UNIT-I | Computer Networks & the Internet Exam Study Guide

Complete Exam Study Guide

Contents
To
pic Section

1 Introduction to Computer Networks

2 Protocols – Syntax, Semantics, Timing

3 What is the Internet?

4 Network Edge & Network Core

5 Delay, Loss & Throughput

6 Reference Models (OSI & TCP/IP)

7 Transmission Media

8 Circuit vs Packet Switching

9 Standards Bodies & ISP Hierarchy

© Computer Networks Study Notes Page 1

UNIT-I | Computer Networks & the Internet Exam Study Guide

1. Introduction to Computer Networks

■ Definition
A computer network is a collection of interconnected computing devices (nodes) that can communicate
and share resources with each other using wired or wireless communication channels following
agreed-upon rules called protocols.

■ Goals of Computer Networks

Goal Explanation

Resource Sharing Share hardware (printers), software and data among users without physical
proximity.

Reliability Multiple copies / redundant paths ensure continued service even if one node fails.

Cost Reduction Cheaper to share resources than equip every user independently.

Communication E-mail, instant messaging, video conferencing, VoIP.

High Speed Data Access Fast retrieval of remote databases, cloud storage, streaming.

Scalability Easy to add more nodes without redesigning the entire network.

Centralised Management Administer software and security policies from one location.

■ EXAM TIP – Goals

• Remember 7 goals: Sharing, Reliability, Cost, Communication, Speed, Scalability, Management

• Use mnemonic: 'SRCCSSM' – Sometimes Robots Communicate Clearly, Sharing Smart Methods

■ Benefits of Networking
• Data sharing – files, databases accessible from anywhere
• Hardware sharing – one printer/scanner for many users
• Faster communication – instant transfer across continents
• Centralised backup – reduced risk of data loss
• Collaboration – real-time co-editing, video calls

■ Types of Networks (Classification)

Typ
e Full Name Span Examples

PAN Personal Area Network <10 m Bluetooth, USB

LAN Local Area Network Building / Campus Ethernet, Wi-Fi

MAN Metropolitan Area Network City / Town Cable TV, Metro Ethernet

WAN Wide Area Network Country / Globe Internet, MPLS, ATM

© Computer Networks Study Notes Page 2

UNIT-I | Computer Networks & the Internet Exam Study Guide

2. Protocols – Definition, Key Elements & Importance

■ Definition of a Protocol
A protocol is a set of rules and conventions that govern how two or more communicating entities
exchange data. Just like human conversation has rules (greet, speak one at a time, respond), networks
need protocols.

Formal definition: A protocol defines the format and the order of messages exchanged between two or
more communicating entities, as well as the actions taken on the transmission and/or receipt of a message
or other event.

■ Three Key Elements of a Protocol

Element Meaning Example

Syntax Structure / format of the data. Specifies IP header: 4-bit version field, 8-bit
how many bits make a field, what order TTL, 32-bit address
fields appear.

Semantics Meaning of each section of the data. What TCP SYN bit = initiate connection;
action to take when certain bits are set. ACK bit = acknowledge receipt

Timing Speed matching & sequencing. When to Flow control, retransmission timers,
send data, how fast, and what order. sequence numbers

■ Memory Trick – Protocol Key Elements

• SST → Syntax (Structure), Semantics (Meaning), Timing (When/Speed)

• Think of it like a language: Grammar=Syntax, Dictionary=Semantics, Conversation-speed=Timing

■ Importance of Protocols in Networking

• Interoperability: Different vendor devices can talk to each other
• Error Detection & Correction: Protocols include checksums, ACKs
• Flow Control: Prevents fast sender from overwhelming slow receiver
• Security: TLS/SSL protocols ensure encrypted communication
• Routing: BGP, OSPF protocols direct packets to destinations
• Standardisation: IEEE, IETF, ISO define protocols globally

■ Common Protocols & Their Layers

Protocol Layer Purpose

HTTP/HTTPS Application Web browsing

FTP Application File Transfer

SMTP/POP3/IMAP Application Email

DNS Application Domain to IP resolution

© Computer Networks Study Notes Page 3

UNIT-I | Computer Networks & the Internet Exam Study Guide

TCP Transport Reliable, connection-oriented data delivery

UDP Transport Fast, connectionless data delivery

IP (IPv4/IPv6) Network Logical addressing & routing

ICMP Network Error reporting (ping)

