Asd
Asd
A Local Area Network (LAN) is a system that connects mul ple devices to communicate and share
resources efficiently. The design of a LAN follows a structured model with different layers that handle
data transmission and network access.
LANs follow the OSI model, which includes seven layers, but the most cri cal ones for LAN design are:
Physical Layer: Defines the hardware and transmission medium (cables, wireless signals, etc.).
Data Link Layer: Ensures reliable data transfer between devices and is divided into:
o Logical Link Control (LLC): Manages communica on, flow, and error control.
o Medium Access Control (MAC): Governs how mul ple devices share the network.
The MAC layer is separated because different networks use different methods to control access, such as
Ethernet (CSMA/CD) or Token Ring.
1. User data moves from the Applica on Layer to TCP/IP Layer and then to the LLC Layer.
2. The LLC Layer adds control informa on, forming an LLC Protocol Data Unit (PDU).
3. The MAC Layer adds more details to create a MAC Frame, which is transmi ed over the
network.
The IEEE 802 commi ee developed widely used LAN standards, later adopted by ISO (Interna onal
Organiza on for Standardiza on). These standards define different transmission mediums, such as:
Coaxial cables
Op cal fiber
Wireless LANs (Wi-Fi, DQDB, etc.) – Use different protocols for wireless communica on.
Conclusion
LAN design ensures efficient and reliable communica on by organizing data transmission through
structured layers. The IEEE 802 standards provide various networking op ons, allowing flexibility in how
data is transmi ed and accessed across different mediums.
LAN Topologies (Simple Explana on)
A LAN topology is the way computers and devices are connected in a network. There are four common
types: bus, tree, ring, and star.
1. Bus Topology
All devices are connected to a single main cable (called the bus).
Devices send data to everyone, but only the intended recipient reads it.
Example: Like a public announcement (PA) system where one person speaks, and everyone
listens.
Advantages:
Simple and cheap to set up
Uses less cable
Works well for small networks
Disadvantages:
If the main cable breaks, the en re network stops working
Only one device can send data at a me (or else signals will collide)
2. Tree Topology
Similar to a bus topology but with branches.
A central point (headend) connects mul ple cables that branch out.
Advantages:
Can cover a larger area
Easier to expand than a bus network
Disadvantages:
If the main cable fails, the en re network is affected
Requires more cable than a simple bus
3. Each device checks if the message is for them (by looking at the address).
To prevent collisions, devices take turns sending messages. This is like raising your hand in class before
speaking.
How It Works
The frame circulates through all devices un l it reaches the correct one.
The des na on device copies the data and lets it con nue moving.
When the frame returns to the sender, it is removed from the ring.
Each device has a repeater, which boosts the signal before passing it to the next device.
Since all devices share the ring, they must take turns sending data.
Single point of failure → If one device or link fails, the whole network stops.
Slower than Star Topology → Data must pass through mul ple devices.
Difficult to expand → Adding or removing a device disrupts the network.
Used in older network designs, especially in token-based networks like Token Ring.
Some fiber-op c networks use a dual-ring setup for redundancy.
Star Topology Summary
In a star topology, all devices connect to a central node (hub or switch) via individual links for
transmission and recep on.
1. Broadcast Mode: The central node forwards incoming data to all connected devices, making it
logically similar to a bus topology. Only one device can transmit at a me.
2. Switching Mode: The central node buffers and directs data only to the intended recipient,
improving efficiency and reducing collisions.
Medium Access Control (MAC) protocols manage how devices share the network’s transmission capacity.
There are two main approaches to control:
1. Centralized Control:
2. Decentralized Control:
Synchronous: Fixed, dedicated capacity (e.g., TDM, circuit switching), not ideal for LANs due to
unpredictable traffic.
Asynchronous: Dynamic alloca on of capacity based on demand, more efficient for LANs, and
further divided into:
In general, asynchronous and decentralized methods offer be er scalability and adaptability for modern
networks.
In Round Robin access control, each sta on takes turns to transmit data. Each sta on can either transmit
data during its turn or skip it, but once it's done (or skips), the next sta on in the sequence gets its turn.
The sequence can be controlled either by a central controller or by the sta ons themselves (distributed).
Efficient for many sta ons: If many sta ons have data to send, Round Robin works well because
it ensures each gets a fair chance to transmit.
Not efficient for few sta ons: If only a few sta ons have data, it becomes inefficient because
many sta ons will just pass their turn without transmi ng, causing unnecessary overhead.
Stream traffic (e.g., voice calls, file transfers) works well with Round Robin because it requires
con nuous transmission.
Bursty traffic (e.g., terminal-host interac ons) might not be efficient with Round Robin, as
transmissions are short and sporadic.
