Unit 1: Network basics

I. Orientation

A computer network is an interconnected system in which devices exchange data through communication links according to agreed rules. Networking combines hardware, software, transmission media, addressing, and protocols so that information can move accurately and efficiently between endpoints.

• Governing principle: Communication succeeds when a sender and receiver use compatible representations, interfaces, addresses, and protocols.
• Basic entities: Nodes include computers, servers, routers, switches, sensors, and mobile devices; links may be wired or wireless.
• Data flow: Information travels as electromagnetic signals carrying bits across a transmission medium.
• Performance measures:
  • Bandwidth: The theoretical capacity of a link, commonly measured in bits per second (bit/s).
  • Throughput: The rate of successful data delivery in practice.
  • Latency: The time taken for data to travel from source to destination.
• Service requirements: Networks are designed around performance, reliability, security, scalability, and cost.
• Standard convention: Layered models divide communication into manageable functions and define interfaces between them.

II. Foundations of Data Communication — Effective Exchange of Information

A. Data communication characteristics

Data communication is the exchange of data between devices through a transmission medium, and its effectiveness depends on four principal characteristics.

• Delivery: Data must reach the correct destination, identified through an address such as an IP address or port number.
• Accuracy: Data must arrive without unauthorized alteration or transmission errors; mechanisms such as checksums and cyclic redundancy checks detect corruption.
• Timeliness: Data must arrive within an acceptable period. Voice calls and online games are particularly sensitive to delay.
• Jitter: Variation in packet arrival time must remain controlled. For example, packets arriving after 20 ms, 25 ms, and 80 ms intervals can produce uneven audio.
• Security: Confidentiality, integrity, and availability protect data from disclosure, modification, and disruption.
• Performance relationship: Transmission time depends on message size and data rate.

Transmission time = Message size / Data rate

Here, transmission time is measured in seconds, message size in bits, and data rate in bit/s. A 1,000,000-bit message sent at 10 Mbit/s requires 0.1 seconds to enter the link.

B. Components

A data communication system requires five components that work together to convey meaningful information.

• Message: The information being communicated, such as text, numbers, images, audio, or video.
• Sender: The source device that creates and transmits the message, such as a laptop or sensor.
• Receiver: The destination device that accepts and interprets the message, such as a server.
• Transmission medium: The physical or wireless path carrying signals.
  • Guided media include twisted-pair cable, coaxial cable, and optical fibre.
  • Unguided media include radio, microwave, infrared, and satellite links.
• Protocol: The set of rules controlling communication, including message format, order, timing, and error handling.
• Concrete exchange: When a browser requests a webpage, the computer is the sender, the HTTP request is the message, Wi-Fi and fibre form the medium, the web server is the receiver, and TCP/IP supplies communication rules.

III. Data Form and Direction — Encoding and Flow

A. Data representation

All information is represented internally as binary digits, although the encoding method depends on the data type.

• Text: Characters are mapped to binary codes. ASCII represents basic English characters, while Unicode supports scripts and symbols worldwide; for example, ASCII capital A has decimal value 65.
• Numbers: Numeric values are stored directly in binary rather than as character shapes; decimal 13 is binary 1101.
• Images: A raster image is a grid of pixels, with each pixel represented by a colour depth. A 24-bit RGB pixel typically allocates 8 bits each to red, green, and blue.
• Audio: Sound is digitized by sampling amplitude at regular intervals and quantizing each sample. CD-quality audio commonly uses 44,100 samples per second and 16 bits per sample per channel.
• Video: Video is a sequence of image frames, usually accompanied by audio. Compression standards reduce the large raw bit rate by removing spatial and temporal redundancy.
• Signal conversion: A network interface converts stored bits into electrical, optical, or radio signals suitable for the selected medium.

B. Transmission modes

Transmission modes describe the permitted direction of data flow between two connected devices.

1. Simplex transmission: Communication occurs in one direction only.

   • Operation: One device always sends and the other always receives.
   • Example: A traditional television broadcast sends content to viewers without a return channel.

2. Half-duplex transmission: Both devices can transmit, but not simultaneously.

   • Operation: The shared channel changes direction as devices take turns.
   • Example: Walkie-talkies require one user to release the channel before another speaks.

3. Full-duplex transmission: Both devices transmit simultaneously.

   • Operation: Separate paths or channel-division techniques support concurrent flows.
   • Example: A telephone call allows both participants to speak and hear at the same time.

• Explicit contrast: Simplex maximizes simplicity, half-duplex shares capacity across alternating directions, and full-duplex provides the most natural interactive communication.

IV. Network Organization — Design and Scope

A. Network design

Network design determines how devices, links, services, and controls are arranged to meet operational requirements.

• Physical topology: The actual arrangement of cables, access points, switches, and devices.
  • A star connects devices to a central switch.
  • A bus uses a shared backbone.
  • A ring connects each node to two neighbours.
  • A mesh supplies multiple paths between nodes.
• Logical topology: The path followed by data, which may differ from the physical layout because of switching and virtual networks.
• Client-server design: Dedicated servers provide centralized services such as authentication, files, or websites to client devices.
• Peer-to-peer design: Devices communicate as equals and may both request and provide resources, suitable for small or distributed systems.
• Design criteria:
  • Scalability accommodates additional users and traffic.
  • Redundancy provides alternative devices or paths after failure.
  • Security uses segmentation, firewalls, encryption, and access control.
  • Quality of service prioritizes delay-sensitive voice or video traffic.
• Trade-off: A mesh offers strong fault tolerance but requires more links; a star is economical but depends heavily on its central device.

