Week 7 - Data-link error control and Data-Link Control (DLC) Flashcards
What causes errors to occur in the first place?
Interference
What are the 2 types of error?
- Single bit
- Burst error
If a codeword is valid, is it uncorrupted?
No, we cannot guarantee that the codeword is uncorrupted even if it’s valid.
What is Block Coding?
It’s the basis of datawords and codewords.
A message is divided into blocks.
Each block has redundant bits added for error control which creates a codeword (ex: hamming code)
What is hamming distance?
The number of bits that differ between 2 words.
(NOTE: min hamming distance for error detection refers to the min HD between VALID codewords.)
What are linear block codes?
XOR of 2 valid codewords creates another valid codeword.
What is linear block code - Parity-check code?
Add 1 bit to the dataword to create codeword with parity bit. Even parity considered
What is linear block code - Cyclic codes?
Code word rotated to the left to create another valid codeword.
(Ex: 1011000 -> 0110001)
Performance-oriented, used in LAN/WAN.
What is Checksum?
Error detection for message of any length, typically used in network layer.
What is Forward error correction (FEC) for?
Used in real-time media, takes too long to retransmit so we add redundant bits that allow us to correct the error at the transport layer.
What are the 3 FEC techniques?
- Hamming distance
- XOR operations
- Chunk interleaving
How does DLC frame data?
- Fixed size frame
- Variable sized (byte/bit oriented approach)
What 2 ways does the data-link layer implement flow control?
- Buffer overflow, drops packets
- Receiver sends feedback to sender
Data-link error control, what happens when error is detected? (2 outcomes)
- frame dropped (udp?)
- frame dropped + ACK sent (tcp?)
What are the 2 DLC protocols for error and flow control?
- Simple (no flow or error control)
- Stop-and-wait (timer used by sender)
