Error Detection and Correction Flashcards
types of errors
single bit errors
burst errors
multiple bit errors
what is a single bit error
when one bit has been changde
what is a burst error
two or more bits had been corrupted/changed in a row
what is a multiple bit error
more than one bit has been changed
no necessarily in a row
3 ways to detect errors
parity check
cyclic redundancy
checsum
what is the concept of even parity
add an additional bit at the end of the data unit to make sure the total number of ones is even
what is the concept of odd parity
add an additional bit at the end of the data unit to make sure the total number of ones is odd
how does two dimensional parity check work
parity bit for every row and every column
why is 2d parity better than 1d parity
can deal with burst errors
to pass check, result of CRC must be
0
what is the concept of hamming code
parity bits for everything to the power of 2
can detect the position of the error
when might simple parity check not work
when more than one bit is flipped in transit
how does checksum work
data divided into k sections
added using 1s complement
checksum sent along with data
receiver also calculates checksum
complements its result with the checksum received
result should be 1
if the result of the checksum is not 0 what happens to the data
it is discarded
how does CRC work
a sequence of redundant bits are appended onto the end of a data unit so that the resulting data unit becomes exactly divisible by a predetermined binary number
