Chapter

Introduction to
Embedded Systems
1.1 What Is an Embedded System?

Embedded systems have some kind of built-in computer system, but

they are not computers by intent. An automobile emissions control
module, also more formally known as an onboard diagnostics (OBD)
system, constitutes an example of an embedded system. An OBD system has, as an integral part of it, a microcontroller, hardware input
and output devices, and its own software. OBD systems were initially
designed to monitor and control the correct operation of a car emissions system. Later they became so successful that more functions
and features were added to them. Features addressed questions such
as these: Are all the spark plugs operating correctly? Is the air-fuel
mixture optimal for the present driving conditions? Is the intake air
temperature sensor reading open or shorted? Is the fuel level circuit
functioning correctly? Todays OBD systems provide hundreds of
such features. Some are generic to all auto manufacturers, while others are manufacturer-defined and specific to their own models. The
heart of an OBD system is its electronic control module (ECM) or its
central processing unit (CPU) with its associated interfacing devices,
such as gas, temperature, and pressure sensors; dashboard displays;
and naturally the software that runs the system.
Many other examples of embedded hardware systems exist in
our daily lives. These include cellular telephones, music players,
global positioning systems (GPSs), and kitchen appliances. There is
no clear-cut definition of what constitutes an embedded system, how
large or fast they are, or where they reside. All embedded systems
share some common characteristics. A computer system and its associated hardware are an integral part of the system, without the embedded system being explicitly a computer system, such as a generalpurpose personal computer. Figure 1.1 depicts a general block
diagram of a generic computer system. All computer systems have
four major building blocks: the brain of the machine, more formally

01_Lipiansky_Ch01_p001-012.indd 1

10/27/11 5:36:32 PM

Chapter One

Central processing unit

(CPU)

Address,
data, and
control
busses
Output devices

Input devices
Main memory
unit

Figure 1.1 Block diagram of a microcomputer system with busses.

called the CPU; the memory devices that hold program instructions,
input data, and application-generated data; the input devices; and
the output devices. These devices allow the system to be interfaced
with the external world.
An address bus, a data bus, and a control bus interconnect all such
major blocks. Greater discussion of busses can be found in Chap. 2. In
the OBD automobile example, some of the input devices can be air,
coolant, and oil temperature sensors; engine oil pressure sensors; and
oxygen sensors. Some of the output devices, for the same example,
can be the dashboard, which alerts the driver to the engine status;
electronic drivers that drive the fuel injectors; and relays and actuators
that control pumps, fans, and miscellaneous mechanical valves.
Figure 1.2 depicts a simplified block diagram of an automobile OBD
system. In the OBD system, many of the input devices are distributed
throughout the engine, some in the intake and exhaust manifolds;
some sensors are in the cooling system, while others are in the exhaust
system; and so on. The most informative output device is on the dashboard, where a malfunction indicator light (MIL) is located. This light
is also referred to as the check engine light.

01_Lipiansky_Ch01_p001-012.indd 2

10/27/11 5:36:33 PM

Introduction to Embedded Systems

Electronic control
module (EMC)
(embedded CPU)
Airflow meter

Check engine light and other

dashboard indicators

Fuel pump control driver

Sensors:
water temperature,
exhaust gas,
detonation, crank
angle

Fuel injection and mixture ratio

control drivers

Switches:
ignition, throttle valve,
neutral, air conditioner,
oil pressure,
power steering, and
miscellaneous others

Ignition timing control

electronics

Idle speed control solenoid

valve driver

Battery voltage

Automatic transaxle control

solenoid valve driver

Figure 1.2 Simplified block diagram of an automobile emissions control system.

