Chapter 2

What are Cyber Physical Systems (CPS)?

 A Cyber-Physical System (CPS) is a system that integrates physical and
computational components to monitor and control the physical processes seamlessly. These
systems combine the sensing, actuation, computation, and communication capabilities, and
leverage these to improve the physical systems' overall performance, safety, and reliability.
 A cyber-physical system is a collection of computing devices communicating with
one another and interacting with the physical world via sensors and actuators in a feedback
loop. Examples include CPS includes self-driving cars.
 Figure : Abstract structure of Cyber-physical Systems

Embedded System vs Cyber-Physical System

Aspect Embedded System (ES) Cyber-Physical System (CPS)
Meaning Small computer built into a Big system where computers, networks,
device to control it. and physical processes work together.
Focus One device / function. A network of devices + communication +
intelligence.
Communication Mostly stand-alone, Connected (IoT, Internet, Cloud, 5G).
limited links.
Intelligence Pre-programmed, fixed Smart, adaptive (AI, data analytics,
tasks. feedback).
Real-Life Washing machine controller Smart laundry system – machine connects
Example 1 – controls motor speed, to phone, adjusts wash mode by fabric
wash cycles. type, orders detergent online.
Real-Life Car’s ABS braking system Self-driving car – integrates sensors, GPS,
Example 2 – ensures wheels don’t AI, and communication with traffic
lock. systems.
Real-Life Digital camera – captures Smart surveillance system – cameras
Example 3 and stores photos. connected to cloud, use AI for face
recognition and alerting.

Embedded System = Brain of one device.

Cyber-Physical System = Many brains + communication +
intelligence working together.
IOT VS CPS :

Architectural overview of Cyber-Physical Systems (CPS)

Cyber-physical systems (CPS) have an architecture that integrates hardware and software
components. Because of this architecture, the dynamic interaction between physical processes
and computational elements becomes possible. This forms the foundation for the intelligent
and responsive behavior of CPS. A CPS architecture consists of the following:
1. Physical Process: The real-world system or environment that CPS interacts with.
2. Sensors: Devices that collect data (e.g., temperature, pressure, motion) from the
physical process.
3. Communication Network: Wired or wireless infrastructure that transmits data
between the physical and cyber domains.
4. Computational Nodes: Devices (e.g., microprocessors, servers) that process and
analyze data from sensors.
5. Actuators: Devices that control the physical process based on commands from the
computational nodes.
6. Control Algorithms: Software that analyzes sensor data, makes decisions, and
generates control signals.
 Figure : The Architecture of CPS
Examples Of Cyber-Physical Systems
The following are the cyber physical systems examples, giving a better understanding of the
utilization of CPS in different industries:
1. Operational Technology (OT): This is a combination of hardware and software
systems that monitor and control physical devices and processes in industrial settings.
OT is important for managing infrastructure like power plants, manufacturing
facilities, and transportation systems.
2. Industrial Internet of Things (IIoT): The IIoT refers to the network of
interconnected sensors, instruments, and other devices used in industrial applications.
It enables real-time data collection, analysis, and automation, leading to increased
efficiency, productivity, and safety in various industries.
3. Smart Manufacturing: In smart factories, robots collaborate with humans.
Algorithms analyze data to predict maintenance needs, preventing failures.
4. Autonomous Vehicles: Self-driving cars and trucks use sensors and AI for
navigation.
5. Smart Grids: The physical power grid infrastructure is interconnected with digital
systems for real-time monitoring and control.
6. Medical Devices: Implantable devices like pacemakers use sensors to monitor and
adjust therapy automatically. These devices integrate sensors and actuators within the
human body to maintain physiological functions.
7. Building Automation: Smart buildings use CPS to control HVAC, lighting, and
security systems.
8. Aerospace and Defense: Drones, missile defense systems, and autonomous
underwater vehicles are all examples of CPS used in these sectors.
9. Traffic Management: Adaptive traffic signals and intelligent transportation systems
use CPS to optimize traffic flow and improve safety.
10. Process Control: Industrial Control Systems (ICS) incorporate CPS to monitor and
control complex processes in critical infrastructure, ensuring safety and efficiency.
Features Of Cyber-Physical Systems
Cyber-physical systems (CPS) have distinct characteristics that set them apart from computer
systems and embedded devices. These features drive their transformative impact across
various domains:
1. Integration of Cyber and Physical Components: CPS are not merely software
running on hardware; they are integrated systems where the cyber and physical
components are deeply intertwined. This integration allows for real-time interaction
and feedback between the digital and physical worlds.
2. Real-Time Operation: CPS operates in real time and responds to changes in the
physical environment with minimal delay. This is important for applications that
require immediate action, such as autonomous vehicles or industrial control systems.
3. Networking and Communication: CPS are connected through networks. This
enables them to exchange data, coordinate actions, and learn from each other. This
interconnectedness leads to scalability and adaptability.
4. Adaptability and Autonomy: CPS adapts to changing conditions and operates
autonomously to some extent. CPS can make decisions based on real-time data, learn
from their experiences, and optimize their behavior over time.
5. Heterogeneity: CPS consists of different components, like sensors, actuators,
processors, and communication devices. This heterogeneity requires sophisticated
integration and coordination mechanisms for seamless operation.
6. Data-Driven Decision Making: CPS relies on data collected from sensors and other
sources. They use this data to make informed decisions and adapt to changing
conditions.

