Filters








5,683 Hits in 4.2 sec

Combining Deep Learning and Linear Processing for Modulation Classification and Symbol Decoding [article]

Samer Hanna, Chris Dick, Danijela Cabric
2020 arXiv   pre-print
Deep learning has been recently applied to many problems in wireless communications including modulation classification and symbol decoding.  ...  In this paper, we propose a novel neural network architecture that combines deep learning with linear signal processing typically done at the receiver to realize joint modulation classification and symbol  ...  In this work, we propose a deep learning approach combined with receiver signal processing for joint modulation classification and symbol recovery.  ... 
arXiv:2006.00729v2 fatcat:djbvygyvgzefdlxbljr4ty2f4m

Signal Processing Based Deep Learning for Blind Symbol Decoding and Modulation Classification [article]

Samer Hanna, Chris Dick, Danijela Cabric
2021 arXiv   pre-print
While deep learning can solve complex problems, digital signal processing (DSP) is interpretable and can be more computationally efficient. To combine both, we propose the dual path network (DPN).  ...  Blindly decoding a signal requires estimating its unknown transmit parameters, compensating for the wireless channel impairments, and identifying the modulation type.  ...  Our focus in this work is on combining DSP with deep learning and not just using improved deep learning modules for higher classification accuracy.  ... 
arXiv:2106.10543v2 fatcat:26buosg7yfacph465cayj4k7wy

To Learn or Not to Learn: Deep Learning Assisted Wireless Modem Design [article]

S. Xue, A. Li, J. Wang, N. Yi, Y. Ma, R. Tafazolli, T. Dodgson
2019 arXiv   pre-print
Through several physical-layer case studies, we argue for a significant role that machine learning could play, for instance in parallel error-control coding and decoding, channel equalization, interference  ...  Deep learning is driving a radical paradigm shift in wireless communications, all the way from the application layer down to the physical layer.  ...  We argue for the merits of performance-complexity trade-off when using deep learning. 3) Current PHY technologies are designed for linear communication channels, and they are not optimized for future wireless  ... 
arXiv:1909.07791v1 fatcat:ggschowl2fc4xemexn2i6pcz5q

O-Net: A Novel Framework With Deep Fusion of CNN and Transformer for Simultaneous Segmentation and Classification

Tao Wang, Junlin Lan, Zixin Han, Ziwei Hu, Yuxiu Huang, Yanglin Deng, Hejun Zhang, Jianchao Wang, Musheng Chen, Haiyan Jiang, Ren-Guey Lee, Qinquan Gao (+3 others)
2022 Frontiers in Neuroscience  
The application of deep learning in the medical field has continuously made huge breakthroughs in recent years.  ...  and classification.  ...  AUTHOR CONTRIBUTIONS TW, JL, ZHa, ZHu, YH, YD, QG, MD, TT, and GC: concept and design. TW, JL, HZ, JW, MC, and TT: acquisition of data. TW, JL, ZHa, ZHu, QG, and TT: model design.  ... 
doi:10.3389/fnins.2022.876065 pmid:35720715 pmcid:PMC9201625 fatcat:lwqkssbtpnctdndyfytstxiedi

Model-based Deep Learning Receiver Design for Rate-Splitting Multiple Access [article]

Rafael Cerna Loli, Onur Dizdar, Bruno Clerckx, Cong Ling
2022 arXiv   pre-print
To assess its practical performance, benefits, and limits under more realistic conditions, this work proposes a novel design for a practical RSMA receiver based on model-based deep learning (MBDL) methods  ...  , which aims to unite the simple structure of the conventional SIC receiver and the robustness and model agnosticism of deep learning techniques.  ...  and the private stream of the user with the highest modulation order, and random symbol combinations for the rest of the private streams allows for a training set size reduction compared to transmitting  ... 
arXiv:2205.00849v1 fatcat:tjifcac75bcxpbbevpqrcgtp7y

Efficient MIMO Detection with Imperfect Channel Knowledge - A Deep Learning Approach [article]

Qian Chen, Shunqing Zhang, Shugong Xu, Shan Cao
2019 arXiv   pre-print
Hence, a deep learning based efficient MIMO detection approach is proposed in this paper.  ...  Then, we compare the end-to-end approach using deep learning with the conventional methods in possession of perfect channel knowledge and imperfect channel knowledge.  ...  robust detection problem with imperfect channel knowledge. • Efficient Deep Learning Framework for Detection Another issue that has not been solved is the efficient deep learning network architecture  ... 
arXiv:1903.07831v1 fatcat:yvysqptyezgopmdligwnxoljmm

Neural Turbo Equalization: Deep Learning for Fiber-Optic Nonlinearity Compensation [article]

Toshiaki Koike-Akino, Ye Wang, David S. Millar, Keisuke Kojima, Kieran Parsons
2019 arXiv   pre-print
Recently, data-driven approaches motivated by modern deep learning have been applied to optical communications in place of traditional model-based counterparts.  ...  The proposed DNN-TEQ is constructed with nested deep residual networks (ResNet) to train extrinsic likelihood given soft-information feedback from channel decoding.  ...  Modulation classification as well as OSNR monitoring was considered in [5] , and a deep CNN showed an accurate performance in [6] .  ... 
arXiv:1911.10131v1 fatcat:7ghe4qetcbejphjkftl54l3zeq

