We investigate in this paper how to optimize the number of iterations required for sliding window decoding of low-density parity-check (LDPC) codes. The aim of this study is to reduce decoding complexity and overcome the phenomenon of decoder blocking. We propose a new design of SC-LDPC codes based on some 5G short LDPC codes to reduce decoding complexity. Simulation results demonstrate a potential reduction in decoding computational complexity of the order of 30% without significant degradation in decoding performance. In addition to taking care of complexity, this mitigates the problem of sliding window decoder blocking by reducing the probability of observing this phenomenon.

We present the reliability confidence level as a way to quantify reliability for ultra reliable connections in uncertain environments subject to random block-error rate fluctuations. As ultra reliability is usually linked to the low latency constraint, we analyze the reliability confidence level using the latest results on block-error rate in the finite blocklength regime. The analysis is carried out for OFDM-based systems over Rayleigh slow frequency block-fading channels. The reliability confidence level is bounded using analytic expressions which are then applied to solve two optimization problems. We first find the minimal resources (or codeword length) to guarantee a target reliability with a given confidence. We then investigate an optimal resource sharing strategy within the context of 5G New Radio. The solutions for optimization problems are obtained without resorting to cumbersome Monte-Carlo simulations. Numerical evaluations validate the relevance as well as the tightness of the proposed bounds and approximations.

We propose a novel method for noise power spectrum estimation in speech enhancement. This method called extended-DATE (E-DATE) extends the d-dimensional amplitude trimmed estimator (DATE), originally introduced for additive white gaussian noise power spectrum estimation in “Robust estimation of noise standard deviation in presence of signals with unknown distributions and occurrences” (D. Pastor and F. Socheleau, IEEE Trans. Signal Processing, vol. 60, no. 4, pp. 1545-1555, Apr. 2012) to the more challenging scenario of non-stationary noise. The key idea is that, in each frequency bin and within a sufficiently short time period, the noise instantaneous power spectrum can be considered as approximately constant and estimated as the variance of a complex gaussian noise process possibly observed in the presence of the signal of interest. The proposed method relies on the fact that the Short-Time Fourier Transform (STFT) of noisy speech signals is sparse in the sense that transformed speech signals can be represented by a relatively small number of coefficients with large amplitudes in the time-frequency domain. The E-DATE estimator is robust in that it does not require prior information about the signal probability distribution except for the weak-sparseness property. In comparison to other state-of-the-art methods, the E-DATE is found to require the smallest number of parameters (only two). The performance of the proposed estimator has been evaluated in combination with noise reduction and compared to alternative methods. This evaluation involves objective as well as pseudo-subjective criteria.

In this paper, we investigate different HARQ strategies with unreliable feedback channel. In classical approaches, retransmitted packets are different versions of the same user data. In order to improve HARQ performance, we propose to construct redundancy packets by jointly encoding several data packets. This strategy is referred to as HARQ Multiple Packet Incremental Redundancy (MP-IR). It can eliminate imperfect feedback effect on performance at medium to high Signal to Noise Ratio (SNR) values. On the other hand, classical HARQ Single Packet Incremental Redundancy (SP-IR) strategy is more robust at low SNR. This motivates us to introduce a hybrid SP-IR and MP-IR strategy. This strategy can benefit from SP-IR performance at low SNR and MP-IR performance at medium to high SNR. The throughput of the different schemes is analyzed using renewal theory and an appropriate Markov model representation. Simulation results confirming the proposed throughput analysis are also presented.

In the absence of the orthogonal design and a very low coding rate channel code, a new bandwidth-efficient multi-user scheme is proposed, which considers random interleavers for different users and multiple antennas at the receiver. Turbo processing is exploited based on the minimum mean square error (MMSE) soft-input soft-output (SISO) detector with interference cancellation (IC) to obtain the receive diversity. The simulation results show that when the number of users K is no more than the product of the number of receive antennas and the reciprocal of the coding rate, the proposed scheme can achieve the single-user performance, which indicates that the uplink user capacity of wireless networks can be dramatically increased with the proposed scheme when more antennas are implemented at the receiver. Meanwhile, it provides a solution to dealing with underdetermined multiple input multiple output (MIMO) systems.

