CSS Prepare

optional · CSS study notes

LAN Technologies

🧭

On this page — click any heading below to jump straight to it. ⬅️ Back to P-II.II. Computer Communications & Networks


OSI Layer: Physical (L1) & Data Link (L2)

Definition

LAN technologies are the standardized combinations of physical media, framing, and medium-access rules that implement a local area network. The dominant wired standard is Ethernet; the dominant wireless standard is Wi-Fi (IEEE 802.11); Token Ring is a largely historical alternative.

Ethernet (IEEE 802.3)

  • Originally a shared-bus broadcast medium using CSMA/CD for access control; modern switched Ethernet gives each device its own dedicated collision domain via a switch, making CSMA/CD largely vestigial on full-duplex links
  • Uses a 48-bit MAC address (physical address) burned into the NIC, written as 6 hex-byte groups (e.g., 00:1A:2B:3C:4D:5E)
  • Frame format: Preamble → Destination MAC → Source MAC → Type/Length → Data (46–1500 bytes) → CRC (Frame Check Sequence)
  • Speed evolution: 10 Mbps (original) → Fast Ethernet 100 Mbps → Gigabit Ethernet 1 Gbps → 10G/40G/100G Ethernet for modern backbones
  • Cabling evolved from shared coaxial (10BASE2/5) to twisted-pair with a central switch (10/100/1000BASE-T) to fibre for backbone links

Wi-Fi (IEEE 802.11)

  • Wireless LAN standard using radio frequencies (2.4 GHz and 5 GHz commonly, increasingly 6 GHz with Wi-Fi 6E)
  • Uses CSMA/CA (not CSMA/CD) for medium access, due to the wireless medium's inability to reliably detect collisions while transmitting, plus the hidden terminal problem
  • Operates in infrastructure mode (devices connect via a central Access Point) or ad-hoc mode (devices connect directly, peer-to-peer)
  • Standards have evolved through 802.11b/g/n/ac/ax (marketed as Wi-Fi 4/5/6), each raising maximum throughput and spectral efficiency

Token Ring (IEEE 802.5)

  • A largely historical LAN technology (once an IBM-backed alternative to Ethernet) where stations are logically (and often physically) arranged in a ring, and a special token frame circulates continuously
  • A station may only transmit once it captures the free token, eliminating collisions entirely by design (deterministic access) — see Token Passing under Multiple Access Protocols
  • Displaced almost entirely by switched Ethernet due to Ethernet's lower cost and simpler equipment, despite Token Ring's more predictable worst-case access delay

Comparison Table

BasisEthernetWi-Fi (802.11)Token Ring
MediumWired (twisted pair/fibre)Wireless (radio)Wired (originally ring-wired)
Access methodCSMA/CD (historical) / switchedCSMA/CAToken passing
TopologyStar (switched) / historical busStar (via AP)Logical ring
Status todayDominant wired LAN standardDominant wireless LAN standardObsolete

CSS Exam Angle

  • A likely question is a straight compare/contrast of Ethernet vs Wi-Fi vs Token Ring on medium, access method, and topology — the table above is exam-ready as is
  • Connect this topic explicitly to Multiple Access Protocols (CSMA/CD ↔ Ethernet, CSMA/CA ↔ Wi-Fi, Token Passing ↔ Token Ring) — examiners reward candidates who show the two topics aren't independent
  • Mention the MAC address format (48-bit, vendor-assigned via OUI + device-specific portion) as a concrete Data Link addressing fact

Example

A university lecture hall wired with Gigabit Ethernet to fixed desktop PCs, alongside a Wi-Fi access point serving the same room for laptops and phones, is a direct real-world pairing of the two dominant LAN technologies — the wired PCs get a dedicated, collision-free full-duplex link per switch port, while every wireless device shares one radio channel via CSMA/CA.

Likely Questions

  1. Compare Ethernet and Wi-Fi (802.11) as LAN technologies on medium, access method, and topology.
  2. Explain the structure of a MAC address and how it differs from an IP address.
  3. Why did Token Ring lose out to switched Ethernet despite its collision-free design?
  4. Explain how a shared-bus Ethernet segment's access method changes once it's replaced by a switch.

📜 Past-Paper Questions & Structured Answers

Questions are paraphrased; answer focus is suggested guidance, not an official marking scheme.

2017 · Paper-II · Q5(b) · 6 marks

Question (paraphrased): Explain Ethernet operation and its types.

Answer focus: Functions + frame/media access + representative Ethernet generations.

Ethernet (IEEE 802.3) provides local communication using frames and normally 48-bit MAC addresses. A sender places source/destination addresses, an EtherType/length field and data in a frame; the FCS detects corruption. A switch learns source MAC addresses and forwards known-unicast frames toward the destination port.

Media access: legacy shared half-duplex Ethernet uses CSMA/CD: sense, transmit, detect collision and retry after random backoff. Modern full-duplex switched Ethernet has no collisions and does not use CSMA/CD.

GenerationTypical variantsRate / medium
Standard10BASE-T; older 10BASE2/510 Mbps; twisted pair or coaxial cable
Fast100BASE-TX, 100BASE-FX100 Mbps; copper or fibre
Gigabit1000BASE-T, SX, LX1 Gbps; copper or fibre
10 Gigabit10GBASE-T, SR, LR10 Gbps; copper or fibre depending on variant

Functions: framing, local addressing, medium access and error detection. Ethernet does not itself provide end-to-end reliable delivery; damaged frames are generally discarded.

2022 · Paper-II · Q5(b) · 6 marks

Question (paraphrased): Design a ≥100 Mbps campus network for four buildings at the corners of a 100 m square, with existing 100BASE-T floors.

Answer focus: Explain distance constraint, draw topology, specify equipment and capacity.

Constraint: a typical 100BASE-TX copper channel is limited to 100 m. Even an adjacent-building route needs allowance for indoor cable runs; a diagonal is √(100²+100²) = 141.4 m before such allowance.

Proposed design: retain copper access links within each building, each within 100 m. Connect building switches by fibre to a core switch located in building A. Choose fibre/transceivers rated for the actual cable route.

B building switch ── fibre ── A core switch ── fibre ── C building switch
                                 |
                               fibre
                                 |
                          D building switch
Each building: floor access switches → local building switch

Use full-duplex switches with at least 100 Mbps access ports and preferably Gigabit fibre uplinks so several floors can share the backbone. Fibre avoids the copper distance constraint and electrical grounding issues between buildings. A second core/uplink can provide resilience, with loop prevention configured. VLANs can separate departments; inter-VLAN routing requires a router or layer-3 switch.

Result: each access link can provide 100 Mbps; simultaneous end-to-end throughput still depends on backbone capacity and traffic load.