Error Correction and Detection - SECDEC

Interview Questions on Error Correction and Detection  ->

1. What is the difference between ECC and CRC ?

2. What are the different technique to detect Error ?

3. Where ECC is useful in design ?

4. How many bits can recover by using ECC ?

5. Why do you need error detection?
Ans: As the signal is transmitted through a media, the signal gets corrupted because of noise and distortion. In other words, the media is not reliable. To achieve a reliable communication through this unreliable media, there is need for detecting the error in the signal so that suitable mechanism can be devised to take corrective actions.

5. Explain different types of Errors?
Ans: The errors can be divided into two types: Single-bit error and Burst error.
• Single-bit Error The term single-bit error means that only one bit of given data unit (such as a byte, character, or data unit) is changed from 1 to 0 or from 0 to 1.
• Burst Error The term burst error means that two or more bits in the data unit have changed from 0 to 1 or vice-versa. Note that burst error doesn’t necessary means that error occurs in consecutive bits.

6. Explain the use of parity check for error detection?
Ans: In the Parity Check error detection scheme, a parity bit is added to the end of a block of data. The value of the bit is selected so that the character has an even number of 1s (even parity) or an odd number of 1s (odd parity). For odd parity check, the receiver examines the received character and if the total number of 1s is odd, then it assumes that no error has occurred. If any one bit (or any odd number of bits) is erroneously inverted during transmission, then the receiver will detect an error.

7. What are the different types of errors detected by parity check?
Ans: If one bit (or odd number of bits) gets inverted during transmission, then parity check will detect an error. In other words, only odd numbers of errors are detected by parity check. But, if two (or even number) of bits get inverted, and then the error remains undetected.

8. How to detect two error bits ?  you can detect single error bit and correct it by using hamming distance , but how will you detect 2 error bits ?

9. What is hamming code distance ?

10. How robust is 2 bit error detection ?

11. How to deal in SECDEC if there is error on overall parity bit ?

12. What are the advantages and disadvantages to use ECC ?
Advantages -:
ECC protects against undetected memory data corruption, and is used in computers where such corruption is unacceptable, for example in some scientific and financial computing applications, or in file servers. ECC also reduces the number of crashes, particularly unacceptable in multi-user server applications and maximum-availability systems

Disadvantages
1. ECC memory usually involves a higher price when compared to non-ECC memory, due to additional hardware required for producing ECC memory modules, and due to lower production volumes of ECC memory and associated system hardware. Motherboards, chipsets and processors that support ECC may also be more expensive
2. It may lower the overall performance of the system , If ECC block is combinational and timings are able to meet then it will not introduce the latency in system , otherwise minimum 1 clock cycle latency will be introduced by ECC.

There are many more questions on ECC, for single bit error correction, it has 100%  error correctable , but when there are more than 1-bit error in data then hamming code turned into worst. different conditions are explained at the bottom.

A SEC-DED Code
For many applications a single error correcting code would be considered unsatisfactory, because it accepts all blocks received. A SEC-DED code seems safer, and it is the level of correction and detection most often used in computer memories.

Different (but equivalent) Hamming codes
Given a specific number N of check bits, there are 2^N equivalent Hamming codes that can be constructed by arbitrarily choosing each check bit to have either "even" or "odd" parity within its group of data bits. As long as the encoder and the decoder use the same definitions for the check bits, all of the properties of the Hamming code are preserved.
Sometimes it's useful to define the check bits so that an encoded word of all-zeros or all-ones is always detected as an error.


What happens when multiple bits get flipped in a Hamming codeword
Multible bit errors in a Hamming code cause trouble. Two bit errors will always be detected as an error, but the wrong bit will get flipped by the correction logic, resulting in gibberish. If there are more than two bits in error, the received codeword may appear to be a valid one (but different from the original), which means that the error may or may not be detected.
In any case, the error-correcting logic can't tell the difference between single bit errors and multiple bit errors, and so the corrected output can't be relied on.


Extended Hamming Code
Extending a Hamming code to detect double-bit errors
Any single-error correcting Hamming code can be extended to reliably detect double bit errors by adding one more parity bit over the entire encoded word. This type of code is called a SECDED (single-error correcting, double-error detecting) code. It can always distinguish a double bit error from a single bit error, and it detects more types of multiple bit errors than a bare Hamming code does.
It works like this: All valid code words are (a minimum of) Hamming distance 3 apart. The "Hamming distance" between two words is defined as the number of bits in corresponding positions that are different. Any single-bit error is distance one from a valid word, and the correction algorithm converts the received word to the nearest valid one.
If a double error occurs, the parity of the word is not affected, but the correction algorithm still corrects the received word, which is distance two from the original valid word, but distance one from some other valid (but wrong) word. It does this by flipping one bit, which may or may not be one of the erroneous bits. Now the word has either one or three bits flipped, and the original double error is now detected by the parity checker.
Note that this works even when the parity bit itself is involved in a single-bit or double-bit error. It isn't hard to work out all the combinations.

Table of Contents