Recursive systematic convolutional codes and application to parallel concatenation

Abstract

This paper deals with a new class of convolutional codes called recursive systematic convolutional (RSC) codes. These codes are very attractive when compared with the best known convolutional codes at low signal to noise ratio (SNR) and high code rate. A comparison with the best known convolutional codes is performed in terms of weight spectrum and bit error rate (BER) on a memoryless Gaussian channel. A new design of parallel concatenation is proposed, well matched to RSC codes and much more attractive than serial concatenation for it needs only one clock timing recovery for the decoding.