ARP Data Link IP to MAC address resolution

Data Link /
Ethernet Physical LAN communication

© Computer Networks Study Notes Page 4

UNIT-I | Computer Networks & the Internet Exam Study Guide

3. What is the Internet?

■ Two Views of the Internet

View 1 – Nuts-and-Bolts (Hardware/Software)
The Internet is a global network of networks connecting billions of computing devices: PCs, servers,
smartphones, IoT sensors, routers, switches, etc.
• End systems (hosts) – run network applications at the 'edge'
• Communication links – fibre, copper, radio, satellite
• Packet switches – forward packets (routers & link-layer switches)
• ISPs (Internet Service Providers) – connect end systems to Internet
• Protocols – TCP, IP, HTTP, SMTP, 802.11 Wi-Fi etc.
• Internet Standards – defined by IETF (RFC documents)

View 2 – Service (Infrastructure)

The Internet is an infrastructure that provides services to distributed applications (Web, email,
streaming, games, e-commerce).
• Provides socket interface – programming API for applications
• Applications exchange messages using Internet protocols

■ Key Internet Components

Component Description Example

Host / End system Runs applications PC, phone, server

Router Forwards IP packets between networks Cisco, Juniper routers

Switch Connects devices within LAN Ethernet switch

Modem Converts digital↔analog signals DSL/Cable modem

Access ISP First mile connection to users Jio, Airtel, Comcast

Backbone ISP Tier-1, global high-speed links AT&T;, Sprint, NTT

IXP Internet Exchange Point – ISPs peer DE-CIX Frankfurt

CDN Content Delivery Network – caches content Akamai, Cloudflare

© Computer Networks Study Notes Page 5

UNIT-I | Computer Networks & the Internet Exam Study Guide

4. Network Edge & Network Core

■ The Network Edge

The network edge consists of all the end systems (hosts) — the devices at the periphery of the Internet
that run application programs.
• Includes: PCs, laptops, smartphones, servers, IoT devices
• Can act as clients (request services) or servers (provide services)
• Connected to the Internet via access networks

Access Network Types

Type Technology Speed Who Uses?

Home DSL (telephone line) Up to 24 Mbps Residential

Home Cable (HFC) Up to 1 Gbps Residential

FTTH (Fibre to
Home Home) 1+ Gbps Residential

Home 5G Fixed Wireless 100+ Mbps Residential

10 Mbps–10
Enterprise Ethernet (LAN) Gbps Offices, Universities

54 Mbps–9.6
Enterprise Wi-Fi (802.11) Gbps Offices, Cafes

Mobile 3G/4G LTE 1–300 Mbps Smartphones

Mobile 5G NR 1–20 Gbps Smartphones, IoT

■ The Network Core

The network core is the mesh of packet switches and links that interconnect the Internet's end systems.
Two fundamental approaches exist:

Packet Switching Circuit Switching

Data split into packets. Each packet travels A dedicated end-to-end circuit is established
independently through the network. Routers use before communication. Resources are reserved
store-and-forward — receive entire packet for the duration of the call.
before forwarding. • FDM – Frequency Division Multiplexing
• Statistical multiplexing of links • TDM – Time Division Multiplexing
• No pre-reservation of bandwidth • Guaranteed bandwidth and quality
• More efficient for bursty data • Wastes capacity when idle
• May cause queuing delays • Used in: Traditional telephone (PSTN)
• Used in: Internet (TCP/IP)

■ EXAM TIP – Switching Comparison

© Computer Networks Study Notes Page 6

UNIT-I | Computer Networks & the Internet Exam Study Guide

• Circuit: Pre-reserved, guaranteed, wasteful, telephone

• Packet: On-demand, efficient, variable delay, Internet
• Store-and-forward delay = L/R seconds (L=packet size bits, R=link rate bps)

