chapter 1

1.1 Data Communications

Definition & Process

• Data Communications is simply sharing information (data) between two devices over some physical connection (like a cable).

• Communication can be local (face-to-face) or remote (over distances).

Four Key Characteristics

1. Delivery: Data must reach the correct destination (only the intended device/user).

2. Accuracy: Data must be sent without errors; any change during transmission can make it unusable.

3. Timeliness: Data needs to arrive on time; delays can render it useless (important for things like video or audio).

4. Jitter: Variation in packet arrival times can cause uneven quality (e.g., in video streams).

Main Components of a Data Communications System

1. Message: The actual data (text, numbers, images, audio, video).

2. Sender: The device that sends the data (e.g., computer, phone).

3. Receiver: The device that gets the data.

4. Transmission Medium: The physical path (like wires or radio waves) that data travels on.

5. Protocol: A set of rules that allows the sender and receiver to understand each other (like speaking the same language).

Representation of Data

• Text: Encoded as bits using systems like Unicode.

• Numbers: Represented directly in binary form.

• Images: Made up of pixels; more pixels mean higher resolution but need more memory. Colors can be represented using systems like RGB or YCM.

• Audio & Video: Audio is a continuous signal (can later be digitized), while video is either a continuous recording or a series of images that create the illusion of motion.

Data Flow Types

• Simplex: One-way communication (e.g., keyboard → monitor).

• Half-Duplex: Both devices can send and receive but not at the same time (e.g., walkie-talkies).

• Full-Duplex: Both devices communicate simultaneously (e.g., telephone calls).

1.2 Networks

What is a Network?

• A network connects devices (like computers, phones, routers) so they can communicate.

Key Criteria for a Network

1. Performance: Measured by how fast data travels (transit time) and how quickly a device responds.

2. Reliability: How often the network fails and how quickly it recovers.

3. Security: Keeping data safe from unauthorized access and damage.

Physical Structure and Connection Types

• Point-to-Point Connection: A dedicated link between two devices.

• Multipoint Connection: One link shared by more than two devices (can be time-shared or space-shared).

Common Network Topologies (Physical Layouts)

1. Mesh Topology: Every device is directly connected to every other device.

2. Star Topology: All devices connect to a central hub; they don’t connect directly with each other.

3. Bus Topology: All devices share a single cable (backbone).

4. Ring Topology: Devices are connected in a circle, and data travels in one direction around the ring.

1.3 Network Types

Local Area Network (LAN)

• Covers a small area (office, building, campus).

• Devices within a LAN have unique identifiers (addresses) for sending and receiving data.

Wide Area Network (WAN)

• Covers larger areas (cities, countries, or even the world).

• Can be a simple direct connection (point-to-point WAN) or more complex (switched WAN) connecting multiple networks.

Internetwork & The Internet

• Internetwork: When two or more networks are connected, they form an internetwork (or “internet” with a lowercase “i”).

• The Internet: A huge network made of many internetworks, including backbone networks (large high-speed networks), provider networks (regional/national ISPs), and customer networks (individual users).

Accessing the Internet

• Telephone Networks:

  • Dial-up: Uses a modem over the telephone line; slow and ties up the line.

  • DSL: Faster than dial-up and allows simultaneous voice/data use.

• Cable Networks: Use upgraded cable TV networks for faster internet (speed may vary with usage).

• Wireless Networks: Increasingly popular for connecting homes and businesses.

• Direct Connection: Large organizations can connect directly by leasing high-speed WAN links.

1.4 Protocol Layering

Why Layer Protocols?

• Protocols are sets of rules for communication.

• For complex communication, the task is split into layers (like building blocks). Each layer handles a part of the process, making systems modular and easier to manage or update.

Examples of Layering

• Single-Layer Communication: Two people talking face-to-face follow basic rules (greetings, taking turns).

• Multi-Layer Communication (e.g., sending an encrypted letter):

  • One layer creates the plain message (plaintext).

  • A second layer encrypts the message (producing ciphertext).

  • A third layer handles packaging (putting it in an envelope with addresses).

• If only one layer has a problem (like encryption), you can replace just that layer without changing the others.

Principles of Protocol Layering

1. Bidirectional Functions: Every layer must be able to both send and receive.

2. Consistency Across Layers: The “object” (like a letter, ciphertext, or packet) must be the same on both ends for each layer.

Logical Connections

• Each layer is logically connected to the corresponding layer on the other side. This makes it easier to manage data flow from one device to another.

1.5 TCP/IP Protocol Suite

Overview

• The TCP/IP protocol suite is a collection of protocols used on the Internet. It is organized into layers, each with specific functions.

Layered Architecture Example

• In a small internetwork with LANs, routers, and switches:

  • End Systems (hosts): Use all five layers of the TCP/IP model.

  • Routers: Typically work on the network layer (and adapt for the link layer depending on the connection).

  • Switches: Work on the data-link and physical layers, managing local connections.

Key Points About Layers

• Application, Transport, and Network Layers: Handle end-to-end communication (from one host to another).

• Data-Link and Physical Layers: Manage communication between immediate hops (local devices or routers).

Each layer transforms the data in a specific way and passes it to the next layer until it reaches the destination, where the process is reversed.

1.5.3 Description of Each TCP/IP Layer

Physical Layer

• Role: Carries individual bits (0s and 1s) across the physical medium (wires, fiber, wireless, etc.).

• Note: Although it works at the bit level, it still interacts with the transmission medium beneath it.

Data-Link Layer

• Role: Transfers frames (data units) across a single link (for example, within a LAN or WAN segment).

• Note: When a router sends data onto the next link, the data-link layer handles that step.

Network Layer

• Role: Establishes host-to-host connections across multiple networks.

• Function: Routers select the best path for each packet as it moves from the source to the destination.

Transport Layer

• Role: Provides end-to-end communication services between processes running on hosts.

• Function: Encapsulates application data into segments (or datagrams) and ensures that a complete message is delivered from one process to another.

Application Layer

• Role: Facilitates process-to-process communication between application programs.

• Function: Handles requests and responses so that applications (like web browsers or email programs) can exchange information.

1.6 The OSI Model

Overview

• The OSI (Open Systems Interconnection) model is a framework defined by the International Organization for Standardization (ISO) to help design and understand network systems.

• Purpose: To allow different computer systems to communicate without changing their underlying hardware or software.

OSI Model Layers

• The OSI model has 7 layers:

  1. Physical

  2. Data-Link

  3. Network

  4. Transport

  5. Session

  6. Presentation

  7. Application

OSI vs. TCP/IP

• Key Difference: The TCP/IP suite has 5 layers because it combines or omits some of the OSI layers:

  1. The Session and Presentation layers of the OSI model are not separate in TCP/IP; their functions are handled within the Transport and Application layers.

• Reasons for the Difference:

  1. TCP/IP supports multiple transport-layer protocols, some of which already provide session-like services.

  2. The application layer in TCP/IP is flexible—developers can add functionalities as needed without being restricted by a fixed layer.

Why OSI Didn't Replace TCP/IP

• TCP/IP was already well-established and widely used.

• OSI's additional layers (session and presentation) were never fully defined or implemented.

• Performance issues in OSI implementations made it less attractive compared to the proven TCP/IP suite.