B. Network classifications

Networks are classified primarily by geographic coverage, ownership, function, and transmission technology.

• Personal Area Network (PAN): Covers a few metres around a person; Bluetooth links between a phone and earbuds are typical.
• Local Area Network (LAN): Connects devices within a home, office, laboratory, or building, usually through Ethernet or Wi-Fi.
• Campus Area Network (CAN): Interconnects several LANs across an institution such as a university campus.
• Metropolitan Area Network (MAN): Extends across a city or large urban region.
• Wide Area Network (WAN): Spans countries or continents using carrier infrastructure; the Internet interconnects many networks globally.
• Specialized classifications:
  • A Wireless LAN (WLAN) uses radio-based Wi-Fi rather than Ethernet cabling.
  • A Storage Area Network (SAN) provides high-speed access between servers and storage systems.
  • A Virtual Private Network (VPN) creates an encrypted logical connection over a public network.
• Ownership contrast: A private network is controlled by one organization, whereas a public network provides services to many subscribers.

V. Communication Rules and Structure — Modular Networking

A. Protocols

A protocol is a formal set of rules governing communication between corresponding entities in a network.

• Syntax: Defines message structure, field order, size, and encoding; an IPv4 header, for example, contains source and destination address fields.
• Semantics: Defines the meaning of fields and the actions taken when particular values are received.
• Timing: Specifies when data may be sent, expected transmission rate, sequencing, and timeout behaviour.
• Core functions: Protocols may perform addressing, routing, segmentation, flow control, error control, authentication, and connection management.
• Examples:
  • HTTP transfers web resources.
  • TCP provides reliable, ordered byte delivery.
  • IP forwards packets across interconnected networks.
  • Ethernet governs local frame delivery.
• Standardization: Open standards permit equipment and software from different manufacturers to interoperate.

B. Layered architecture

Layered architecture divides networking into ordered modules, with each layer providing services to the layer above it.

• Service relationship: A layer uses lower-layer services and hides their implementation details from higher layers.
• Peer communication: Corresponding layers on different devices communicate logically through a shared protocol.
• Encapsulation: Each sending layer adds control information, normally as a header; the receiver removes it through decapsulation.

Application data → Segment → Packet → Frame → Bits

• Advantages: Modularity simplifies design, testing, maintenance, standardization, and replacement of technologies.
• Limitations: Layer boundaries may duplicate functions, add headers, and reduce efficiency.
• Concrete example: A web application can use HTTP over TCP regardless of whether the underlying link is Ethernet, Wi-Fi, or fibre.

VI. Reference Models — OSI and Internet Protocol Architecture

A. OSI model

The ISO Open Systems Interconnection model organizes communication into seven conceptual layers, numbered from Physical at Layer 1 to Application at Layer 7.

• Layer 7—Application: Supplies network services to user applications, including web, email, and file-transfer functions.
• Layer 6—Presentation: Handles data translation, compression, and encryption.
• Layer 5—Session: Establishes, manages, synchronizes, and terminates communication sessions.
• Layer 4—Transport: Provides end-to-end delivery, segmentation, reliability, and flow control.
• Layer 3—Network: Performs logical addressing and routing between networks.
• Layer 2—Data Link: Provides framing, physical addressing, media access, and local error detection.
• Layer 1—Physical: Transmits raw bits through signals, connectors, frequencies, and media.
• Significance: The model is primarily a reference framework for describing responsibilities and troubleshooting faults layer by layer.

B. TCP/IP protocol suite

The TCP/IP suite is the practical architecture of the Internet and is commonly represented using four layers.

• Application layer: Combines OSI application, presentation, and session functions; protocols include HTTP, DNS, SMTP, and DHCP.
• Transport layer: Supports process-to-process delivery.
  • TCP is connection-oriented, reliable, ordered, and congestion-controlled.
  • UDP is connectionless and has lower overhead but does not guarantee delivery.
• Internet layer: Uses IP for logical addressing and routing; ICMP reports network conditions and errors.
• Network access layer: Handles local framing, media access, and physical transmission through technologies such as Ethernet and Wi-Fi.
• Model correspondence: TCP/IP’s Internet layer broadly matches the OSI Network layer, while its network access layer combines OSI Data Link and Physical functions.
• Practical path: An HTTP message may be encapsulated inside a TCP segment, an IP packet, and an Ethernet frame before transmission.

VII. Layer Identification — Delivery at Different Scopes

A. Addressing mechanisms in layers

Each layer uses an addressing mechanism suited to a particular delivery scope, from local hardware to application processes.

• Data-link address: A Media Access Control address identifies an interface for delivery within a local network; Ethernet commonly uses 48-bit MAC addresses such as 00:1A:2B:3C:4D:5E.
• Network-layer address: An IP address identifies a host interface and its network location for routing.
  • IPv4 addresses are 32 bits, as in 192.0.2.10.
  • IPv6 addresses are 128 bits, as in 2001:db8::10.
• Transport-layer address: A 16-bit port number identifies an application process. HTTPS conventionally uses destination port 443.
• Application-layer identifier: Human-oriented names and resource identifiers include domain names, email addresses, and URLs; DNS can translate a domain name into an IP address.
• Address resolution: On an IPv4 LAN, Address Resolution Protocol maps a known IPv4 address to a MAC address; IPv6 uses Neighbor Discovery.
• Scope contrast: MAC addresses support next-hop local delivery, IP addresses support end-to-end routing across networks, and port numbers direct received data to the correct process.
• Combined endpoint: A socket is commonly identified by an IP address and port, such as 192.0.2.10:443, enabling multiple applications to share one host.