Figure 1.3 depicts a functional block diagram of a microcontrollercontrolled DC motor. This system automatically generates an appropriate output signal to drive a motor, typically a stepper motor with
great accuracy.
The electric motor is the plant or the entity that needs its shaft
position to be accurately controlled. The digital computer on the left of
Fig. 1.3 is the error amplifier. The ADC and DAC are, respectively,

External
input

Digital
computer

Digital-to-analog
converter
(DAC)

Analog-todigital
converter

Power
amplifier

Motor

Shaft
position
being
controlled

Shaft position
encoder/
tachometer

Figure 1.3 Simplified block diagram of a dc motor position control system.

01_Lipiansky_Ch01_p001-012.indd 3

10/27/11 5:36:34 PM

Chapter One

the analog-to-digital converter and the digital-to-analog converter.

The power amplifier provides the required signal strengths to drive
the DC motor. The output of the DC motorits shaft positionis
sensed, and its position is encoded and fed back to the error amplifier. The error amplifiers external positive input is a signal that is
proportional to the final position desired. Such systems are referred
to as digital control systems, sometimes also called sample data systems.
Such systems are self-correcting; i.e., the output or outputs are
being sensed to optimize the operation of the plant in a dynamic
fashion. Unfortunately sample data systems as well as their characteristics, stability requirements, and other features are beyond
the scope of this book.
Another very common feature is that many embedded systems
operate in real time, i.e., at the speed at which the environment
changes. Embedded systems can be entirely within the body of a
product; however, some can be geographically distributed, such as
with industrial controls in a gasoline refinery. Others can be implanted
in the human body, such as pacemakers or a left ventricular assist
system (LVAS) to keep an ailing heart functioning better.

1.2 What Is the Embedded Hardware?

The embedded hardware consists of all the electronic components
such as integrated circuits, inductors, capacitors, terminating resistors, and crystal oscillators. The hardware also comprises mechanical
or electro mechanical components such as pressure, temperature, and
airflow sensors; relays; actuators; displays; and warning lights. Additionally the embedded system hardware includes the mechanical
framework or chassis of the system, when it is applicable, or a printedcircuit board. This book deals fundamentally with the electrical and
electronic circuits used in embedded systems, their function, and how
they can be interfaced to other devices. It is beyond the scope of this
book to treat in any depth the mechanical, chemical, hydraulic, physiological, or any other technical aspect that may be present in an
embedded system. To write a book with such coverage would be
overly ambitious, and the book would become obsolete very quickly.
This book starts with elements of computer architecture, microcomputers, microcontrollers, and finally embedded hardware and how to
interface it to the microcontroller or microprocessor.

1.3 Everyday Examples of Embedded Hardware

Earlier a handful of examples were mentioned. One can visualize a
larger number of embedded systems when one thinks about the
various industries and their applications. The automobile industry

01_Lipiansky_Ch01_p001-012.indd 4

10/27/11 5:36:34 PM

Introduction to Embedded Systems

has many applications of embedded systems; perhaps hundreds of

examples abound in the consumer electronics and entertainment
industries. Additionally we find many applications in industrial
controls, metallurgy, and avionics. Medical sciences have a large
number of applications of embedded hardware, such as hearing
aids, pacemakers, respirators, liquid and gas chromatographs, and
laboratory automation equipment. The list can go on and on. Basically just about every application or industry has or may have an
embedded system and hardware solution to solve a particular problem. Table 1.1 does not intend to be a complete list of industries or

Industry

Embedded Hardware or Device

Automobile

OBD systems, ABS braking systems,

safety systems

Civil engineering

Automated cranes, automated levelers,

heavy-duty earthmoving equipment

Consumer electronics

DVD, Blu-ray players

Movie and photographic cameras
Cellular, wireless, and land telephones
Calculators, smart power tools
GPS

Electrical engineering

Test equipment such as oscilloscopes,

logic analyzers, network analyzers

Entertainment (movie, TV,

theater)

Professional audio and video recording

equipment

Farming

Farming equipment, such a circular tillers

Food industry

Mixers, packaging equipment,

programmable ovens

Office equipment

Calculators, interoffice communication

systems, video conferencing equipment

Manufacturing

Robots for assembly, painting, washing,

and drying equipment

Life sciences, medicine,

chemistry, and
biochemistry

Automated blood analyzers, culture

analyzers, chromatographs, glucose
and lactate analyzers, and electronic
microscopes

