Data Link Layer (OSI Model - Layer 2)

1. Introduction to the Data Link Layer

The Data Link Layer is the second layer in the OSI model and is responsible for node-to-node data trans-
fer and error-free communication between devices on the same physical network. It ensures that data is
properly framed and transmitted over the physical medium (Layer 1) without errors, providing mecha-
nisms for addressing, flow control, and error control.

Main Functions of the Data Link Layer:

Framing: The process of breaking the data from the Network Layer into manageable chunks called
frames. Each frame is encapsulated with necessary headers (like source and destination addresses) for
proper routing.

Addressing: The Data Link Layer adds Media Access Control (MAC) addresses to the frame, ensuring that
data reaches the correct device in a network.

Error Control: It detects and corrects errors that may occur during data transmission, ensuring the in-
tegrity of the data.

Flow Control: Prevents data overflow in slow receivers by regulating the rate of data transfer.

Access Control: Determines how devices gain access to the physical medium (like Ethernet or Wi-Fi) and
manage collisions when two devices attempt to send data simultaneously.

Example: In Ethernet networks, the Data Link Layer defines how devices are addressed using MAC ad-
dresses and how they handle data collisions using Carrier Sense Multiple Access with Collision Detection
(CSMA/CD).

2. Data Link Control Concepts

a) Framing
Framing refers to the process of breaking the data into frames, which are the basic units of transmission
at the Data Link Layer. Each frame contains the data, along with necessary control information, such as
the start and end of the frame, source and destination addresses, error-checking data, and control flags.

Framing Techniques:

Character Count: The first field in the frame specifies how many characters (or bytes) are in the frame. It
is a simple method but not very reliable since if the character count itself gets corrupted, the frame be-
comes unrecognizable.

Byte-Oriented Protocol (Byte Stuffing): Special escape characters are inserted before any special control
byte (like 0x7E in PPP). This ensures that data bytes do not conflict with control bytes.

Bit-Oriented Protocol (Bit Stuffing): In protocols like HDLC, bit stuffing is used where a '0' bit is inserted
after five consecutive '1's to prevent confusion with special flag sequences like 01111110 (used to de-
note the start and end of a frame).
Example: HDLC (High-Level Data Link Control) uses bit stuffing, where a '0' is inserted after five consecu-
tive '1's to avoid misinterpretation of the frame boundaries.

b) Flow Control
Flow control mechanisms prevent the receiver from being overwhelmed with data from the sender.
These techniques manage the pacing of data transmission.

Flow Control Techniques:

Stop-and-Wait Protocol: The sender sends one frame and waits for an acknowledgment before sending
the next. Although easy to implement, this method can be inefficient as the sender must wait after ev-
ery transmission.

Sliding Window Protocol: The sender and receiver maintain a window (a set of frames that can be sent
before receiving an acknowledgment). This method supports full-duplex communication and allows the
sender to transmit multiple frames at once, improving throughput.

Example: TCP (Transmission Control Protocol) uses a sliding window mechanism to manage flow control
in a connection, allowing multiple packets to be sent and acknowledged without waiting for one packet
at a time.

c) Error Control
Error control ensures that data is transmitted without corruption by detecting errors in transmitted
frames and requesting retransmission if necessary.

Error Control Mechanisms:

Error Detection: Methods like Parity, Checksum, and Cyclic Redundancy Check (CRC) are used to detect
errors in transmitted data.

Error Correction: In cases where errors are detected, data can be corrected using techniques like Auto-
matic Repeat Request (ARQ), where frames are retransmitted when an error is detected.

Acknowledgment and Retransmission: The receiver sends an acknowledgment (ACK) to the sender if the
frame was received correctly, or a negative acknowledgment (NAK) to request a retransmission.

Example: The Go-Back-N ARQ protocol uses sequence numbers to track frames. If an error is detected in
a frame, all subsequent frames are resent, even if they were transmitted correctly.

3. Types of Communication Channels

a) Noiseless Channel
In a noiseless channel, no transmission errors occur. This is an idealized scenario where data is trans-
mitted perfectly.

Features:

No need for error detection or correction.

Protocols are simpler since the assumption is that no errors will occur during transmission.

Use Case: Noiseless channels are often used in theoretical scenarios or simulations where focus is placed
on flow control and framing without complicating factors like noise.

b) Noisy Channel
A noisy channel is a real-world scenario where data can get corrupted due to noise, interference, or sig-
nal attenuation. In this case, error detection and correction mechanisms are necessary.

ARQ Protocols:

Stop-and-Wait ARQ: After sending a frame, the sender waits for an acknowledgment (ACK). If no ACK is
received, the frame is retransmitted.

Go-Back-N ARQ: The sender can send several frames, but if any frame is lost or corrupted, all frames
from that point are retransmitted.

Selective Repeat ARQ: Only the frames that were detected to be in error are retransmitted, making it
more efficient than Go-Back-N.

Example: Wi-Fi and Ethernet use error detection codes like CRC to handle errors and ensure the integrity
of data over noisy channels.

4. Error Detection and Correction Techniques (Expanded)

a) Error Detection
Error detection techniques are used to identify corrupted data during transmission.

Parity Check: A single bit (parity bit) is added to the data to make the total number of 1's either even or
odd. If the total number of 1's doesn't match the expected parity, an error is detected.

Checksum: The sender divides data into equal-sized parts, calculates a checksum (the sum of all parts),
and sends it along with the data. The receiver recalculates the checksum and compares it with the sent
value.

Cyclic Redundancy Check (CRC): A more robust error detection technique where data is treated as a
polynomial and divided by a pre-defined generator polynomial. The remainder of this division is ap-
pended to the data as a CRC value.

b) Error Correction
Error correction goes a step further by not only detecting errors but also correcting them without need-
ing retransmission.

Hamming Codes: These are a class of codes that can correct single-bit errors and detect two-bit errors
by adding redundant bits at specific positions.

Linear Block Codes: These codes add redundant bits to data to form a codeword (a combination of data
and redundant bits) that can be used to detect and correct errors.
Cyclic Codes (CRC): Used primarily for error detection, but when paired with error correction methods,
they can correct errors as well.

Example: Ethernet uses CRC-32 for error detection and correction in frames.

5. Data Link Layer Protocols

a) HDLC (High-Level Data Link Control)
HDLC is a bit-oriented synchronous data link layer protocol used to control communication between net-
work devices.

Frame Types in HDLC:

I-frames (Information Frames): These carry the actual data and sequence numbers.

S-frames (Supervisory Frames): Used for flow control and error control.

U-frames (Unnumbered Frames): Used for link management, such as establishing or terminating a con-
nection.

Modes of Operation:

Normal Response Mode (NRM): In this mode, the primary device controls the communication link.

Asynchronous Balanced Mode (ABM): Both devices can initiate communication, making it more efficient.

Asynchronous Response Mode (ARM): A less commonly used mode.

Use Case: HDLC is commonly used in point-to-point and multipoint communication links, such as in WAN
(Wide Area Network) communication.

b) PPP (Point-to-Point Protocol)

PPP is a byte-oriented protocol used to establish a direct connection between two nodes, often over se-
rial connections like telephone lines or ISDN.

Features of PPP:

Supports multiple network layer protocols like IP and IPX.

Provides authentication through PAP (Password Authentication Protocol) and CHAP (Challenge Hand-
shake Authentication Protocol).

Uses LCP (Link Control Protocol) for link management and error checking.

Use Case: PPP is commonly used in dial-up internet connections and VPNs (Virtual Private Networks).