In this paper, we present a construction method of non-binary low-density parity-check (LDPC) convolutional codes. Our construction method is an extension of Felström and Zigangirov construction [1] for non-binary LDPC convolutional codes. The rate-compatibility of the non-binary convolutional code is also discussed. The proposed rate-compatible code is designed from one single mother (2,4)-reg...

In this paper, we suggest a new technique for LDPC parity-check matrix (H-matrix) generation and a corresponding decoding process. The key idea is to construct LDPC H-matrix by using a convolutional encoder. It is easy to have many different coderates from a mother code with convolutional codes. However, it is difficult to have many different coderates with LDPC codes. Constructing LDPC Hmatrix...

In this paper we study periodically time-varying convolutional codes by means of input-state-output representations. Using these representations we investigate under which conditions a given time-invariant convolutional code can be transformed into an equivalent periodic time-varying one. The relation between these two classes of convolutional codes is studied for period 2. We illustrate the id...

Algebraic methods for the construction, design and analysis of series of convolutional codes using row or block structures of unit schemes are developed. The general methods lead to the construction and analysis of series and infinite series of types of convolutional codes and of codes with specific properties. Explicit examples are given and properties may be shown algebraically. Algebraic dec...

Computational units induced by convolutional kernels together with biologically inspired perceptrons belong to the most widespread types of units used in neurocomputing. Radial convolutional kernels with varying widths form RBF (radial-basis-function) networks and these kernels with fixed widths are used in the SVM (support vector machine) algorithm. We investigate suitability of various convol...

It has long been known that convolutional codes have a natural, regular trellis structure that facilitates the implementation of Viterbi's algorithm [30,10]. It has gradually become apparent that linear block codes also/]ave a natural, though not in general a regular, "minimal" trellis structure, which allows them to be decoded with a Viterbi-1ike algorithn] [2,31,22,11,27,14,12,16,24,25,8,15]....

A deep learning approach has been widely applied in sequence modeling problems. In terms of automatic speech recognition (ASR), its performance has significantly been improved by increasing large speech corpus and deeper neural network. Especially, recurrent neural network and deep convolutional neural network have been applied in ASR successfully. Given the arising problem of training speed, w...

A general method for constructing convolutional codes from units in Laurent series over matrix rings is presented. Using group rings as matrix rings, this forms a basis for in-depth exploration of convolutional codes from group ring encodings, wherein the ring in the group ring is itself a group ring. The method is used to algebraically construct series of convolutional codes. Algebraic methods...

Convolutional kernels are basic and vital components of deep Convolutional Neural Networks (CNN). In this paper, we equip convolutional kernels with shape attributes to generate the deep Irregular Convolutional Neural Networks (ICNN). Compared to traditional CNN applying regular convolutional kernels like 3× 3, our approach trains irregular kernel shapes to better fit the geometric variations o...

Constructions of (classical) convolutional codes and their corresponding properties have been presented in the literature [1, 3–8, 12, 15–20]. In [3], the author constructed an algebraic structure for convolutional codes. Addressing the construction of maximum-distance-separable (MDS) convolutional codes (in the sense that the codes attain the generalized Singleton bound introduced in [18, Theo...