Title :
Maximum slope convolutional codes
Author :
Jordan, Ralph ; Pavlushkov, Victor ; Zyablov, Victor V.
Author_Institution :
Dept. of Telecommun. & Appl. Inf. Theor., Univ. of Ulm, Germany
Abstract :
The slope is an important distance parameter for a convolutional code. It can be used to obtain a lower bound on the active burst distance and in this respect essentially determines the error-correcting capability of the code. An upper bound on the slope of rate R=b/c convolutional codes is derived. A new family of convolutional codes, called the maximum slope (MS) code family, is introduced. Tables for rate R=1/2 MS codes with memory 1≤m≤6 are presented. Additionally, some new rate R=(c-1)/c, 3≤c≤6, punctured convolutional codes with rate R=1/2 optimum free distance (OFD) and MS mother codes are presented. Simulation results for the bit error performance of serially concatenated turbo codes with MS component codes are presented.
Keywords :
concatenated codes; convolutional codes; error correction codes; error statistics; turbo codes; active burst distance; bit error performance; component code; error-correction code; maximum slope convolutional code; mother code; optimum free distance code; serial concatenated turbo code; Concatenated codes; Convolutional codes; Decoding; Error probability; Feedback; Information theory; Mathematics; Memoryless systems; Reliability theory; Upper bound; Active distance; convolutional code; free distance; slope;
Journal_Title :
Information Theory, IEEE Transactions on
DOI :
10.1109/TIT.2004.834780