Soft Output Viterbi Algorithm Decoder Research Articles

Efficient SOVA decoding and enhanced early termination mechanism based on new attenuation factors

This paper proposes two attenuation factors (AF’s) to improve the performance of the soft-output Viterbi algorithm (SOVA) as well as to enhance the early termination mechanism in turbo decoding. The mean square difference between the systematic bipolar coded symbols and the a-posteriori information is used to estimate the first AF. The second AF is computed online for each iteration based on the correlation coefficient between the extrinsic and a-priori information instead of intrinsic information as customary to calculate in literature. The second factor is used in the early termination (ET) scheme which is practically useful to terminate iterations when there is no significant improvement is achieved. In addition, a method for offline computing the AF’s that cover a specific range of signal-to-noise power ratio is provided which results in a reduction of utilization and latency estimates with a shallow degradation in performance. The results show that the proposed scheme outperforms the previous related works by about 0.2 dB at bit error rate (BER) of 10 -5 using interleaver depth of 512 and reducing the average number of iterations (ANI) by about 3 iterations.

Performance Analysis of Serial Concatenation Convolutional Codes Based on Amplitude Matrix

Serial Concatenation Convolutional Codes (SCCCs) encoder is built using a serial concatenation of two Recursive Systematic Convolutional (RSC) encoders, separated by an interleaver. These two RSC encoders depending on the trellis termination criteria are software implemented and the performance of each one of them is analyzed under different conditions and circumstances. The output data from the encoder are multiplied by an amplitude matrix (AM) at the transmitter side and the Inverse of Amplitude Matrix (IAM) at the receiver side. The reliability estimation, log-likelihood algebra, and soft channel outputs for Soft Output Viterbi Algorithm (SOVA) are examined. Then the modified Viterbi metric that incorporates a-priori information used for SOVA decoding is derived. A low memory implementation of the SOVA decoder is presented. The iterative SOVA for SCCCs is described with illustrative examples. The behavior of the SCCCs encoder-decoder scheme is tested under different circumstances with AM and without AM at the AWGN and Rayleigh fading channels with unlike frame sizes (FS) and constraint length (K). The results show that the performance of system with AM outperforms the other conventional system that worked without AM.

Non‐binary SOVA algorithms for decoding of block codes on a sectionalized trellis

AbstractIn this paper, we explore computationally efficient implementations of the soft output viterbi algorithm (SOVA) applied to Soft‐Input Soft‐Output (SISO) decoding of linear block codes. In order to simplify the trellis‐based decoding of binary block codes with SOVA, we use the technique of sectionalization of the trellis, which has been successfully applied to the simplification of the MAP and Max‐Log‐MAP algorithms. Due to the branch complexity of the sectionalized trellis, we define a generalization of a non‐binary version of SOVA. However, the computational complexity of directly applying this approach remains too high for efficient implementation; we thus introduce the concept of non‐binary SOVA (NSOVA) with propagation of bit‐level reliabilities (BLR). This new algorithm is analyzed from a computational complexity viewpoint. Both serial and parallel implementations are explored. Finally, optimal sectionalizations are derived for selected codes; since the normal SOVA decoding is a particular case of NSOVA with BLR, we show that our approach is more efficient than a bit‐level trellis by showing that, for all the codes tested, the optimal trellis is a sectionalized one. Copyright © 2007 John Wiley & Sons, Ltd.

Complexity Reduction in SISO Decoding of Block Codes

SISO decoding for block codes can be carried out based on a trellis representation of the code. However, the complexity entailed by such decoding is most often prohibitive and thus prevents practical implementation. This paper examines a new decoding scheme based on the soft-output Viterbi algorithm (SOVA) applied to a sectionalized trellis for linear block codes. The computational complexities of the new SOVA decoder and of the conventional SOVA decoder, based on a bit-level trellis, are theoretically analyzed and derived for different linear block codes. These results are used to obtain optimum sectionalizations of a trellis for SOVA. For comparisons, the optimum sectionalizations for Maximum A Posteriori (MAP) and Maximum Logarithm MAP (Max-Log-MAP) algorithms, and their corresponding computational complexities are included. The results confirm that the new SOVA decoder is the most computationally efficient SISO decoder, in comparisons to MAP and Max-Log-MAP algorithms. The simulation results of the bit error rate (BER) performance, assuming binary phase -- shift keying (BPSK) and additive white Gaussian noise (AWGN) channel, demonstrate that the performance of the new decoding scheme is not degraded. The BER performance of iterative SOVA decoding of serially concatenated block codes shows no difference in the quality of the soft outputs of the new decoding scheme and of the conventional SOVA.

Application of the Berrou SOVA algorithm in decoding of a turbo code

We present a turbo decoder that is based on a low complexity soft output Viterbi algorithm (SOVA) introduced by Claude Berrou et al. in ICC'93. It is conceptually different from the conventional turbo decoders, where the iteration is based on the a posteriori probabilities (APP). The constituent Berrou SOVA decoders make the maximum likelihood (ML) decision on the correct path in the trellis, and they have lots of similarities with a conventional hard decision Viterbi decoder—their complexity is double that of the Viterbi due to extraction of the soft output. The iterative Berrou SOVA decoder is able to achieve a relatively high speed: the constituent decoders can work at the same pace as their conventional Viterbi counterparts, and no back tracing is needed. The performance is competitive with the traditional decoding algorithms. In comparison with Max-Log-MAP, we have attained a slightly better performance. Copyright © 2003 AEI.

Modification of branch metric calculation to improve iterative SOVA decoding of turbo codes

It is known that the performance of a SOVA (soft output Viterbi algorithm) turbo decoder can be improved, as the extrinsic information that is produced at its output is over-optimistic. A new parameter associated with the branch metrics calculation in the standard Viterbi algorithm is introduced that affects the turbo code performance. Different parameter values show a simulation improvement in the AWGN channel as well as in an uncorrelated Rayleigh fading channel. By choosing the best value of this parameter, a coding gain improvement of 0.25 dB at BER of 10−5 is achieved compared to existing schemes.

Performance of parallel and serial concatenated codes on fading channels

The performance of parallel and serial concatenated codes on frequency-nonselective fading channels is considered. The analytical average upper bounds of the code performance over Rician channels with independent fading are derived. Furthermore, the log-likelihood ratios and extrinsic information for maximum a posteriori (MAP) probability and soft-output Viterbi algorithm (SOVA) decoding methods on fading channels are developed. The derived upper bounds are evaluated and compared to the simulated bit-error rates over independent fading channels. The performance of parallel and serial codes with MAP and SOVA iterative decoding methods, with and without channel state information, is evaluated by simulation over independent and correlated fading channels. It is shown that, on correlated fading channels, the serial concatenated codes perform better than parallel concatenated codes. Furthermore, it has been demonstrated that the SOVA decoder has almost the same performance as the MAP decoder if ideal channel state information is used on correlated Rayleigh fading channels.

Comparative Study of Turbo Codes in AWGN Channel Using MAP and SOVA Decoding

A final-year undergraduate project in error-control coding is discussed. The project consists of the study, software implementation and computer simulation of turbo codes using two different decoding approaches, the Maximum a Posteriori (MAP) and Soft-Output Viterbi Algorithm (SOVA). It combines theory with computer-aided design and provides hands-on experience in the field of channel coding.