In this letter we consider the ensemble of codes formed by the serial concatenation of a Hamming code and two accumulate codes. We show that this ensemble is asymptotically good, in the sense that most codes in the ensemble have minimum distance growing linearly with the block length. Thus, the resulting codes achieve high minimum distances with high probability, about half or more of the minimum distance of a typical random linear code of the same rate and length in our examples. The proposed codes also show reasonably good iterative convergence thresholds, which makes them attractive for applications requiring high code rates and low error rates, such as optical communications and magnetic recording.

In this paper, we propose a rate-compatible serially concatenated structure consisting of an outer linear extended BCH code and an inner recursive systematic convolutional code. Rate flexibility is achieved by puncturing the inner code. A two- step code design procedure combining analytical union bounds with Extrinsic Information Transfer charts is used to obtain codes offering very good performance in both the waterfall and the error floor regions over a wide range of code rates. The resulting codes show interesting advantages in terms of convergence and error floor compared to similar structures using convolutional codes as outer codes.

We study the minimum distance of the binary expansion of high-rate Reed-Solomon (RS) codes and product codes in the polynomial basis and show that the binary codes obtained in this way usually have minimum distance equal to the designed symbol minimum distance. We then show that a judicious choice for the code roots may yield binary expansions with larger binary minimum distance and better asymptotic performance. This result is used to design high-rate RS product codes with significantly lower error floors compared to classical constructions.

This letter considers high-rate block turbo codes (BTC) obtained by concatenation of two single-error-correcting Reed-Solomon (RS) constituent codes. Simulation results show that these codes perform within 1 dB of the theoretical limit for binary transmission over additive white Gaussian noise with a low-complexity decoder. A comparison with Bose-Chaudhuri-Hocquenghem BTCs of similar code rate reveals that RS BTCs have interesting advantages in terms of memory size and decoder complexity for very-high-data-rate decoding architectures.

Reed-Solomon codes are block-based error correcting codes with a wide range of applications in digital communications and storage. Recently, block turbo codes using Reed-Solomon component codes have been introduced. This was motivated by the highest code rate property of Reed-Solomon codes and their efficiency for burst error correction. In fact, the main advantage of Reed-Solomon block turbo codes is for high code rate applications. For these applications, the code length and consequently the decoder complexity are smaller than for usual Bose-Chaudhuri-Hocquenghem block turbo codes. This paper presents a block turbo decoder architecture using Reed-Solomon component codes. Our elementary soft input soft output decoder is dedicated to Reed-Solomon codes (31, 29, 3) with single error correction power. To the authors' knowledge, this is the first published architecture implementing this type of decoder. Experimentation has been done on a Stratix-based NIOS development board

This paper deals with a low complexity receiver scheme where equalization and channel decoding are jointly optimized in an iterative process. We derive the theoretical transfer function of the infinite length linear minimum mean square error (MMSE) equalizer with a priori information. A practical implementation is exposed which employs the fast Fourier transform (FFT) to compute the equalizer coefficients, resulting in a low-complexity receiver structure. The performance of the proposed scheme is investigated for the enhanced general packet radio service (EGPRS) radio link. Simulation results show that significant power gains may be achieved with only a few (3-4) iterations. These results demonstrate that MMSE turbo equalization is an attractive candidate for single-carrier broadband wireless transmissions in long delay-spread environments.

We describe the implementation of a low-complexity minimum mean-square error (MMSE) turbo-equalizer on the Texas Instruments (TI) TMS320VC5509 device, a low-cost 16-bit fixed-point DSP typically designed for mobile terminals. A data rate of 207 Kb/s per iteration has been achieved. With carefully optimized data quantization, the resulting fixed-point receiver exhibits virtually no performance degradation with respect to an ideal (unquantized) floating-point turbo-equalizer.