Circuit-Switched vs Packet-Switched Networks

Circuit-Switched Packet-Switched

Dedicated reserved path Data broken into packets

Guaranteed bandwidth Shared links – store & forward

© Computer Networks Study Notes Page 7

UNIT-I | Computer Networks & the Internet Exam Study Guide

5. Delay, Loss & Throughput in Packet-Switched Networks

■■ Four Types of Packet Delay

Total nodal delay = d_proc + d_queue + d_trans + d_prop

Delay Type Formula Typical Value Cause

Processing Delay – Microseconds Error checking, routing table

d_proc lookup

Queueing Delay Depends on traffic Microseconds–ms Congestion; packets wait in

d_queue intensity router buffer

Transmission L/R (L=bits, R=bps) Microseconds–ms Time to push all bits onto the link
Delay d_trans

Propagation Delay d/s (d=distance, Milliseconds Physical travel time through

d_prop s=speed) medium (≈2×10■ m/s)

Four Types of Packet Delay

Processing Queueing Transmission Propagation

Delay Delay Delay Delay

Check bits, Wait in buffer Push bits Travel through

route lookup Traffic dependent onto link medium
~microseconds Variable L/R seconds d/s seconds

d_proc d_queue d_trans=L/R d_prop=d/s

■ Delay Formulas to Remember for Exam

• Transmission Delay = L / R (packet length L bits, link rate R bps)

• Propagation Delay = d / s (distance d metres, propagation speed s ≈ 2×10^8 m/s)
• End-to-end delay (N links) = N × (L/R) [store-and-forward, no prop delay]
• Traffic Intensity = La/R (λ=packets/sec, a=avg packet arrival rate) – keep < 1!

■ Packet Loss
When router buffers are full, arriving packets are dropped.
• Lost packets may be retransmitted by sender (TCP) or ignored (UDP)
• Loss probability increases as traffic intensity approaches 1
• Measured as: Packet Loss Rate = lost_packets / total_sent × 100%

■ Throughput
Throughput = rate at which bits are successfully delivered from sender to receiver.
• Instantaneous throughput – rate at a specific moment

© Computer Networks Study Notes Page 8

UNIT-I | Computer Networks & the Internet Exam Study Guide

• Average throughput – total bits / total time

• Bottleneck link – the link with minimum capacity limits end-to-end throughput
Formula: Throughput = min(R_s, R_c) where R_s = server link, R_c = client link

■ EXAM TIP – Throughput & Bottleneck

• End-to-end throughput = minimum bandwidth along the path

• The bottleneck is usually the access network (first/last mile), not the backbone
• Distinguish: Bandwidth (capacity) vs Throughput (actual rate) vs Goodput (useful data rate)

© Computer Networks Study Notes Page 9

UNIT-I | Computer Networks & the Internet Exam Study Guide

6. Reference Models – OSI & TCP/IP

■ Why Layered Architecture?

• Complexity Management: Break the problem into manageable pieces
• Modularity: Change one layer without affecting others
• Standardisation: Each layer has a well-defined interface
• Interoperability: Different vendors implement layers independently

■ OSI Reference Model (7 Layers) – ISO Standard

OSI Reference Model (7 Layers)
Application
7 HTTP, FTP, SMTP, DNS

Presentation
6 Encryption, Compression

Upper Session
5 Session Mgmt, RPC

Transport
4 TCP, UDP (Segments)

Network
3 IP, ICMP (Packets)

Lower Data Link

2 Ethernet, MAC (Frames)

Physical
1 Bits, Cables, Wi-Fi

# Layer PDU Key Function Protocols

7 Application Message Network services to applications HTTP, FTP, SMTP, DNS,

DHCP

6 Presentation Message Data translation, encryption, SSL/TLS, JPEG, MPEG,

compression ASCII

5 Session Message Session establishment, NetBIOS, RPC, PPTP

maintenance, termination

4 Transport Segment End-to-end delivery, flow/error TCP, UDP, SCTP

control

3 Network Packet Logical addressing, routing IP, ICMP, OSPF, BGP

2 Data Link Frame Physical addressing (MAC), error Ethernet, Wi-Fi, PPP
detection

1 Physical Bits Bit transmission over physical NIC, Cables, Hubs,

medium Repeaters

© Computer Networks Study Notes Page 10

UNIT-I | Computer Networks & the Internet Exam Study Guide

■ OSI Layer Memory Mnemonic

• Top to Bottom (7→1): 'All People Seem To Need Data Processing'

• A=Application, P=Presentation, S=Session, T=Transport, N=Network, D=Data Link, P=Physical
• Bottom to Top (1→7): 'Please Do Not Throw Sausage Pizza Away'