Cyber-Physical Systems (CPS): Faces of Engineering

CPS are smart systems where the physical world (machines, devices, processes) interacts
closely with the cyber world (computation, communication, control, AI).
They are the backbone of modern technologies such as autonomous vehicles, smart grids,
healthcare monitoring, industrial robots, and IoT.

To build CPS, we need many faces of engineering, meaning the different engineering
domains/perspectives that come together:

1. Computer Science / Software Engineering (Cyber Face)

1. Embedded software, programming, and algorithms.

2. Artificial intelligence (AI) and machine learning for decision-making.

3. Real-time operating systems and simulation tools.

2. Electrical / Electronics Engineering (Electronics Face)

1. Sensors and actuators to connect cyber and physical worlds.

2. Microcontrollers, processors, and VLSI systems.

 3. Signal processing for data acquisition and conditioning.

3. Mechanical Engineering (Physical Face)

1. Robotics, mechatronics, and control systems.

2. Dynamics of machines and processes.

3. Design of physical structures interacting with cyber systems.

4. Communication / Networking Engineering (Connectivity Face)

1. IoT and wireless communication (5G, Wi-Fi, Bluetooth).

2. Cloud–edge computing integration.

3. Reliable, low-latency networks for real-time control.

5. Systems Engineering (Integration Face)

1. Modeling and simulation of whole CPS.

2. Reliability, safety, fault tolerance, and performance evaluation.

3. Human–machine interaction and ergonomics.

6. Data Engineering & Cybersecurity (Information Face)

 Handling big data from sensors and IoT devices.

 Securing CPS against cyberattacks.

 Privacy protection and trustworthy computing.

Making Processor Chips from Sand

[Link]

[Link]

How are Microchips Made? CPU Manufacturing Process

Steps
[Link]
Making Processor Chips from Sand
1. Starting Material – Sand (Silicon Dioxide, SiO₂)

 The main element used in chips is silicon, and silicon comes from sand (quartz).

 Sand is purified to produce very high-purity silicon (over 99.9999%).

2. Purification of Silicon

 Raw sand is chemically processed to remove impurities.

 The result is polycrystalline silicon (chunks of pure silicon).

3. Formation of Silicon Ingots

 Pure silicon is melted in a furnace.

 A small seed crystal is dipped into the molten silicon and slowly pulled out while
rotating → this grows into a large cylindrical single crystal ingot (Czochralski
process).

 Ingot size: often 300 mm (12 inches) diameter.

4. Cutting into Wafers

 The ingot is sliced into very thin discs called wafers (as thin as a credit card).

 Wafers are polished to be perfectly flat and mirror-like.

5. Photolithography – Creating Circuit Patterns

 Wafers are coated with light-sensitive material (photoresist).

 UV light passes through a mask (like a stencil) to transfer the microscopic circuit
pattern onto the wafer.

 This is repeated many times to build different layers of the chip.

6. Doping and Etching

 Doping: Adding controlled amounts of other elements (boron, phosphorus) to silicon

to change its electrical properties (making p-type and n-type semiconductors).

