Error Checking Methods (Checksum, CRC, and Parity Bits) (1 Viewer)

[elliott07]

Hey guys, ive been having a hard time trying to understand these three types of error checking methods and questions such as this:

could anyone who's good at this explain in baby terms how each works? Thanks <3

 

[wizzkids]

Parity bit is the least accurate form of error checking, but it can be performed extremely fast in hardware and therefore it has an important place in data transmission. Parity bit is added to the data packet to ensure that there is an even number of ones in the packet. It can be fooled by a 2-bit error.
Check-sum will catch a two-bit error, but check-sum can be fooled if two eight-bit words are received out of order, as in the example above, or if there is multiple zero-bit inserted that leaves the check-sum intact.
CRC performs a mathematical manipulation of all the bits (e.g. a polynomial formula) and therefore it is sensitive to the position of all the bits, as well as their values. Therefore CRC will detect that two eight-bit words have been received in the wrong order, but it takes a lot more computing power to perform CRC.

 

[ranjeet.singh]

Parity bit is the least accurate form of error checking, but it can be performed extremely fast in hardware and therefore it has an important place in data transmission. Parity bit is added to the data packet to ensure that there is an even number of ones in the packet. It can be fooled by a 2-bit error.
Check-sum will catch a two-bit error, but check-sum can be fooled if two eight-bit words are received out of order, as in the example above, or if there is multiple zero-bit inserted that leaves the check-sum intact.
CRC performs a mathematical manipulation of all the bits (e.g. a polynomial formula) and therefore it is sensitive to the position of all the bits, as well as their values. Therefore CRC will detect that two eight-bit words have been received in the wrong order, but it takes a lot more computing power to perform CRC.

How does CRC perform the check to find if two eight bit words in wrong order? I don't understand how the polynomial formula actually works

 

[wizzkids]

The exact mathematics of the CRC is outside the scope of the IPT Course. If you have a copy of Samuel Davis' book then you can find a few examples on pages 253-256. There are many CRC standards, but they all require a "generator polynomial" which is just a binary number. The sender and the receiver must agree in advance what generator polynomial they will use. The length of the generator polynomial determines how rigorous the error detection will be. Typical lengths are 16 bits and 32 bits for the generator polynomial. With this many bits, the CRC formula will easily detect if two 8-bit words are received out-of-order.