■ TCP/IP Reference Model (4 Layers) – DoD Model

TCP/IP Model vs OSI Model – Side by Side
7OSI Model
Application TCP/IP Model
Key Differences
OSI: 7 layers | TCP/IP: 4 layers
Application
6 Presentation OSI: Presentation+Session separate
HTTP FTP SMTP DNS
TCP/IP: Physical+DataLink merged
TCP/IP is practically implemented
5 Session
OSI is a theoretical reference model

Transport
4 Transport
TCP UDP

Internet
3 Network
IP ICMP ARP

2 Data Link
Network Access
Ethernet Wi-Fi
1 Physical

# TCP/IP Layer Maps to OSI Protocols Devices

4 Application OSI 5+6+7 HTTP, FTP, SMTP, DNS, DHCP, Hosts, Servers
SSH

3 Transport OSI 4 TCP (reliable), UDP (fast) Hosts

2 Internet OSI 3 IP, ICMP, ARP, RARP Routers

1 Network Access OSI 1+2 Ethernet, Wi-Fi, PPP, DSL NIC, Switches,
Hubs

■■ OSI vs TCP/IP – Detailed Comparison

Feature OSI Model TCP/IP Model

Full Name Open Systems Interconnection Transmission Control

Protocol/Internet Protocol

Developed by ISO (International Standards Org.) US Department of

Defense (DoD)

Layers 7 4

© Computer Networks Study Notes Page 11

UNIT-I | Computer Networks & the Internet Exam Study Guide

Type Reference / Theoretical model Practical / Implemented

model

Protocol dependency Protocol independent Protocol specific

Session & Presentation Separate dedicated layers Merged into Application

layer

Physical & Data Link Separate dedicated layers Merged into Network
Access layer

Transport layer Provides both connection-oriented & TCP (reliable), UDP (fast)
connectionless

Network layer Connection-oriented & connectionless Connectionless only (IP)

Reliability Less reliable (theoretical) More reliable (widely

deployed)

Usage Teaching & reference Real-world Internet

Standard ISO 7498 RFC 1122, RFC 1123

■ EXAM TIP – OSI vs TCP/IP

• OSI=7 layers, theoretical, ISO, teaches concepts

• TCP/IP=4 layers, practical, DoD, powers the Internet
• TCP/IP merges: App+Pres+Session → Application; Physical+DataLink → Network Access
• Key question: 'Which layer handles routing?' → OSI: Network(3); TCP/IP: Internet(2)

© Computer Networks Study Notes Page 12

UNIT-I | Computer Networks & the Internet Exam Study Guide

7. Guided & Unguided Transmission Media

Transmission Media
Guided (Wired) Unguided (Wireless)
Twisted Pair Radio Waves
UTP/STP, cheap, common AM/FM, Wi-Fi, long range

Coaxial Cable Microwaves

Better shielding, TV/cable Line-of-sight, satellite

Fibre Optic Infrared

Light, fastest, immune to EMI Short range, TV remotes

■ Guided (Wired) Media – Detailed

Medium Bandwidth Distance Advantages Disadvantages

Twisted Pair Up to 10 100 m per Cheap, widely used, easy Susceptible to EMI,
(UTP/STP) Gbps (Cat 8) segment to install limited distance

Coaxial Cable Up to 1 Gbps Several km Better shielding, higher Stiffer, more expensive
bandwidth than UTP than UTP

Fibre Optic Up to 100+ 100+ km Immune to EMI, ultra-fast, Expensive, fragile, hard
(Single/Multi Tbps secure, low loss to splice
mode)

■ Unguided (Wireless) Media – Detailed

Type Frequency Range Examples Issues

Radio Waves 3 Hz – 1 GHz km–global AM/FM radio, Wi-Fi Interference,

(2.4 GHz), Bluetooth eavesdropping

Microwaves 1 GHz – 300 km (LOS) Satellite, cellular 5G, Requires line-of-sight,

GHz radar rain fade

Infrared 300 GHz–400 <10 m TV remotes, IrDA Blocked by walls,

THz short range

Visible Light / 400–700 THz Room Li-Fi, optical Requires clear path,
Li-Fi networking light-sensitive

