Cagri C. Tanriover

Improving turbo code error performance by multifold coding

This article presents a simple turbo coding technique to improve the error performance of a convolutional rate-1/3 turbo code by shaping its weight spectrum closer to the binomial weight distribution of a random code. This technique can be applied to both symmetric and asymmetric rate 1/3 turbo codes to achieve additional coding gain.

Image transmission using unequal error protection combined with two-fold turbo coding

Two-fold turbo codes have recently been introduced as the least complex member of the multifold turbo code family, and they have been shown to outperform their conventional symmetric and asymmetric cousins in the waterfall region. The article introduces a simple method that allows two-fold turbo codes to offer bilevel unequal error protection. Also, the application of the unequal error protection to a still image coding technique, i.e. hue-saturation-luminance coding, is explained and the pixel error performance results under Gaussian conditions are presented. According to the simulation results, the proposed scheme outperforms its conventional counterpart by up to 0.7 dB. Furthermore, with the new unequal error protection coding scheme, under certain channel conditions, significant savings in decoder iterations are achieved, without compromising the received image quality. Although the unequal error protection is limited to bilevel within the scope of this article, the same technique can be used to provide multilevel error protection.

Multifold turbo codes

The multifold turbo coding, designed to provide multiple extrinsic estimates and LLRs on a bit basis during decoding iterations, is introduced. Component encoders accept various combinations of information segments, which are obtained by equally dividing the information block prior to encoding. Grouped segments are then interleaved and encoded separately, while the information block is left uncoded. This novel sectionalization of the information sequence, combined with separate interleaving and encoding, bring the weight distribution closer to the binomial distribution of a random code. The effectiveness of the new technique is illustrated by the error performance of several symmetric and asymmetric 2-fold turbo codes in comparison with their equivalent turbo codes.

Enhanced Image Coding for Noisy Channels

This paper explores the application of a combined error resilient coding scheme to image transmission over time-varying noisy channels. To improve performance at low signal-to-noise ratios, turbo coding is incorporated into the system. Demonstrated through simulations, this novel combination of source and channel coding is shown to correct and restrict errors incurred during transmission over Additive White Gaussian Noise (AWGN) and Rayleigh Fading channels. The error correcting capability of the coding scheme also illustrated with compressed and uncompressed image transmission results which are comparable in terms of their visual quality.

Unequal Two-Fold Turbo Codes

Two-fold turbo codes have been recently introduced as the least complex member of the multifold turbo code family, and they have been shown to outperform their conventional symmetric and asymmetric cousins in the waterfall region. This article introduces a simple method that allows two-fold turbo codes to offer bi-level unequal error protection. Also, the application of the unequal error protection to a still image coding technique, i.e. huesaturation-luminance coding, is explained and the pixel error performance results under Gaussian conditions are presented. According to the simulation results, the proposed scheme outperforms its conventional counterpart by up to 0.7 dB. Furthermore, with the new unequal error protection coding scheme, under certain channel conditions, significant savings in decoder iterations are achieved, without compromising the received image quality. Although the unequal error protection is limited to bi-level within the scope of this article, same technique can be used to provide multilevel error protection.

Crossover codeword algorithm for iteration control

During iterative decoding, a monotonic increase in the magnitude of the log-likelihood ratios is a sign of convergence to the maximum likelihood solution. When the log-likelihood ratios undergo frequent sign changes during decoding, the decoder generally fails to converge and the number of bit errors within a frame increases. In such cases, little gain is achieved by performing further iterations. Therefore, utilizing iteration termination methods is beneficial for minimizing the decoding latency, and reducing the operational complexity of turbo decoders. The stopping criteria that have been proposed until now are mainly based on the convergence of log-likelihood ratios of the constituent component decoders. In this paper, we propose a new criterion based on the prediction of the number of bit errors in a decoded frame. The iterations are stopped when the bit error rate of the turbo decoder falls below a specified threshold. This technique is found to be particularly effective in the waterfall region.