MedVision ad

Parity Checking (1 Viewer)

goan_crazy

Hates the waiting game...
Joined
Sep 28, 2004
Messages
6,225
Location
Sydney
Gender
Male
HSC
2005
Can someone please help me 2 understand what parity checking is and what it involves
i know its a method of error checking
the other 2 methods are checksum? and cyclic redundancy checking
but can some1 explain parity?
thanks
 

goan_crazy

Hates the waiting game...
Joined
Sep 28, 2004
Messages
6,225
Location
Sydney
Gender
Male
HSC
2005
odd and even yes
its like 101010101 ODD


100101100 EVEN

i dont get it
*sigh*
 

Lazarus

Retired
Joined
Jul 6, 2002
Messages
5,965
Location
CBD
Gender
Male
HSC
2001
I can give you the general idea.

Let's say you want to transmit to someone the following stream of bits: 1010101

The parity of a number refers to whether the number is odd or even.

The parity of that bit stream is the parity of the sum of the bits:

1 + 0 + 1 + 0 + 1 + 0 + 1 = 4 = EVEN. (Because four is an even number.)

The sender encodes the parity of the stream into the bit sequence itself by designating a specific bit (usually the eighth) to be the parity bit. If the parity is even, the parity bit is set to 1. If the parity is odd, the parity bit is set to 0.

So, the stream might be sent as follows: 11010101

The parity bit has been bolded.

This allows the recipient of the data to check for minor transmission errors. If a single bit becomes garbled or messed up during the transfer (i.e. switching from 1 to 0 or vice versa), the parity of the stream will be different to the parity indicated by the parity bit.

Note that it's only a very simple check - if TWO bits switched (or any even number of bits), the parity wouldn't change, and the error(s) might go unnoticed.
 

lourai*87

~"*_*"~
Joined
Jun 30, 2004
Messages
745
Location
in the wigwam of a Marsh-wiggle
Gender
Female
HSC
2005
Lazarus said:
1 + 0 + 1 + 0 + 1 + 0 + 1 = 4 = EVEN. (Because four is an even number.)

The sender encodes the parity of the stream into the bit sequence itself by designating a specific bit (usually the eighth) to be the parity bit. If the parity is even, the parity bit is set to 1. If the parity is odd, the parity bit is set to 0
I thought if if the parity is even, the parity bit had to be 0 to keep it even But if the parity was odd then it had to be 1 to keep it odd?
 

Lazarus

Retired
Joined
Jul 6, 2002
Messages
5,965
Location
CBD
Gender
Male
HSC
2001
It could be the other way around. I don't know. I was only describing the general idea.

I don't know whether the parity bit itself is included in the count - wouldn't that mean it would always have to be set to 0? A value of 1 is always going to change the sum from odd to even or vice versa.

Try asking Google. :)
 

Huratio

Moderator - UTS
Joined
Jun 24, 2004
Messages
2,504
Gender
Male
HSC
2005
Lazarus said:
I can give you the general idea.

Let's say you want to transmit to someone the following stream of bits: 1010101

The parity of a number refers to whether the number is odd or even.

The parity of that bit stream is the parity of the sum of the bits:

1 + 0 + 1 + 0 + 1 + 0 + 1 = 4 = EVEN. (Because four is an even number.)

The sender encodes the parity of the stream into the bit sequence itself by designating a specific bit (usually the eighth) to be the parity bit. If the parity is even, the parity bit is set to 1. If the parity is odd, the parity bit is set to 0.

So, the stream might be sent as follows: 11010101

The parity bit has been bolded.

This allows the recipient of the data to check for minor transmission errors. If a single bit becomes garbled or messed up during the transfer (i.e. switching from 1 to 0 or vice versa), the parity of the stream will be different to the parity indicated by the parity bit.

Note that it's only a very simple check - if TWO bits switched (or any even number of bits), the parity wouldn't change, and the error(s) might go unnoticed.

i think in IPT the parity bit goes at the end so: 10101011
 

_muse_

Come on join the joyride
Joined
May 19, 2004
Messages
2,019
Location
hunter valley
Gender
Female
HSC
2005
alright, in my effort to be nice.. ill write whats in my textbook about parity..and whats in the excel book about parity.. then u get 2 general ideas which SHOULD help you guys to understand.. if it doesnt, then shoot me.

TEXT BOOK:
Parity checking is a method of checking for errors in data transmission using an additional bit called a parity bit. This bit is used only for the purpose of identifying whether the bits being moved have arrived successfully. When data is represented using an 8-bit ASCII, then a ninth bit is used that holds a 0 or a 1. When an odd parity is chosen, the number of 1s in the 9 bits must be odd. This means that
. The sender and receiver can also decide to send an even parity, in which case the number of 1s in the 9 bits must be even. If an error has occured in a single bit, then the parity will be different and an error in transmission has occured.

EXCEL BOOK:
Seven bits represent the normal ASCII character set when an 8-bit byte is used in a computer. The 8th bit can be used to check whether that byte has been received correctly by using the ideas of parity.
It works as follows:

- the originating computer counts the number of 1s and determines whether there is an odd or even number.
- the 8th bit in the byte is then used to create a byte with an even number of 1s if even parity is set, or an off number of 1s if odd parity is set in the protocol governing the particular communication session.

The transmission of the ASCII character 'R' demonstrated this process.

0 1 0 1 0 0 1 0 (Parity = bold)

Because the number of 1s is three, which is odd, we add 0 as the parity bit to maintain odd parity checking.



Problems with parity checking

- it can detect 1 error but is stumped if there are two.
- if two bits flip value, then the parity bit will not alert the receiver.
- Any byte with an even numbers of errors will slip through undetected.


make sense?
 

Huratio

Moderator - UTS
Joined
Jun 24, 2004
Messages
2,504
Gender
Male
HSC
2005
why can't the IPT course just omit the parity and checksum methods.. its not like many things use it anymore seeing those many disadvantages. Cyclic redundancy check is widely used now and should be the only on on the IPT syllabus.
 

Lwaxana

Betazed Ambassador
Joined
Oct 16, 2002
Messages
430
Location
in a galaxy far far away
Gender
Female
HSC
N/A
That's what I have been saying about System flowcharts

they are rarely used in Industry anymore
so why include them in the syllabus
 

_muse_

Come on join the joyride
Joined
May 19, 2004
Messages
2,019
Location
hunter valley
Gender
Female
HSC
2005
system flowcharts are evil. ive never been able to do them...

maybe they jus put parity and everything in there to fill in space on the syllabus - we dont need it, but they do to fill up the missing spots :)
 

melsc

Premium Member
Joined
Aug 17, 2004
Messages
6,365
Location
Chasing ambulances in the Inner West...
Gender
Female
HSC
2005
_muse_ said:
system flowcharts are evil. ive never been able to do them...

maybe they jus put parity and everything in there to fill in space on the syllabus - we dont need it, but they do to fill up the missing spots :)

They are evil i can never remember the symbols
 

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Top