Chapter 2: Communications

1. Network Types & Topologies

Key Concepts:

LAN (Local Area Network):
- Small geographic area (e.g., school, home).
- High-speed connections (Ethernet/Wi-Fi).
- Example: Star topology in a classroom (all devices connect to a central switch).
WAN (Wide Area Network):
- Large geographic area (e.g., internet, multi-site businesses).
- Uses leased lines/satellites.
- Example: University campuses linked across cities.

Topologies:

Topology	Description	Advantages	Disadvantages
Star	All devices connect to a central switch/router.	Easy to add devices; fault isolation.	Single point of failure (central device).
Bus	Single backbone cable connects all devices.	Cheap to set up.	Collisions (CSMA/CD needed); hard to troubleshoot.
Mesh	Each device connects to every other device.	Redundant paths; high reliability.	Expensive; complex cabling.

Past Paper Example (2023 Q8a):
Question: Is a school network a LAN or WAN? Justify.
Marking Scheme Answer:

LAN – confined to one building; uses Ethernet/Wi-Fi.

2. Hardware Components

Key Devices:

Switch: Connects devices in a LAN; forwards packets to correct MAC address.
Router: Connects LANs/WANs; routes packets using IP addresses.
Wireless Access Point (WAP): Allows Wi-Fi connections.
Network Interface Card (NIC): Enables wired/wireless connectivity.

Past Paper Example (2024 Q2d):
Question: Identify 3 LAN hardware components.
Marking Scheme Answer:

Switch
Router
Ethernet cables

3. Transmission Media

Cable Type	Description	Advantages	Disadvantages
Copper (Twisted Pair)	Cheap; flexible.	Easy to install.	Low bandwidth; EMI interference.
Fibre-Optic	Glass strands; light signals.	High bandwidth; no EMI.	Expensive; fragile.
Wireless (Wi-Fi)	Radio waves.	Mobility; no cables.	Signal interference; slower than wired.

Past Paper Example (2023 Q9b):
Question: Compare fibre-optic and copper cables.
Marking Scheme Answer:

Fibre-optic has higher bandwidth.
Fibre-optic is immune to EMI.
Copper is cheaper but shorter range.

4. Protocols & Collision Management

CSMA/CD (Carrier Sense Multiple Access/Collision Detection)

1. Definition (1 mark):
CSMA/CD is a protocol used in Ethernet networks (typically bus topology) to manage data transmission and resolve collisions when multiple devices attempt to transmit simultaneously.

2. Step 1: Carrier Sense (1 mark):
Before transmitting, a device checks the network cable to determine if it is free (no ongoing transmissions).

3. Step 2: Transmission/Collision Detection (1 mark):

If the cable is free, the device begins transmitting.
If two devices transmit simultaneously, a collision occurs, and the data packets are corrupted.

4. Step 3: Collision Response (1 mark):

Devices detect the collision via voltage spikes.
Both devices stop transmitting and send a jamming signal to alert other devices.

5. Step 4: Random Backoff (1 mark):

Each device waits a random time interval (backoff algorithm) before retrying transmission to avoid repeated collisions.

Example (2023 Q12):
Question: Describe how CSMA/CD works.
Answer:

Devices check if the cable is free before transmitting.
If two devices transmit at once, a collision occurs.
Collisions are detected via voltage spikes.
Devices send a jamming signal and stop transmitting.
Each device waits a random time before retrying.

5. IP Addressing & Subnetting

Private IP: Used within LAN (e.g., 192.168.x.x).
Public IP: Assigned by ISP; visible on internet.
Subnetting: Divides network to reduce traffic.
- Subnet Mask: Identifies network/host parts (e.g., 255.255.255.0).

Finding the Network ID

Write the IP and subnet mask in binary.
- Example:
  - IP: 192.168.1.10 → 11000000.10101000.00000001.00001010
  - Subnet: 255.255.255.0 → 11111111.11111111.11111111.00000000
Perform a bitwise AND operation:
- Compare each bit (1 AND 1 = 1; all others = 0).
Result = Network ID:
- 11000000.10101000.00000001.00000000 → 192.168.1.0

Shortcut:

For /24 subnet (mask 255.255.255.0), the first 3 octets are the Network ID.

Example:

IP: 10.50.20.30 with mask 255.0.0.0 → Network ID = 10.0.0.0.

Key Points

Subnet mask 1s = Network portion.
AND operation isolates the Network ID.
Host ID = Remaining bits after Network ID.

This method works for IPv4 addresses with any subnet mask.

Past Paper Example (2023 Q15c):
Question: Find network ID for 10.10.12.1 with mask 255.0.0.0.
Marking Scheme Answer:

Network ID = 10.0.0.0 (first octet).

6. Cloud Computing

Types:

Public Cloud: Shared resources (e.g., AWS, Google Cloud).
Private Cloud: Dedicated to one organization.

Advantages:

Scalability.
Remote access.

Disadvantages:

Dependency on internet.
Security risks.

Past Paper Example (2024 Q11c):
Question: Give 2 benefits of cloud computing.
Marking Scheme Answer:

Cost-effective (no hardware maintenance).
Access files anywhere.

7. Client-Server vs. Peer-to-Peer (P2P)

Feature	Client-Server Model	Peer-to-Peer (P2P) Model
Centralization	Centralized (dedicated server).	Decentralized (all devices are equal).
Scalability	✅ High (server handles load).	❌ Limited (depends on peers’ resources).
Security	✅ High (centralized control, backups).	❌ Low (no central authority, vulnerable).
Cost	❌ Expensive (server hardware/maintenance).	✅ Cheap (no dedicated server needed).
Performance	✅ Stable (server manages resources).	❌ Unpredictable (depends on peers).
Reliability	✅ High (single point of failure is server).	❌ Low (peers may disconnect).
Setup Complexity	❌ Complex (requires server setup).	✅ Simple (no central setup needed).
Use Cases	Websites, databases, enterprise systems.	File sharing (Torrent), blockchain.

Key Takeaways

Choose Client-Server for reliability, security, and scalability (e.g., banking systems).
Choose P2P for cost savings and decentralization (e.g., BitTorrent, Bitcoin).

Past Paper Example (2023 Q4a):
Question: Describe P2P drawbacks.
Marking Scheme Answer:

No centralized security.
Slower if peers disconnect.

Exam Tips

Topologies: Always link to context (e.g., “Star is best for schools due to fault isolation”).
CSMA/CD: Mention “random wait time” for full marks.
Subnetting: Use binary AND operation to find network ID.

Worked Example (2024 Q13b)

Question: Why use wired over wireless for gaming?
Marking Scheme Answer:

Lower latency (no interference).
Higher bandwidth (stable connection).