■ EXAM TIP – Transmission Media

• Fibre Optic = fastest, most expensive, immune to EMI → used in backbone

• UTP = most common in LANs (Cat5e/6/6a/7/8 categories)
• Wireless = convenience vs security & interference trade-off
• Single-mode fibre: 1 light path, long distance; Multi-mode: multiple, shorter range

© Computer Networks Study Notes Page 13

UNIT-I | Computer Networks & the Internet Exam Study Guide

8. Circuit vs Packet Switching – Deep Dive

■ Circuit Switching – In Detail

A circuit-switched network reserves end-to-end resources (bandwidth, buffers) for the entire duration of
a communication session before transmission begins.

Two Multiplexing Methods in Circuit Switching:

• FDM (Frequency Division Multiplexing): Each call gets a dedicated frequency band. All frequencies
used simultaneously. E.g., traditional radio, cable TV.
• TDM (Time Division Multiplexing): Each call gets a time slot in a repeating frame. Only one call
transmits at a time in its slot. E.g., ISDN, T1 lines.

Advantages of Circuit Switching:

• Guaranteed bandwidth – no competition during call
• Predictable and constant delay
• Good for real-time, continuous data (voice calls)

Disadvantages of Circuit Switching:

• Wasted capacity during silent periods (e.g., pauses in conversation)
• Setup time needed before communication begins
• Not suited for bursty data (web browsing)

■ Packet Switching – In Detail

In packet switching, data is divided into discrete chunks called packets. Each packet is transmitted
independently and may take different routes.

Store-and-Forward Mechanism:
Each router must receive the complete packet before forwarding it to the next link. This introduces
transmission delay at each hop.
End-to-end delay (N routers, 1 packet) = N × (L/R)

Statistical Multiplexing:
Link capacity is shared on demand. Packets from multiple users are interleaved. More efficient than
FDM/TDM but can cause queuing congestion.

Parameter Circuit Switching Packet Switching

Setup Required (connection establishment) Not required

Resource allocation Reserved end-to-end On-demand (shared)

Bandwidth Fixed, guaranteed Variable (best-effort)

Delay Constant (predictable) Variable (depends on traffic)

Efficiency Low (idle slots wasted) High (statistical multiplexing)

Example PSTN (telephone network) Internet

© Computer Networks Study Notes Page 14

UNIT-I | Computer Networks & the Internet Exam Study Guide

Cost Higher Lower

Data suitability Continuous streams (voice/video) Bursty data (web, email)

Congestion Not possible once circuit set up Possible – packets

queued/dropped

© Computer Networks Study Notes Page 15

UNIT-I | Computer Networks & the Internet Exam Study Guide

9. Standards, ISPs & Their Hierarchy

■ Protocols & Standards

Standards ensure interoperability by defining common rules that all vendors and manufacturers follow.
Without standards, devices from different companies could not communicate.

Two Types of Standards:

• De Facto Standards – Standards that emerge from widespread use without formal approval.
Example: TCP/IP became a de facto standard before formal adoption.
• De Jure Standards – Standards approved and ratified by official standards bodies. Example: ISO
7498 (OSI model), IEEE 802.3 (Ethernet).

■■ Standards Bodies
Organis Full Name Focus Area Examples
ation

ISO International Organisation OSI model, network ISO 7498 (OSI), ISO 9660
for Standardisation standards (CD-ROM)

IETF Internet Engineering Task Internet protocols RFCs: HTTP (RFC 2616), TCP
Force (TCP/IP) (RFC 793)

IEEE Institute of Electrical & LAN/WLAN standards 802.3 (Ethernet), 802.11 (Wi-Fi),
Electronics Engineers 802.15 (Bluetooth)

ITU International Telecom standards V.92 (modem), H.264 (video)

Telecommunication Union

ANSI American National US national standards ANSI C, FDDI

Standards Institute

W3C World Wide Web Web standards HTML, CSS, XML, HTTP
Consortium

ICANN Internet Corporation for Domain names, IP DNS root, IP allocation

Assigned Names & addresses
Numbers

■ EXAM TIP – Standards Bodies

• ISO → OSI Model (7 layers theory)

• IETF → TCP/IP protocols (RFCs – Request for Comments)
• IEEE → Ethernet (802.3) and Wi-Fi (802.11)
• Remember: IETF publishes RFCs, not ISO. TCP/IP not an ISO standard!

