Odd vs Even Parity in Binary Data Transmission

Parity checking is a fundamental mathematical method used in telecommunications and computer systems to detect single-bit errors in binary data transmission. By appending a single parity bit to a string of binary data, the system ensures that the total count of logic high states (“1”s) across the entire frame adheres to a predefined rule of either evenness or oddness. This article details the mathematical foundations—specifically modulo-2 arithmetic and Boolean logic—underpinning both even and odd parity schemes.

The Mathematical Basis: Modulo-2 Arithmetic

At the core of parity calculations is modulo-2 arithmetic, where any integer sum is divided by 2 to leave a remainder of either 0 or 1. Given a sequence of \(n\) data bits represented as \(b_1, b_2, \dots, b_n\) (where each \(b_i \in \{0, 1\}\)), the arithmetic sum \(S\) is:

\[S = \sum_{i=1}^{n} b_i\]

The parity check relies on the value of \(S \pmod 2\): * If \(S \pmod 2 = 0\), the number of 1s in the data sequence is even. * If \(S \pmod 2 = 1\), the number of 1s in the data sequence is odd.

In digital logic circuits, modulo-2 addition is implemented using the Exclusive OR (XOR) operation (\(\oplus\)).

Even Parity

In an even parity scheme, the parity bit \(p_{\text{even}}\) is chosen so that the total number of 1s in the transmitted block (data bits plus parity bit) is always an even number:

\[\left( \sum_{i=1}^{n} b_i + p_{\text{even}} \right) \pmod 2 = 0\]

Using Boolean logic, the even parity bit is computed as the chained XOR sum of all the data bits:

\[p_{\text{even}} = b_1 \oplus b_2 \oplus \dots \oplus b_n\]

• Example: For the 7-bit data byte 1011001, the number of 1s is 4 (already even). Therefore, \(p_{\text{even}} = 0\), producing the transmitted 8-bit stream 10110010.
• Example: For the 7-bit data byte 1010001, the number of 1s is 3 (odd). Therefore, \(p_{\text{even}} = 1\), producing the transmitted 8-bit stream 10100011.

Odd Parity

In an odd parity scheme, the parity bit \(p_{\text{odd}}\) is selected so that the total number of 1s in the transmitted block is always an odd number:

\[\left( \sum_{i=1}^{n} b_i + p_{\text{odd}} \right) \pmod 2 = 1\]

Mathematically, this corresponds to the logical inversion (XNOR) of the even parity calculation:

\[p_{\text{odd}} = \neg (b_1 \oplus b_2 \oplus \dots \oplus b_n) = 1 \oplus (b_1 \oplus b_2 \oplus \dots \oplus b_n)\]

• Example: For the 7-bit data byte 1011001, the number of 1s is 4. To make the total sum odd, \(p_{\text{odd}} = 1\), producing the transmitted stream 10110011.
• Example: For the 7-bit data byte 1010001, the number of 1s is 3. To maintain an odd sum, \(p_{\text{odd}} = 0\), producing the transmitted stream 10100010.

Error Detection at the Receiver

Upon receiving a transmitted word consisting of data bits \(b_1', b_2', \dots, b_n'\) and the received parity bit \(p'\), the receiver computes the total syndrome \(C\):

\[C = \left( \sum_{i=1}^{n} b_i' + p' \right) \pmod 2\]

• Even Parity Validation: An error is flagged if \(C \neq 0\).
• Odd Parity Validation: An error is flagged if \(C \neq 1\).

Because altering any single bit changes the result of a modulo-2 sum from 0 to 1 or from 1 to 0, parity checks reliably detect any single-bit transmission error (or any odd number of bit errors). However, because an even number of simultaneous bit flips preserves the original modulo-2 sum, double-bit or even-count bit errors will pass undetected.