Error connection
RAID Recovery - Error connection
Automatic repeat request:
It is a process which control error for data communication which uses error detection codes, receive the acknowledgement positive or negative massages and timeouts to save consistent data transmission. Generally, if the transmitter fails to receive the acknowledgement massage before the timeout, it retransmits the frame till it determines number of retransmissions.
Stop-and-wait ARQ, Go-Back-N ARQ, and selective Repeat ARQ are ARQ protocols. ARQ and forward error correction formed Hybrid ARQ.
Error correcting Code:
Error correcting codes are used in computer data storage such as in dynamic RAM and data transmission. It is a code where every data signal fits to particular construction rules which enter in the received signal that can be automatically corrected.
The essential for the transmitter to apply one or more detecting codes then users use the exact code to make out the exact error location. But decoder can fix the error in case any error occurs in transmission.
1. Repetition scheme: Users can correct single bit error, if the transmitter repeats the error for three times. It is called triple modular redundancy.
2. Parity schemes: If the transmitter sends parity bits covering overlapping groups of data bits, any single error can cause parity error. But the users are also capable of correcting the single-bit mistake by switching the one bit fails the check excepting the group that has passed the check.
3. Cyclic redundancy checks: Here, transmitter adds a CRC code to a massage which will appear only in mistakes and give a clue to users to correct it.
4. Hamming distance based checks: It transform many bit of errors into a valid hamming code word to some other valid hamming code word. Here, users can correct any bit of error with the help of “closest” valid hamming code.
There are some codes that can correct a limited number of bit errors. Single error correcting or SEC can rectify one error, Hamming Code can rectify double error and some advanced codes are capable of correcting more errors. Two major types of codes are: Convolutional codes and block codes. Others such as Hamming code, BCH code, Reed-Solomon code, Binary Golay code, and Low-density parity –check codes came into existence later.
The importance of Shannon’s theorem is crucial in error correction. It describes the utmost attainable efficiency of an error-correcting scheme against the desired levels of noise interference. Further, the efficiency of the coding is determined in terms of code rate and the coding gain.
Error correcting memory:
Soft errors are common in computer’s Dynamic Random Access Memory (DRAM) used mostly in satellites and space probes. Such type of memory controllers employs a Hamming Code or Triple Modular Redundancy. It is designed in such a way that in case of a Single Event Upset (SEU), only a single error occurs, and further corrected by a single-bit error correcting code.
