Open AccessCorrection Impulse Method for Turbo Decoding over Middleton Class-A Impulsive Noise
Author(s) -
Lucian Trifina,
Daniela Tărniceriu,
Mihaela Andrei
Publication year - 2016
Publication title -
advances in electrical and computer engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.254
H-Index - 23
eISSN - 1844-7600
pISSN - 1582-7445
DOI - 10.4316/aece.2016.04011
Subject(s) - decoding methods , computer science , impulse noise , turbo code , impulse (physics) , speech recognition , class (philosophy) , turbo , turbo equalizer , algorithm , artificial intelligence , error floor , engineering , low density parity check code , physics , pixel , quantum mechanics , automotive engineering
The correction impulse method (CIM) is very effective to achieve low error rates in turbo decoding. It was applied for transmission over Additive White Gaussian Noise (AWGN) channels, where the correction impulse value must be a real number greater than the minimum distance of the turbo code. The original version of CIM can not be used for channels modeled as Middleton additive white Class-A impulsive noise (MAWCAIN), because of nonlinearity of channel reliability. Thus, in this paper we propose two ways to modify the method such that it improves the system performances in the case of aforementioned channels. In the first one, the value of the correction impulse is chosen to maximize the channel reliability. It depends on the signal to noise ratio (SNR) and the error rates are significantly improved compared to those obtained by using the correction impulse value applied for AWGN channels. The second version is based on the least squares method and performs an approximation of the correction impulse. The approximated value depends on the SNR and the parameter A of the MAWCAIN model. The differences between the error rates obtained by the two proposed methods are negligible