Physical sciences

Atomic particle accelerators, telescopes

Retail

Cash registers, inventory control, price

tag scanning systems, vending machines

Table 1.1 Some Industries and Examples of Their Embedded Hardware

Devices

01_Lipiansky_Ch01_p001-012.indd 5

10/27/11 5:36:34 PM

Chapter One

embedded hardware examples. It is just a starting point to allow the

reader to develop an appreciation of the large number of embedded
system examples that exist in just a few industries.

1.4Embedded System Design Process:

The Hardware Perspective
Embedded system design is a process that requires, at least initially, continuous refinement, sometimes redefinition, and possibly multiple iterations. The single most difficult question for the
engineer is, What do we want to design? This task is not just in the
hands of the hardware and software engineers; it is also driven
greatly by marketing engineers, many times with inputs from customers. So the initial document that needs to be created is what the
electronic industry refers to as the Product Requirements Document, or PRD. Typically drafted by marketing with some engineering guidance, this document contains the key features, functions
of the to-be-designed product, characteristics, limitations, applications, configurations, and the national and international agency
requirements that the product has to comply with, in order to be
sold throughout the world.
This section is primarily focused on the processes that follow
once the PRD is available. This section addresses the general
steps that need to be followed from cradle to grave of the product. These steps are the conception of the product (the mentioned
PRD), engineering specifications, design, implementation, softwarehardware integration, bring-up, testing, debugging, release to manufacturing, sustaining engineering, and sustaining of the product
during its obsolescence period.
Figure 1.4 depicts a simplified flowchart of the engineering
design process. Typically the second most important set of documents that need to be created right after the PRD is the hardware
engineering specification. The hardware specification minimally
contains a general description of the product, list of features, list of
functions, functional block diagrams, detailed description of the
implementation of its various functional blocks, interfacing connectors, and connector pin descriptions. Additionally and at the discretion of the engineering team, a detailed theory of operation section
may be included. Sometimes this is presented as a totally separate
document, referred to as the software guide or software document.
Once the hardware specification is created, the actual design
begins. Hardware design starts on several fronts. Architects along
with software engineers consider the most appropriate processor
or microcontroller to use based on the requirements. The design is
partitioned functionally and physically. Hardware engineers determine how many printed-circuit boards there will be; mechanical

01_Lipiansky_Ch01_p001-012.indd 6

10/27/11 5:36:34 PM

Introduction to Embedded Systems

Product requirements document

Technical specification definition
Schematics design and entry
Printed-circuit board placement & layout
Printed-circuit board fabrication & testing
Electronic components assembly
Assembly test
Engineering bring-up & debug
Hardware-software integration

Design verification & test (DVT)

Engineering verification & test (EVT)
Production verificaton & test
First customer shipment
Full production release
Sustaining engineering
End of life

Figure 1.4 Embedded hardware design: a simplified and generalized

flowchart.

engineers in conjunction with electrical engineers make design

decisions about the enclosure of the product, chassis, or simply the
box. Detailed functional blocks are assigned to various hardware
designers. When an embedded system is designed, usually analog