■ ISPs & Their Hierarchy

© Computer Networks Study Notes Page 16

UNIT-I | Computer Networks & the Internet Exam Study Guide

ISP Hierarchy

Tier-1 ISP (AT&T, Sprint, NTT)

Tier-2 ISP A Tier-2 ISP B

Local ISP Local ISP Local ISP Local ISP

U U U U U U U U

U = End User / Home / Enterprise

Tier Name Role Examples

Tier 1 National/Global ISP Own global backbone, peer with AT&T;, Verizon, Sprint,
each other for free (settlement-free NTT, Deutsche Telekom
peering)

Tier 2 Regional ISP Connect to Tier-1 (pay for transit), Regional telcos, large
peer with other Tier-2 for free cable companies

Tier 3 Local ISP / Access Last mile connection to Small local ISPs,
ISP homes/businesses; pay Tier-2 for campus networks
transit

IXP Internet Exchange Physical location where ISPs DE-CIX (Frankfurt),

Point exchange traffic directly AMS-IX (Amsterdam),
LINX (London)

CDN Content Delivery Cache content close to users; may Akamai, Cloudflare,
Network bypass ISP hierarchy Netflix Open Connect

■ EXAM TIP – ISP Hierarchy

• Tier 1 → Tier 2 → Tier 3 → End users (money flows upward for transit)
• Tier-1 ISPs peer for FREE (settlement-free); Tier-2 pay Tier-1
• IXPs allow direct peering → reduces latency and transit cost
• The Internet is a 'network of networks' – no single owner

© Computer Networks Study Notes Page 17

UNIT-I | Computer Networks & the Internet Exam Study Guide

10. Quick Revision – Exam Cheat Sheet

Q: What are the 3 key elements of a protocol?

A: Syntax (structure/format), Semantics (meaning/action), Timing (speed/sequence)

Q: How many layers does OSI have? Name them.

A: 7 – Physical, Data Link, Network, Transport, Session, Presentation, Application

Q: How many layers does TCP/IP have? Name them.

A: 4 – Network Access, Internet, Transport, Application

Q: What is store-and-forward?

A: Router must receive entire packet before forwarding; delay = L/R per link

Q: What is the bottleneck link?

A: Link with minimum bandwidth on the path; determines end-to-end throughput

Q: Difference between transmission delay and propagation delay?

A: Trans = L/R (push bits out); Prop = d/s (signal travel time)

Q: What is statistical multiplexing?

A: Packet switching sharing of link capacity based on demand, not fixed allocation

Q: FDM vs TDM?

A: FDM = different frequency bands; TDM = different time slots in a frame

Q: What organisation manages OSI standard?

A: ISO (International Organisation for Standardisation)

Q: What organisation manages TCP/IP?

A: IETF (Internet Engineering Task Force) via RFCs

Q: What is an IXP?

A: Internet Exchange Point – where ISPs directly connect and exchange traffic

Q: Guided vs Unguided media?

A: Guided = physical cables (UTP, coax, fibre); Unguided = wireless (radio, microwave)

Q: Which fibre mode is used for long-distance?

A: Single-mode fibre (one light path, very long distances, used in backbone)

Q: What is traffic intensity?

A: La/R – if ≥1, infinite queue builds; keep well below 1 for acceptable performance

© Computer Networks Study Notes Page 18

UNIT-I | Computer Networks & the Internet Exam Study Guide

Q: Circuit switching vs Packet switching – internet uses which?

A: Internet uses Packet Switching (more efficient for bursty data)

■ Final Exam Strategy

• Know all 7 OSI layers + their PDU names + 1-2 protocols each
• Know all 4 TCP/IP layers and how they map to OSI
• Memorise delay formulas: d_trans=L/R, d_prop=d/s, end-to-end=N×L/R
• Be ready to draw OSI & TCP/IP diagrams side by side in exam
• Understand ISP hierarchy: Tier-1→2→3→users; money flows upward
• Know difference between de facto (TCP/IP) and de jure (OSI) standards
• Guided: Twisted pair (UTP/STP), Coaxial, Fibre Optic
• Unguided: Radio waves, Microwaves, Infrared

© Computer Networks Study Notes Page 19