 Etching: Removing unwanted silicon or photoresist to form tiny transistors, wires, and
components.

7. Layering and Interconnection

 Many layers of transistors, metal wiring, and insulators are built up.
  Modern chips have billions of transistors layered in a wafer only a few millimeters
thick.

8. Testing and Dicing

 The completed wafer is tested.

 It is then cut into individual chips (dies).

9. Packaging

 Each chip is placed in a protective package with connectors (pins or solder bumps).

 The package allows the chip to be mounted on circuit boards and connected to other
components.

10. Final Product

 The result is a processor chip (like Intel, AMD, or ARM processors) that powers
computers, phones, and countless devices.

[Link]

Wafer Manufacturing
The semiconductor chip manufacturing process begins with the creation of silicon wafers,
which are made of sand. These wafers are typically considered the foundation for any
semiconductor chip. Producing silicon wafers from tiny grains of sand (also known as
Silicon dioxide (SO2) requires quite a sophisticated process:
1- First, sand is heated until it melts into a high-purity liquid and then gets solidified by
crystallization.
2- After getting the resultant of high-purified silicon pillars, also called ingot cylindricals,
these ingots are then sliced into ultra-thin, precise size wafers using advanced diamond saws.
With the technology development, typical wafer thicknesses now range from 675 to 725
micrometers.
3- Wafer slices then remain rough and contain defects after cut can affect the electronic IC
performance, in terms of precision and function. Therefore, wafers undergo precision
polishing to achieve a flawless, mirror-like surface.

Why call them “wafer”? The pattern grinds on the surface with the appearance of a wafer-like
is the most simple, yet obvious answer. Depending on the size of the wafer, the number of
chips produced is indicated. In simple words, larger equals better capability.
Oxidation
The results wafers are not yet semiconductive to be used. To enhance the electrical properties
of the silicon wafer, a thin layer of silicon dioxide (SiO2) is formed on its surface through
oxidation. This oxidation stage in the semiconductor chip manufacturing process involves

1- Exposing the wafer to high temperatures in the presence of oxygen or water vapor (also
known as dry and wet oxidation – the most two common types of oxidation to create a thin,
dense layer of silicon dioxide (SiO2) on the wafer’s surface
2- The thickness oxide layer directly affects the performance of the chip therefore, it need to
be highly controlled. Modern oxide thickness can now remain at a few nanometers.
3- The silicon dioxide layer serves as an insulator a protective shield, preventing unwanted
electrical leakage between different chips on the wafer.
Photolithography
Photolithography is a key step in defining the intricate patterns that form the chip’s circuitry.
In other words, it likes “printing” the circuit design onto the wafer, using a technique similar
to creating a film camera photo.
The photography process:
1- First, a thin layer of light-sensitive photoresist is applied to the wafer
2- Then, a photomask – a glass substrate containing the desired circuit pattern design is
prepared to selectively expose the photoresist.
3- Let light transfer through the photomask, imprinting the chip design on the wafer surface
Make sure to conduct an additional check to ensure the circuits are drawn well. This step
defines the locations for later stages of the semiconductor manufacturing process like etching
and deposition.
Etching
The next step in the semiconductor manufacturing process is called etching.
As the chip design is printed on the wafer surface in the photolithography process, it is now
time to remove unnecessary material from the wafer’s surface according to the desirable
patterns defined by photolithography.
There are two common methods of etching: wet etching and dry etching
Wet etching – A chemical solution, using liquid reactive chemicals to dissolve unwanted materials
Dry etching – Solution using plasma, using ionized gases to selectively remove unwanted material
It can be said that this step shapes the various components of the chip, including transistors
and interconnects, set the fundamental for the miniaturization of electronic IC, which
indicates the performance, efficiency and cost.
Deposition
In a chip manufacturing, photolithography and etching process are repeated serveral times,
thin layers by thin layers on the wafer. To build the chip’s components, thin layers of
materials such as metals and insulators are deposited onto the wafer’s surface.

