PARITY BIT: WHAT IT IS FOR, HOW IT WORKS - TECHNOLOGY - 2024

The parity bit is a parameter with a value of 0 or 1 that is used in a transmission error detection method in which a 0 or a 1 is added to each group of 7-8 bits (byte). The aim is that each byte always has an odd total quantity of “1” or an even total quantity of “1”, according to the established parity.

Parity is an error detection technique used in asynchronous communications. It is used to verify the integrity of each byte within the transmitted stream. For example, if odd parity is set, any byte that is received from a transmission with a total number of "1s" that is even must contain an error.

Source: pixabay.com

Two types of parity are used: even parity, where a parity 1 bit is added if there is an odd total number of “1” bits in the previous byte, and odd parity, where the opposite is done. With this method you will only be able to know that an error has occurred, but you will not know where the error occurred.

What is the parity bit for?

When sending digital data, there may be an error between the transmitted code and the received code. There are many sources of error in the form of different types of noise, such as EM noise or thermal noise.

Therefore, it is necessary to implement some method to verify if the codes or bytes that are received are in error or not.

However, how can the receiver know if the received code is in error or not? It is impossible for the receiver to know the code before receiving it.

For example, suppose the sender transmits the code 01100110, but after passing through a noisy line, the receiver receives the code 00100110. The receiver will not know that it has received a code with an error in the second bit.

It is impossible for the receiver to know that the message has an error in the first bit, because that would imply that the receiver already knows the message from the transmitter before transmission.

Error control

The problem that the receiver has of being able to verify that there is an error can be solved by using an error control encoding.

The central idea of ​​error control coding is to add an extra bit in the information to be sent so that the error is detected and corrected. There are many error handling encodings. The simplest is the parity bit.

The parity bit is added to each byte that is transmitted. This bit is used to check that the information has been delivered accurately.

The parity bit for each byte is set so that all bytes have an odd number or an even number of "1" bits.

Example

Suppose two drives communicate with even parity, which is the most common form of parity check.

Depending on the transmission unit, it sends the bytes and first counts the number of bits "1" in each group of seven bits (byte). If the number of bits “1” is even, set the parity bit to 0. If the number of bits "1" is odd, set the parity bit to 1. In this way, each byte will have an even number of bits "1".

By the receiver, each byte is verified to ensure that it has an even number of "1" bits. If an odd number of "1" bits are found in the byte, the receiver will know that an error occurred during transmission.

Previously, both the receiving entity and the sender must have agreed on the use of parity verification and whether the parity should be odd or even. If both sides are not configured with the same sense of parity, it will be impossible to communicate.

Error detection

Parity checking is the simplest technique for detecting communication errors.

However, although it can detect many errors, it is not infallible, since it is not capable of detecting the arrangement when an even number of bits is changed in the same byte by electrical noise.

Parity checking is used not only in communications, but also to test memory storage devices. For example, many personal computers perform a parity check whenever a byte of data is read from memory.

How does it work?

Suppose you have 7-bit data codes and an additional bit, which is the parity bit, is added to form an 8-bit data code. There are two methods that can be used: even parity and odd parity.

As a sample, the even parity method can be taken. You would do the opposite if you took the odd parity method.

Even parity method

This method indicates that the parity bit to add must be such that the total amount of "1" in the final code is even. For example:

Therefore, for the first 7-bit code: 0010010, with an even quantity of “1” (2), the transmitted 8-bit code will be: 00100100, with an even quantity of “1” (2).

For the 7-bit code 1110110, with an odd quantity of "1" (5), the transmitted 8-bit code will be 11101101, with an even quantity of "1" (6).

After the receiver gets the 8 bits, it will check the quantity of “1” in the received code, if the quantity of “1” is even, that means there is no error, if the quantity is odd, that means that a error.

When the calculated parity of the received byte does not match the value of the received parity bit, a parity error is said to have occurred and normally the byte is discarded.

In the event of an error, the receiver will alert the transmitter to send the code again.

Not infallible

However, there is a drawback with these parity methods, if the code 1110110 is converted by line noise to 11111001, causing a 2-bit error, then this method cannot detect that an error has occurred.

Parity is good at detecting errors and will always detect any odd number of errors in a received byte. However, if there are an even number of errors, the parity checker will not be able to find the error.

References

1. Vangie Beal (2019). Parity checking. Webopedia. Taken from: webopedia.com.
2. Electronics Research Group (2019). Character Parity. Taken from: erg.abdn.ac.uk.
3. Vocabulary (2019).. Parity bit. Taken from: vocabulary.com.
4. Angms (2013). The most simple error control code - Parity Bit. Taken from: angms.science.
5. Christensson, (2011). Parity Bit Definition. Techterms. Taken from: techterms.com.