Data Link Layer Functions and Frame Structures

Data Link Layer Frames

In the link layer, data is organized into units called frames. Each frame has a header, which includes an address and control information, and a tail that is used for error detection.

The header of a Local Area Network (LAN) contains the physical addresses of the source and destination. The header of a frame that is spread by a Wide Area Network (WAN) contains a circuit identifier in the direction field.

Remember that a link is a local area network, a peer-to-peer facility, or another wide-area facility by which systems can communicate using a data link layer protocol.

Core Functions of the Data Link Layer

The data link layer is responsible for the reliable transfer of information through a data transmission circuit. Data transmission is performed by frames, which are units of information with logical sense for the exchange at the data link layer. In the TCP/IP model, it corresponds to the second layer.

Main Functions Include:

• 1. Initiation, termination, and identification
• 2. Targeting and blocking
• 3. Byte and character synchronization
• 4. Demarcation of plot and transparency
• 5. Error control
• 6. Flow control
• 7. Failover
• 8. Management and coordination of communication

Flow Control Mechanisms

Flow control is necessary to avoid overwhelming the receiver. It is normally performed at the transport layer, but sometimes also at the link level. It uses feedback mechanisms, which are often associated with error correction and should not limit the efficiency of the channel. Flow control involves two critical actions: error detection and error correction.

Error Detection and Correction

Error Detection

Error detection is used to identify errors when sending frames to the receiver and attempting to solve them. This is done through various types of codes. Key methods include the CRC (Cyclic Redundancy Code), simple parity (which may be even, where the number of 1s is even, or odd), cross-parity (horizontal and vertical parity), and checksums.

Error Correction

Error correction arises from detection to fix identified errors. It requires adding a number of useful redundant bits to detect and retransmit. Techniques are varied. The Hamming code repeats each bit 3 times, and on failure, the bit with the most repeats is taken. Correction can also be done through cross-parity checks, Reed-Solomon, and Golay.

High-Level Data Link Control (HDLC)

The HDLC protocol is a high-level data link control that is bit-oriented and reflects a sliding window ARQ or continuous flow. There are also character-oriented protocols.