Chemical Vapor Deposition (CVD) and Physical Vapor Deposition (PVD) are common
techniques. These coating layers contribute to the formation of transistors and interconnects,
with each layer carefully designed to meet specific performance requirements.
Ion Implantation
Semiconductor made of silicon doesn’t conduct electrical characteristics. Therefore, this is
the time for ion to be implemented, by adding impurities, altering the ability to conduct
electrical properties.
This precise doping process is essential for creating the necessary conductivity variations, cn
be considered as the ending phase to make the wafer conductive with a number of chip drawn
on it.
Metal Wiring
The addition of metal wires is crucial for establishing electrical connections, creating
pathways between various components of the chip. These metal layers, typically made of
materials like aluminum, titanium, or stungsten serve as the conductive pathways for
electrical signals to flow in the circuit, ensuring efficient communication within the
semiconductor device.

Note: Among those above materials, Aluminum is the most used option as it adheres well to
the oxide layer (silicon dioxide) and is easily workable.
Sorting
Once the manufacturing steps are completed, wafers undergo rigorous testing to identify and
sorting out defective chips. This stage is also called EDS, or Energy Dispersive Spectroscopy.

Automated testing equipment assesses each chip’s functionality, ensuring that only fully
operational units progress to subsequent stages. This step is critical for maintaining high
yields and quality in semiconductor production.
Packaging
Coming to the last phase in the semiconductor chip manufacturing process, the wafer is diced
into individual chips. Those chips are then encapsulated in protective casings, connecting it to
external pins or leads, enabling electrical exchange with the outside, and ensuring effective
heat dissipation. Packaging not only protects the chip from various environmental factors but
also influences its overall performance and reliability.

After final test, sealing and labeling of the product name, manufacturer name, the
semiconductor chip we commonly see is completed.
Structural units in Embedded Processor
The processor in any embedded system contains various units for its successful and fast
operation. Buses, memories, processing units, instructions, registers, control units are some of
the structural units that an embedded processor should possess.

Internal Buses
It internally connects all the structural units inside the processor. Its width can be of 8, 16, 32
or 64 bits. The internal buses include address bus, data bus and control bus. All the internal
units of the processor are communicated with the external buses through an interface unit
called Bus Interface Unit(BIU).
 Address bus is an external bus that carries the address from the Memory Address
Register(MAR) to the memory as well as to the IO devices and other units of the
system.
 Data bus is an external bus that carries the data from or to the address determined by
Memory Address Register(MAR).
 Control Bus is an external bus that carries, control signals to or between the processor
and memory.
Instruction Units
The processor executes several operations with the help of a piece of code or program called
instructions. All these instruction are arranged in a queue called Instruction Queue(IQ). This
helps the Instruction Register to execute the instructions without the need to wait.
Instruction Register(IR) takes the instruction codes(opcodes) sequentially to the execution
unit of the processor. Instruction decoder(ID) decodes the instruction opcode received at the
Instruction Register and passes it to the processor Control unit for its execution..

