arXiv subjects

Hossein Najafi

Publications and source records attributed to Hossein Najafi.

4 recordsLinked to original sources

arXiv2024

Application of Transformers for Nonlinear Channel Compensation in Optical Systems

In this paper, we introduce a new nonlinear optical channel equalizer based on Transformers. By leveraging parallel computation and attending directly to the memory across a sequence of symbols, we show that Transformers can be used effectively for nonlinear compensation (NLC) in coherent long-haul transmission systems. For this application, we present an implementation of the encoder part of the Transformer and analyze its performance over a wide range of different hyper-parameters. It is shown that by proper embeddings and processing blocks of symbols at each iteration and also carefully selecting subsets of the encoder's output to be processed together, an efficient nonlinear equalization can be achieved for different complexity constraints. To reduce the computational complexity of the attention mechanism, we further propose the use of a physic-informed mask inspired by nonlinear perturbation theory. We also compare the Transformer-NLC with digital back-propagation (DBP) under different transmission scenarios in order to demonstrate the flexibility and generalizability of the proposed data-driven solution.

cs.IT↗

arXiv2023

Ordered Reliability Direct Error Pattern Testing Decoding Algorithm

We introduce a novel universal soft-decision decoding algorithm for binary block codes called ordered reliability direct error pattern testing (ORDEPT). Our results, obtained for a variety of popular short high-rate codes, demonstrate that ORDEPT outperforms state-of-the-art decoding algorithms of comparable complexity such as ordered reliability bits guessing random additive noise decoding (ORBGRAND) in terms of the decoding error probability and latency. The improvements carry on to the iterative decoding of product codes and convolutional product-like codes, where we present a new adaptive decoding algorithm and demonstrate the ability of ORDEPT to efficiently find multiple candidate codewords to produce soft output.

cs.IT↗

arXiv2023

Neural Network Architectures for Optical Channel Nonlinear Compensation in Digital Subcarrier Multiplexing Systems

In this work, we propose to use various artificial neural network (ANN) structures for modeling and compensation of intra- and inter-subcarrier fiber nonlinear interference in digital subcarrier multiplexing (DSCM) optical transmission systems. We perform nonlinear channel equalization by employing different ANN cores including convolutional neural networks (CNN) and long short-term memory (LSTM) layers. We start to compensate the fiber nonlinearity distortion in DSCM systems by a fully connected network across all subcarriers. In subsequent steps, and borrowing from fiber nonlinearity analysis, we gradually upgrade the designs towards modular structures with better performance-complexity advantages. Our study shows that putting proper macro structures in design of ANN nonlinear equalizers in DSCM systems can be crucial for practical solutions in future generations of coherent optical transceivers.

cs.IT↗

arXiv2011

Asynchronous Interference Alignment

A constant K-user interference channel in which the users are not symbol-synchronous is considered. It is shown that the asynchronism among the users facilitates aligning interfering signals at each receiver node while it does not affect the total number of degrees of freedom (DoF) of the channel. To achieve the total K/2 DoF of the channel when single antenna nodes are used, a novel practical interference alignment scheme is proposed wherein the alignment task is performed with the help of asynchronous delays which inherently exist among the received signals at each receiver node. When each node is equipped with M > 1 antennas, it is argued that the same alignment scheme is sufficient to achieve the total MK/2 DoF of the medium when all links between collocated antennas experience the same asynchronous delay.

cs.IT↗