Some Extended Results on the Design of Punctured Serially Concatenated Convolutional Codes

Published online: Apr 5, 2017 Full Text: PDF (308 KiB) DOI: 10.24138/jcomss.v2i3.286
Cite this paper
Authors:
Massimiliano Laddomada, Bartolo Scanavino

Abstract

The aim of this paper is twofold. On one hand, it presents the results of the search for good punctured systematic recursive convolutional encoders suitable for application in serially concatenated convolutional codes (SCCCs) operating in two different target regions: at low-to moderate signal-to-noise ratios (SNRs), i.e., in the so called waterfall region, and at high SNRs. On the other hand, it provides some useful design guidelines for choosing the constituent encoders in an SCCC. The results of the search for good SCCCs operating in the waterfall region rely upon an effective algorithm, based on density evolution technique, first proposed in a companion paper. Good punctured SCCCs were obtained through considerations deduced by the behaviour of the bit error probability of an SCCC for high values of both SNR and interleaver length, i.e., through asymptotic considerations. The mother codes in the serial concatenation are rate 1/2 recursive convolutional encoders (RCC) found by an exhaustive search for encoders tailored to SCCC schemes, using two different selection criteria. Extensive tables of optimized puncturing patterns for various mother codes and SCCCs are presented along with sample simulation results.

Keywords

Convolutional codes, high rate, punctured, serially concatenated convolutional codes, turbo codes
Creative Commons License 4.0
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.