Recent research has shown that the principal cause of errors on some digital data networks is due to cycle slips that occur during clock recovery, instead of erroneous decisions made at the detector. Conventional codes are unable to tolerate cycle slips, and can make the problem worse. The objectives of this project are to develop new and efficient coding algorithms that can correct slips. In addition, the coding algorithms studied will have a limited error correction capability, so that the occasional error can be corrected and so that the resulting algorithms are not sensitive to errors. The research has two different directions. Previous research in the area has led to the development of some fairly efficient codes that can tolerate cycle slips at the boundary between codewords, but nowhere else. The research examines ways to extend or modify these codes so that they can correct cycle slips that occur anywhere in the codeword. Another direction for the research is the study of novel coding methods for correcting cycle slips, including source coding and convolutional code structures.***//