Reserva on
For stream traffic, reserva on techniques are well suited. In general, for these tech- niques, me on the
medium is divided into slots, much as with synchronous TDM. A sta on wishing to transmit reserves
future slots for an extended or even an in- definite period. Again, reserva ons may be made in a
centralized or distributed fashion.
Conten on techniques are used for bursty traffic, where devices need to send small, irregular bursts of
data. In this method, there is no control over which sta on gets to send data first. Instead, all devices try
to send data at the same me, leading to poten al conflicts or collisions, which can cause problems if
many devices try to transmit at once.
Advantages: It's simple and works efficiently when there are only a few devices trying to
transmit at the same me (light to moderate network traffic).
Disadvantages: Performance can become poor if many devices compete to send data
simultaneously, leading to delays or conges on.
The MAC (Medium Access Control) layer handles data transmission and medium access in networking,
using a MAC frame as the Protocol Data Unit (PDU). Here's a summarized breakdown:
o The MAC layer receives data from the LLC (Logical Link Control) layer and is responsible
for managing access to the network medium, as well as transmi ng the data.
o MAC Control: Contains protocol control informa on, such as priority levels.
o Des na on MAC Address: Specifies where the frame is to be delivered on the LAN.
o CRC (Cyclic Redundancy Check): Used to detect errors in the transmi ed frame. If errors
are found, the MAC layer discards the frame.
3. Error Handling:
o The MAC layer is responsible for detec ng errors using the CRC and discarding
erroneous frames.
o The LLC layer is op onally responsible for retransmi ng any frames that were not
successfully received.
In summary, the MAC layer formats data into a frame, manages access to the transmission medium,
handles error detec on, and works with the LLC layer for error recovery if necessary.
The Logical Link Control (LLC) layer is a part of the data link layer in LANs. It helps in transmi ng data
between two devices without needing an intermediate switch.
2. Works with the MAC layer, which handles physical network access.
LLC Services
LLC helps in addressing devices and managing data exchange. (hdlc)It provides three types of services:
o Used in cases where occasional data loss is acceptable (e.g., sensor readings, monitoring
systems).
o Used for important messages (e.g., emergency alarms) that must be confirmed quickly.
LLC Protocol
LLC is based on the HDLC (High-Level Data Link Control) protocol and follows a similar structure. It has
three types of opera ons:
2. Type 2: Supports connec on-mode service. Ensures error-free data delivery using
acknowledgments.
Uses different message types (PDUs) for sending, receiving, and managing connec ons.
In connec on-mode service (Type 2), a connec on is requested, accepted, used, and then
closed when done.
In acknowledged connec onless service (Type 3), each message is confirmed using small
sequence numbers (0 and 1) to ensure delivery.
Conclusion
LLC helps LAN devices communicate efficiently, providing op ons based on reliability and speed needs.
It ensures flexibility and efficiency while working with the MAC layer to manage network access.
IEEE 802
Physical:
• Encoding/decoding
• Bit transmission/recep on
ALOHA is a random access protocol used for wireless and satellite communica on. It allows mul ple
devices to share a common channel without coordina on.
Types of ALOHA
1. Pure ALOHA
2. Slo ed ALOHA
Applica ons
Limita ons
Low efficiency compared to modern MAC protocols like CSMA/CD (Ethernet) and CSMA/CA (Wi-
Fi).
1. Time is divided into fixed slots equal to the frame transmission me.
4. If a collision occurs, the sender waits for a random me and retries in the next slot.
CSMA is a network access method used in shared communica on channels where mul ple devices
compete to transmit data. The key idea is "listen before talk" to reduce collisions.
4. If a collision occurs (in some CSMA variants), the device takes correc ve ac on.
Non-Persistent CSMA
Non-Persistent CSMA is a collision avoidance technique where a device checks the channel before
transmi ng and waits a random me if the channel is busy, instead of con nuously sensing it.
Advantages
Overview
1-Persistent CSMA is a medium access control (MAC) protocol where a device con nuously senses the
communica on channel and transmits immediately when the channel becomes idle. It is called "1-
Persistent" because the device has a 100% (i.e., 1.0 probability) chance of transmi ng as soon as the
channel is free.
This protocol is aggressive and works well in low-traffic networks, but in high-traffic condi ons, it
increases the risk of collisions due to mul ple devices trying to send data at the same me.
o The device keeps sensing con nuously (i.e., it does not wait randomly).
4. If a collision occurs:
o Since mul ple devices may try to transmit at the same me (as they all sense the idle
channel), collisions can happen.
o The device then waits for a random backoff me before rea emp ng transmission.
Lower wai ng me: Since devices transmit immediately when the channel is idle, there is minimal
delay.