Caches
In general, cache is the temporary storage place in the processor, where the data can easily be
retrieved by the CPU for quick operation. Cache is available in your computers, laptop,
mobile phone, browser, app, etc. The embedded processor has three cache memory for fast
operation.
Similar to an instruction queue, Instruction Cache(I-Cache) sequentially stores the
instructions in FIFO mode. It lets the processor to execute the instructions at a greater speed.
Branch Target Cache(BT Cache) facilitates the ready availability of the next instruction set,
when a branch instruction like jump, call or loop is encountered. Its fetch unit foresees a
branching instruction in the I – cache.
Data Cache(D – Cache) stores the pre – fetched data from the external memory. A data cache
holds both the address and data together at a location. It also stores the write-through data
that is to be transferred to the external memory addresses. Write-through data is nothing but
the output data from the execution unit.
Control Units
The control unit(CU) is one of the important structural units in an any embedded processor. It
is responsible for all the control of the bus activities and unit functions needed for processing.
It directs all the units of a processor to respond to the instruction that is executed.
Pre fetch control Unit(PFCU) controls the fetching of data into the I-Cache and D-Cache in
advance from the memory unit. The instructions and data are delivered when needed to the
processor’s execution units.
Memory Management Unit(MMU) manages the memories such that the instructions and data
are readily available for processing. There are several memory management methods for
managing the memory in an embedded processor such as fixed block allocation, dynamic
block allocation, dynamic page allocation, etc.
Processing Units
Processing units receives the input data, process it and produces the output. Depending on the
function of the embedded system, the processing is done through different units like ALU,
FLPU, AOU and advanced units.
Arithmetic Logic Unit(ALU) is used for the execution the arithmetic and logic instructions
according to the instruction present at the Instruction Register.
Floating point processing unit(FLPU) is different from ALU, used primarily for floating
point processing. It is essential for fast processing mathematical functions in a
microprocessor or Digital Signal Processor.
Atomic operation unit(AOU) lets a user (compiler) instruction when broken into number of
processor instructions called atomic operations, finish before an interrupt of the processor
occurs.
The embedded processor also has Advanced processing units for multistage pipeline
processing, multi-line superscalar processing to obtain processing speeds higher than one
instruction per cycle.
Register Sets
Application Register Set(ARS) is a set of on-chip registers used during processing of
instructions of the application program of the user. A register window consists of a subset of
registers with each subset storing static variables of software – routine.
System Register Set(SRS) is a set of registers used while processing the instructions of the
supervisory system program.
Floating point register set(FRS) is dedicated for storing floating point numbers in the
standard format and used by Floating point Processing Units for its data.
Memory Registers
Registers are smallest part of the CPU, where the data and address can be stored and accessed
quickly. There are various registers like accumulator, general purpose registers, flag registers,
memory address register, memory data register, program counter, stack pointer, etc.
Memory Address Register(MAR) holds the address of the data byte or word that is to be
fetched from external memories. Memory Data Register(MDR) holds the data byte or word
that is fetched from external memory or stored to the external memory.
Program Counter and Stack Pointer
Program counter is a register that stores the address of the next instruction to be executed
from the memory. It generates the instruction cycle to fetch the address from memory through
MAR.
Stack pointer is a register that stores the address of the last program request in a stack. It is a
pointer for an address which corresponds to a stack top in the memory.

Sensors and Actuators in Cyber Physical Systems

Sensors are devices that transform physical or environmental values into electrical impulses.
Not only are sensors widely used in cyber-physical systems and the Internet of Things, but
they are also present in many independent electronic gadgets.
Take your mobile phone for instance, there are multiple sensors embedded in it. The
fingerprint sensor to unlock your phone, Google Voice assistance that gets activated based on
voice commands, radio sensors, and many more. With various applications of sensors, it is
crucial to understand their role, especially in CPS (Cyber-Physical systems).
What are sensors and actuators?
 Sensors are devices that transform physical or environmental values into electrical
impulses. They are essential for gathering data from the actual world and making it
possible for machines to communicate with and comprehend their environment.
 Conversely, actuators are machines that convert electrical information or orders into
mechanical motion. They oversee carrying out commands from a controller or
computer system to make things happen in the real world.
 Basically, a system processes the data that sensors gather from their surroundings. The
controller instructs actuators on what action to take by sending signals to them based
on this data. The ability of machines to perceive, interpret, and react to their
environment is made possible by the smooth communication between sensors and
actuators. Cyber-physical systems, which are essential to contemporary automation
and technology, are built on this foundation.
How does it impact CPS?
 CPS is a sophisticated combination of physical elements that interact directly with the
outside environment and computing algorithms. These systems use sensors and
actuators to collect information from their surroundings, process it utilizing
calculating factors, and further respond by impacting or managing physical processes.
 The backbone of the functioning of CPS is the integration of sensors and actuators.
These systems are based on a dynamic feedback circle in which sensors collect data
from the real world and a computing system processes it. The regulator of the system
uses this data to give commands to the actuators, telling them what to do. CPS is
made up of interdependent digital, analog, physical, and human factors that are
designed to work together using integrated physics and logic.
 CPS is incredibly intelligent and adaptive because of this coordination between
sensors and actuators. They diligently shape and impact the environment around them
rather than simply gathering data. Whether in transportation, manufacturing,
healthcare, or any other industry.
 The capacity of sensors in CPS to deliver real-time environmental data is fascinating.
Sensors are always collecting data on a range of characteristics, including stir,
pressure, temperature, and more. The regulator of the system also processes this data
in real-time. They also help make better use of resources in industries like
automobiles and healthcare.