01_Lipiansky_Ch01_p001-012.indd 7

10/27/11 5:36:34 PM

Chapter One

experts will design the front-end analog signal conditioning circuitry, signal amplifiers, and system decoupling. The digital and
analog experts will work on the hardware interface of the microcontroller with the other functional blocks. Memories and network
interfaces are a few more functional blocks that the system may
require.
Hybrid, analog, and digital designers work on the analog-to-digital
and digital-to-analog interfacing. Other hardware blocks are clocks,
resets, power converters, and decoupling capacitors. Lets assume for
the sake of this example that a single printed-circuit board (PCB) will
contain the core of the design.
A computer-aided design (CAD) schematics capture tool is used
to enter the design. All the circuit components from microcontrollers,
memory chips, network interfaces, reset circuitry, data acquisition
components, decoupling capacitors, and filtering and all other components that need to sit on the PCB are entered into a schematics
database. This process can take from about a week or so to multiple
weeks depending on the complexity and number of circuits that need
to be interconnected. Schematics may vary from as little as a single
page to a few hundred pages. The components count may be in the
several thousands for very complex systems. Upon completion of
the schematics, not only does the engineering team who designed
them review them, but also several cross-functional groups participate in design reviews. Firmware and software will be mainly interested in the correct hardware support of the functions that the software
needs to implement. A manufacturing group will be interested in the
design following manufacturability guidelines, and a test group will
ensure that the design is testable. In parallel, other groups such as
purchasing or commodities work on the availability, lead time, cost,
and ultimately procurement of the components needed for a predetermined number of prototypes.
The next step is the physical design phase of the design. Within
this phase the PCB layout phase takes place. The first step is to produce a physical placement of all the electronic components on the
board. The components placement is by no means arbitrary. Fundamentally, functionality, electrical, timing, signal integrity, manufacturability, testing, and electromagnetic abatement considerations go
into producing a good PCB component placement. Greater discussion of those factors can be found in Chap. 12. Once the placement is
done, the routing of the copper traces that interconnect component
pins to other component pins and connectors begins. The routing,
like the placement, is also greatly driven by electrical, timing, signal
integrity, and electromagnetic interference abatement requirements
and considerations. It is important to observe that a good placement
is fundamental for a successful PCB design. If the placement is done
poorly, it will be next to impossible to have a good functional PCB.

01_Lipiansky_Ch01_p001-012.indd 8

10/27/11 5:36:34 PM

Introduction to Embedded Systems

Some design engineers state that a good placement counts for more
than 70% or 80% of the design. Moreover, once the placement is good,
the routes pretty much fall into place. The physical design is done
with layout CAD tools that allow the PCB designer to do placement
and routing. Once the board file that contains all the physical information and the board are complete, the file is sent to a PCB manufacturer who fabricates the printed-circuit board described by its board
file. The PCB is just the physical structure where mostly electronic
and mechanical components are held together in place not only for
mechanical reasons but for electrical functionality as well. The blank
or unpopulated PCB is checked for basic connectivity, such as opens
and shorts.
After the board is fabricated, an assembly house populates the
PCB with all its components. Typically components are wave-soldered
to the PCB by automated machines; some are press-fit and sometimes
a few components are hand-installed. Electronic designers and manufacturers try to minimize the need for any hand operations in order
to achieve higher manufacturing volumes per time unit.
The bring-up is commonly performed by circuit design engineers
or by bring-up engineers in large organizations. The engineer brings
up the board little by little. Typically power is the first item on the
bring-up list, since nothing else would work without power. Once
power is reasonably stable, i.e., it works, the reset and clock circuit
verification follows. Every board subsystem is checked in an incremental fashion and without firmware running on the machine. The
idea is to validate the correct operation of the most fundamental
hardware functions. Without any firmware or software the testing
cannot go on much longer. There is a need to begin verifying the
board operation with some minimal amount of firmware and later on
application software. This part of the bring-up is also referred to as
hardware-software integration. During this phase, hardware and
software engineers work very closely to verify step by step the functional features of the board. This step, depending on the complexity
of the product, may last from days to weeks or even months. Once a
PCB correctly performs at least most of the critical hardware functions required by the product, the software engineers continue to
bring up and test the board. In many practical situations the software
and firmware are not fully developed, so that testing, bring-up, and
code development may take place concurrently. At times there are
setbacks and the hardware needs to be corrected. Once a handful of
prototype boards are believed to be working reasonably, two more
verification phases ensue. A larger number of prototypes are built,
and the engineering verification and test (EVT) phase begins.
When the board has some mature level of functionality, component placement is considered stable and automatic test equipment
(ATE) development begins. ATE typically involves checking that the

