hysteresis

Mechatronics Engineering – Sensors and Signal Processing 62 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
9. Nonlinearity:
Example

Mechatronics Engineering – Sensors and Signal Processing 63 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
10. Saturation:
• Every sensor has its operating limits

• At some levels of the input stimuli, its output

signal no longer will be responsive.

• further increase in stimulus does not produce

a desirable output.

Linear span Saturation

Mechatronics Engineering – Sensors and Signal Processing 64 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
11. Repeatability:
• The repeatability is expressed as a maximum difference
between the output readings as determined by two run
measurement cycles under identical conditions First time measure curve

• Repeatability error describes the inability of a sensor to

measure the same value under identical conditions.
E second time measure
curve

max s
r
(FS ) .100%, r
: Repeatability error
FS

max E
r
(FSO ) .100%, r
: Repeatability error s
FSO

Mechatronics Engineering – Sensors and Signal Processing 65 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
11. Repeatability:
• Example:

FSO = 12V

10 15 20 25 30 40

10 15 20 25 30 40
Run 1 2.2V 2.7V 3.1V 3.5V 3.8V 4.2V
Run 2 2.1V 2.5V 2.9V 3.3V 3.5V 3.3V
Find Repeatability error%(FSO)?

Mechatronics Engineering – Sensors and Signal Processing 66 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
12. Dead zone (dead band):
• Dead zone is the insensitivity area of a sensor. The output may remain near a certain value (often
zero)

Mechatronics Engineering – Sensors and Signal Processing 67 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
13. Resolution:
• Resolution describes the smallest increment of a stimulus
that can be sensed by sensor
• Or the magnitude of the input variation, which results in
the output smallest step increase.
• Resolution also can be described by %Full Scale (FS)

Example:
- Encoder has 200 slot → resolution: 1.8deg (minimum
detectable angle)
- Described in FS%: 200 slot → 360 deg, 1.8 deg resolution
can be described by: resolution = 1.8/360 = 0.5% of FS

• Resolution may vary over the range, hence, the resolution

may be specified as typical, average, or worst

Mechatronics Engineering – Sensors and Signal Processing 68 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
13. Resolution: can be describes in Number of ADC bits

- Quantizer: lượng tử hóa (làm tròn)

- T: smaller is better,
- Number of bit of coder is larger, better

Ex. coder 3 bit

LSB: Least Significant Bit

(Count)
T LSB

Analog signal Discrete step Quantizer and coder step

Mechatronics Engineering – Sensors and Signal Processing 69 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
13. Resolution: Process of convert from Physical phenomenon to ADC
convert to convert to
s (physical input) E (V , I ) AD
• Corresponding a smallest increment of a stimulus (Resolution: ∆s), the obtained output is VLSB
CONVERT FROM Physic to Electrical signal CONVERT Electrical signal to Digital by ADC
E
(analog) VFS Vmax Vmin
Vmax

…
VLSB
VFS LSB

∆s

Vmin s VLSB VFS

Smin Smax 0
Mechatronics Engineering – Sensors and Signal Processing 70 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
13. Resolution:
• Resolution of a sensor with a digital output is the number of bits
• If output of sensor is 0 – VFS is converted to digital signal ADC
(N bits)
Analog(V) Digital (count) 1..111 2N
convert to N 0..111 2N 1

VFS 2 1 (N : coder bits )

4094
VLSB ? 1 LSB 4093

VFS VRef
resolution : VLSB
2N 1 2N
LSB: Least Signifcant Bit LSB
0..101
0..100

- LSB is the smallest level that an ADC can convert

0..011
0..010
or is the smallest increment a DAC outputs (1 count number). 0..001
VRef
- VFS is maximum output signal 0..000
0 VFS
VLSB : resolution

Mechatronics Engineering – Sensors and Signal Processing 71 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
13. Resolution:
• Example 0: The resolution of an n-bit ADC with maximum input voltage of 5 V is 5 m V. Find the
value of n

Solution:

Mechatronics Engineering – Sensors and Signal Processing 72 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
13. Resolution:
• Example 1:
If a sensor releases a maximum output 5V, with ADC 12 bits, 14 bits, 16 bits,
a) Compute resolution?
b) Compute VRef ?

Mechatronics Engineering – Sensors and Signal Processing 73 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
13. Resolution:
• Example 2: weight scale AN3183
- Full-scale weight = 2000g
- Output sensitivity = 1mV/V
- Working voltage (input excitation) = 3.3V
- VRef= 3.3V
Question:
Find weight scale resolution of ADC 16-bits, 22-bits
(question means ADC change 1 LSB (1 count), how many weight change?)
Solution:

Mechatronics Engineering – Sensors and Signal Processing 74 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
14. Precision:
- A sensor performs several measurements of the same quantity under the same conditions, but the
results obtained are not necessarily the same.

x xn

Precision can be expressed mathematically by:

Mechatronics Engineering – Sensors and Signal Processing 75 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
14. Precision:

Mechatronics Engineering – Sensors and Signal Processing 76 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR*
14. Precision:
- Compare accuracy and precision

Mechatronics Engineering – Sensors and Signal Processing 77 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
15. Noise:
- Sensor noise is random variations of sensor
output unrelated to variations in sensor input.
- All sensor produce some noise along with the
output signal.
- Sensors are often placed in noisy
environments such as factories, trains,…
- Noise is unavoidable feature of a dynamical
system. But there are some solutions to
minimize this interference such as filter
techniques: low pass filter, average filter,
Kalman filter,….

Mechatronics Engineering – Sensors and Signal Processing 78 © Ha Le Nhu Ngoc Thanh, Ph.D.
 GENERAL CHARACTERISTIC OF SENSOR
16. Calibration and Calibration error
- Many sensors require calibration prior to their use.
- Calibration is to determine the coefficients in their transfer functions (offset, sensitivity, or both).
- OR calibration means to convert the sensor’s electrical output (e.g., voltage) to a measured value
(e.g., temperature), 𝒔 = 𝒇−𝟏 (𝑬)

- The general calibration procedure involves comparing the measured values with true values and
then adjusting the parameters to match the actual values.
- The true values (or the standard curve) may be obtained by:
+ (1) using a more accurate sensor as the calibration reference.
+ (2) provided by the sensor’s manufacturer.
- Smart or advanced sensors may have self-calibration and adaptive features to adjust themselves
to different operating conditions

Mechatronics Engineering – Sensors and Signal Processing 79 © Ha Le Nhu Ngoc Thanh, Ph.D.
 HCMUTE Univ. - FME
https://fme.hcmute.edu.vn/

Lecture notes info

Sensors and Signal Processing
Fall Semester, 2023
------------------------------------------------

Copy right © Ha Le Nhu Ngoc Thanh, Ph.D.

Email: [email protected]
Sites: https://sites.google.com/view/halenhungocthanh/courses?authuser=0

Mechatronics Engineering – Sensors and Signal Processing 94 © Ha Le Nhu Ngoc Thanh, Ph.D.