Study on the GA-Based Decoding Algorithm for Convolutional Turbo Codes

  • Authors:

  • Xingcheng Liu;Shishuang Zhang;Zerong Deng

  • Affiliations:

  • Department of Electronic and Communications Engineering, Sun Yat-sen University, Guangzhou, Guangdong, China 510275;Department of Electronic and Communications Engineering, Sun Yat-sen University, Guangzhou, Guangdong, China 510275;Department of Electronic and Communications Engineering, Sun Yat-sen University, Guangzhou, Guangdong, China 510275

  • Venue:
  • ISNN 2009 Proceedings of the 6th International Symposium on Neural Networks: Advances in Neural Networks - Part II
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

A new decoding algorithm for convolutional Turbo codes that is called the Soft-Output Genetic Algorithm (SOGA) is proposed. With good individuals' diversity, wide searching region and global optimizing ability, the SOGA performs better than the Soft-Output MA in terms of BER (Bit Error Rate) with the similar complexity. Simulation results show that when 1/3 code rate, 16-state convolutional Turbo codes are decoded, at BER=10*** 5, the SOGA achieves about 0.2dB gains over the SOMA algorithm and nearly performs the same as the SOVA when BER*** 4. Besides, the SOGA only deals with M states in the total 2v states, so it can save 2v-M registers compared with the SOVA when 1 bit is decoded.