01_Lipiansky_Ch01_p001-012.indd 9

10/27/11 5:36:35 PM

10

Chapter One

inputs and outputs of all electronic components are connected as

expected or as described by the board schematic diagram. ATE also
verifies the components that must be installed as well as those that
are not to be installed on the board. Some components may not be
installed, e.g., because some circuit boards may be assembled with
more or fewer functional features than others. This means the same
PCB design supports a basic model as well as one or more enhanced
models with additional features.
The EVT prototypes are generally the responsibility of the design
engineering organization, and software and hardware engineers do
the functional validation. A subsequent phase with even a larger number of prototypes is built. This time, the prototypes are built by a manufacturing organization; however, it is still customary for the hardware
and software designers to continue working on this functional verification. This phase is commonly referred to as production verification and
test (PVT) phase. Throughout the validation processes there may be a
need to fabricate a second or third version of the PCB to correct initial
errors in the hardware design. When a new board is implemented, the
process basically brings up and hardware-software integration restarts,
although the bring-up of a newer revision board usually moves faster
than it did with the first prototype. After the product reaches a certain
level of confidence and maturity with various organizations, such as
hardware, software, manufacturing, and test, the product is released to
manufacturing. At this point very large numbers of products are built
and tested; then after a greater confidence is obtained, the product goes
into FCS or first customer shipment (FCS). During the lifetime of the
product the manufacturing organization handles the assembly and
testing of the products. A sustaining engineering organization, usually
not the original designers, supports the product. Sustaining engineering handles the need to continue to improve the correct functionality of
the product as well as its manufacturability. Electronic components
may become obsolete because of decisions of their chip manufacturers,
and newer components need to replace the obsolete ones. Typically the
product remains in production for some years. This time continues to
shrink as newer technologies and newer products are developed. After
its useful product lifetime, the system company may decide to announce
the product end-of-life (EOL). At this time the support of the product is
scheduled to end within a period of time, e.g., six months to a year. At
that time the defective product is replaced or upgraded with a newer
product.

1.5 Summary
This chapter describes the fundamentals of embedded hardware at a
high level. Its intent is to present to the reader the general scope of
where hardware fits within embedded design. Most of the available

01_Lipiansky_Ch01_p001-012.indd 10

10/27/11 5:36:35 PM

Introduction to Embedded Systems

11

literature covers embedded systems largely from an embedded software design standpoint. Our intent is to provide the essential elements of hardware design to system software and other high-level
embedded system designers. The simplified and generalized chart
given in Fig. 1.4 is just a guideline or general document to give an
idea of the hardware development phases that exist in a hardware
design and how they interrelate with the software development.
From company to company and from industry to industry, clearly
variations and deviations are expected. In the design and development of a product, quality, schedule, and cost are three key factors that
drive the design process. In an ideal world, all three are equally
important; however, the most successful product manufacturers
know how to prioritize each of those three factors throughout the
development process, to obtain the highest possible quality in a product produced at a reasonable cost within the intended schedule.

1.6 References
1. Tim Wescott, Applied Control Theory for Embedded Systems, Newnes, 2006.
2. Jack Ganssle, Tammy Noergaard, Fred Eady, Lewin Edwards, David J. Katz, Rick
Gentile, Ken Arnold, et al., Embedded Hardware, Newnes, 2008.
3. Raj Kamal, Embedded Systems: Architecture, Programming and Design, McGrawHill, 2008.
4. Tammy Noergard, Embedded Systems Architecture: A Comprehensive Guide for
Engineers and Programmers, Newnes, 2005.
5. James K. Peckol, Embedded Systems: A Contemporary Design Tool, Wiley, 2008.

01_Lipiansky_Ch01_p001-012.indd 11

10/27/11 5:36:35 PM

01_Lipiansky_Ch01_p001-012.indd 12

10/27/11 5:36:35 PM