In data communication systems, while transferring binary data from the source to destination the data bits may be corrupted due to the channel noise. These errors in the data bits can be detected and/or corrected using error control codes . The main idea of error control codes is to append some appropriate redundant check bits to the information bits at the transmitter so that at the receiver, the errors in the information and check bits can be detected/corrected using these extra redundant bits.

The data errors can be broadly classified as symmetric, asymmetric and unidirectional. In symmetric errors both 1 to 0 and 0 to 1 errors can occur in a received word. In asymmetric error case only one type, say 1 to 0 errors can occur. Finally, in the case of unidirectional case, the errors within a data word are of same type, either all 1 to 0 or all 0 to 1, but the error type is not known a priori.

In this proposal, some efficient design of error control codes for asymmetric and unidirectional errors are investigated. Furthermore, some novel error control techniques using automatic repeat request (ARQ) method are also studied. In ARQ protocol, error detecting codes are used in the system. Whenever a receiver receives a word, it checks whether there are some errors in the word. If none, it accepts the word and sends a positive acknowledgement to the transmitter; the transmitter then sends the next word. However, if there are some errors then the receiver rejects the word and sends a negative acknowledgement to the transmitter; the transmitter then sends the same word again. This process continues till the receiver gets the correct word. When the errors are of asymmetric or unidirectional type, some novel schemes of code combing are proposed. Here, instead of discarding the corrupted word at the receiver end, it is combined with the next transmitted (same) word using bit-by-bit OR or AND operation. The performance analysis of these methods will be investigated. The applications of these low cost error control schemes for many computer and communication systems will be evaluated.

Project Start
Project End
Budget Start
2007-07-01
Budget End
2011-06-30
Support Year
Fiscal Year
2007
Total Cost
$152,764
Indirect Cost
Name
Oregon State University
Department
Type
DUNS #
City
Corvallis
State
OR
Country
United States
Zip Code
97331