Efficient in low-traffic condi ons: When there are few devices, collisions are rare, and the protocol
works efficiently.
p-Persistent CSMA – Detailed Explana on
Overview
p-Persistent CSMA is a probabilis c channel access method used in slo ed me networks to reduce
collisions while maintaining efficiency. It is a hybrid of 1-Persistent and Non-Persistent CSMA, offering a
balance between collision probability and channel u liza on.
It is mainly used in slo ed- me systems like Wi-Fi (IEEE 802.11), where me is divided into slots, and
transmission decisions are based on probability.
o With probability (1 - p), the device waits for the next me slot and checks again.
3. If busy, the device waits un l the next me slot and repeats the process.
4. If a collision occurs, the device applies a random backoff me and retries a er some me.
Similar to 1-Persistent CSMA → It listens con nuously and transmits if the channel is idle.
Similar to Non-Persistent CSMA → It does not always transmit immediately, reducing collision
chances.
Unique Feature → Uses a probability factor (p) to decide when to transmit, balancing speed and
efficiency.
Advantages
Disadvantages
Best suited for slo ed- me networks, such as IEEE 802.11 (Wi-Fi).
np<1
Dynamic p adjustment is used in some networks (e.g., Wi-Fi) to op mize performance based on
network load.
CSMA/CD
CSMA/CD (Carrier Sense Mul ple Access with Collision Detec on) is used in wired Ethernet
(IEEE 802.3) to detect and manage collisions.
Steps:
3. Collision Detec on (CD) – If a collision occurs, stop and send a jamming signal.
4. Backoff & Retry – Wait for a random me (Binary Exponen al Backoff) before
retransmi ng.
A er the 10th collision, the wait me stops increasing and remains at a maximum limit.
A er the 16th collision, the device gives up and drops the packet (no retransmission).
Used in Ethernet (CSMA/CD) and Wi-Fi (CSMA/CA) to reduce repeated collisions and improve
network efficiency.
Ethernet
Topology: Bus.
Topology: Bus.
The 100BASE-T family includes mul ple standards based on different cabling types and signaling
methods, as shown in the diagram.
100BASE-T Categories:
100BASE-TX
o Requires two twisted-pair cables (one for sending, one for receiving).
100BASE-FX
Used in networks where only Category 3 cables are available, as it doesn't require all pairs to
support high-speed signaling.
Less common today since 100BASE-TX (with Cat 5) became the standard.
Conclusion:
100BASE-T4 was an alterna ve for older Cat 3 cabling but is now obsolete.
Half-duplex Ethernet: Can either transmit or receive, using CSMA/CD to handle collisions.
Full-duplex Ethernet: Allows simultaneous transmission and recep on, effec vely doubling the
data rate.
Uses the IEEE 802.3 MAC frame format, ensuring compa bility.
Fast Ethernet allows both old (10 Mbps) and new (100 Mbps) networks to work together.
Example Setup:
o Hubs link to switching hubs that support both 10 Mbps and 100 Mbps.
o Powerful computers and servers connect directly to fast switches (10/100 Mbps).
o 100 Mbps hubs act as the main network backbone and connect to a router for
internet/WAN access.
Benefit: Allows older and newer devices to work together smoothly while upgrading to faster
speeds.
Gigabit Ethernet (GbE) is a high-speed networking standard that provides 1 Gbps (1000 Mbps) data
transfer rates. It is widely used in enterprise networks, data centers, and home networking for fast and
reliable communica on.
1⃣ 1000BASE-SX (Short-range)
2⃣ 1000BASE-LX (Long-range)
Key Takeaway: Gigabit Ethernet supports both fiber op c and copper cables, offering flexibility for
short- and long-distance networking needs.
Overview
IEEE 802.5 Token Ring is a MAC protocol that uses a token-passing technique for network access. It
operates in a round-robin fashion, ensuring fair access to all sta ons.
How It Works:
A small frame (token) con nuously circulates the network when no sta on is transmi ng.
The frame travels around the ring and is received by the intended sta on.
A new token is generated a er transmission is complete and the original frame returns.
Advantages:
Disadvantages:
Token Maintenance Required – Loss or duplica on of the token disrupts the network.
Delays Under Light Load – Sta ons must wait for the token.
Requires a Monitor Sta on – To prevent token loss and duplica on.
Conclusion: The token ring protocol ensures fair access and efficiency under heavy loads, but it
requires strict token management to func on properly.
The IEEE 802.5 MAC frame consists of several fields used for controlling data transmission in a Token
Ring network.
Frame Structure
1⃣ Star ng Delimiter (SD)
Format: PPPTMRRR
o T (Token Bit):
o RRR (Reserva on Bits): Allows sta ons to reserve a token for later use.
Defines the type of frame (LLC data frame or MAC control frame).
6⃣ Data Unit
Contains:
Contains: