Error Correction vs Error Detection

Two Different Goals

Error detection (covered by parity, checksums and CRC in this section) only tells a receiver that something went wrong. Error correction goes a step further - it can actually identify and fix the error, without needing to ask the sender to resend anything.

What Happens After Detection, Without Correction

Error detected -> data discarded -> receiver requests a retransmission -> sender resends

Detection-only schemes rely on this "detect and retransmit" pattern, which works well when retransmission is cheap and delay isn't critical - exactly the assumption TCP relies on.

How Error Correction Achieves More

Error-correcting codes add extra redundant information calculated in a way that not only flags an error, but pinpoints exactly which bit (or bits) is wrong, allowing the receiver to flip it back to the correct value itself - no retransmission required at all.

The Trade-off: More Overhead, for a Reason

Detection onlyDetection + Correction
Extra data requiredLessMore
Response to an errorDiscard and request retransmissionFix it immediately, no retransmission
Best suited forSituations where retransmission is fast/cheapSituations where retransmission is slow, expensive, or impossible

When Paying That Extra Overhead Makes Sense

Error correction is especially valuable when retransmission is impractical - like a satellite link with several seconds of delay, or a live broadcast where "please resend that" simply isn't an option. In those cases, the extra overhead of correction is well worth avoiding a costly or impossible retransmission.

Written by Lorens Mishra

Software Engineer Notes Management System Administrator

Discussion

0 comments
Sign in to join the discussion.

No comments yet. Be the first to say something.