ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Turbo Codes in AMC Systems For Blind Identification

Journal: International Journal of Electronics, Communication & Instrumentation Engineering Research and Development (IJECIERD) (Vol.7, No. 6)

Publication Date:

Authors : ; ;

Page : 1-10

Keywords : Blind Identification; Turbo Codes; zero Insertion Decoding; LLR;

Source : Downloadexternal Find it from : Google Scholarexternal


Adaptive modulation and coding (AMC) systems are essential in Blind identification for channel codes. Since Turbo codes are popular in AMC systems, it's necessary to identify its parameters. In this paper, we focus on the identification for Turbo codes from a closed-set. The proposed approach firstly identifies the first component code by accumulating Log-Likelihood Ratio (LLR) for syndrome a posteriori probability, then the interleaver and the other component code are identified by decoding based on zero insertion and LLR accumulation. This approach is robust to noise due to LLR. Moreover, it applies to both symmetric Turbo codes with two same component codes and asymmetric Turbo codes with two different component codes. Simulation results demonstrate that the proposed blind identification scheme is able to identify Turbo codes at signal-to-noise ratio (SNR) larger than 3.5dB. The improvement for identification of RSC2 and the application for Turbo codes with various rates. we design a novel blind encoder parameter estimator for turbo codes. In this scheme, we propose to separate the feedback components from the forward path in a recursive convolution encoder so as to blindly estimate the parameters. The simulations demonstrate the average estimation performance can lead to more than 95% accuracy for the channel condition with the signal-to noise ratio at 5 dB. Once the encoder parameters are blindly estimated, the corresponding decoder can be implemented to retrieve the original information sequences.

Last modified: 2018-01-17 16:16:17