Deep Neural Networks based Modrec: Some Results with Inter-Symbol Interference and Adversarial Examples [article]

S. Asim Ahmed, Subhashish Chakravarty, Michael Newhouse
2018 arXiv   pre-print
Recent successes and advances in Deep Neural Networks (DNN) in machine vision and Natural Language Processing (NLP) have motivated their use in traditional signal processing and communications systems.  ...  In this paper, we present results of such applications to the problem of automatic modulation recognition.  ...  CONCLUSION Deep neural networks typically use linear optimization techniques to learn parameters in extremely high dimensions and most activations in the network are carefully controlled during the training  ... 
arXiv:1811.06103v1 fatcat:feg7hjfrhneq5chroel4ttl46m

Advances in Joint CTC-Attention based End-to-End Speech Recognition with a Deep CNN Encoder and RNN-LM [article]

Takaaki Hori, Shinji Watanabe, Yu Zhang, William Chan
2017 arXiv   pre-print
We learn to listen and write characters with a joint Connectionist Temporal Classification (CTC) and attention-based encoder-decoder network.  ...  During the beam search process, we combine the CTC predictions, the attention-based decoder predictions and a separately trained LSTM language model.  ...  Finally, these modules are integrated into a Weighted Finite-State Transducer (WFST) for efficient decoding.  ... 
arXiv:1706.02737v1 fatcat:cdswlmukebhnvepo6phfezmnae

OFDM-guided Deep Joint Source Channel Coding for Wireless Multipath Fading Channels [article]

Mingyu Yang, Chenghong Bian, Hun-Seok Kim
2021 arXiv   pre-print
Inspired by recent works on deep learning based JSCC and model-based learning methods, we combine an autoencoder with orthogonal frequency division multiplexing (OFDM) to cope with multipath fading.  ...  The proposed encoder and decoder use convolutional neural networks (CNNs) and directly map the source images to complex-valued baseband samples for OFDM transmission.  ...  For all datasets, we apply linear learning rate decaying for the second half of the training process. As for the OFDM system, the parameters are set to = 64, = 16, = 8, and = 4.  ... 
arXiv:2109.05194v1 fatcat:67cvmgo4rbehvknyy3xlnl7deq

Autoencoder-Based Error Correction Coding for One-Bit Quantization [article]

Eren Balevi, Jeffrey G. Andrews
2019 arXiv   pre-print
This paper proposes a novel deep learning-based error correction coding scheme for AWGN channels under the constraint of one-bit quantization in the receivers.  ...  Our results show that the proposed coding scheme at finite block lengths outperforms conventional turbo codes even for QPSK modulation.  ...  We fill this gap by developing a novel deep learning-based coding scheme that combines turbo codes with an autoencoder.  ... 
arXiv:1909.12120v1 fatcat:qb66tzpx5rf2lifgcftbxskepy

Hybrid Neural Coded Modulation: Design and Training Methods [article]

Sung Hoon Lim, Jiyong Han, Wonjong Noh, Yujae Song, Sang-Woon Jeon
2022 arXiv   pre-print
The inner code is designed using a deep neural network (DNN) which takes the channel coded bits and outputs modulated symbols.  ...  The outer-code can be any standard binary linear code with efficient soft decoding capability (e.g. low-density parity-check (LDPC) codes).  ...  In particular, end-to-end learning for designing encoders and decoders have been proposed in [1] that utilizes a deep neural network (DNN) autoencoder.  ... 
arXiv:2202.01972v1 fatcat:gfq4mntji5dall5xpulrwzsdou

Deep Learning for Wireless Communications [article]

Tugba Erpek, Timothy J. O'Shea, Yalin E. Sagduyu, Yi Shi, T. Charles Clancy
2020 arXiv   pre-print
Next, we present the benefits of deep learning in spectrum situation awareness ranging from channel modeling and estimation to signal detection and classification tasks.  ...  These applications demonstrate the power of deep learning in providing novel means to design, optimize, adapt, and secure wireless communications.  ...  such as modulation and coding, and a decoder for the receiver functionalities such as demodulation and decoding.  ... 
arXiv:2005.06068v1 fatcat:6lklqhjyxjdu5bkq4jpzr3cfd4

Neural Sequence-to-grid Module for Learning Symbolic Rules [article]

Segwang Kim, Hyoungwook Nam, Joonyoung Kim, Kyomin Jung
2021 arXiv   pre-print
Logical reasoning tasks over symbols, such as learning arithmetic operations and computer program evaluations, have become challenges to deep learning.  ...  In particular, even state-of-the-art neural networks fail to achieve out-of-distribution (OOD) generalization of symbolic reasoning tasks, whereas humans can easily extend learned symbolic rules.  ...  Acknowledgements The authors appreciate Hyunkyung Bae for assistance with experiments. K. Jung is with ASRI and ECE, Seoul National University, Korea.  ... 
arXiv:2101.04921v2 fatcat:wct7xacsunaddl47j2tljt6w6a

Configuration Learning in Underwater Optical Links [article]

Xueyuan Zhao, Zhuoran Qi, Dario Pompili
2020 arXiv   pre-print
The proposed configuration learning framework can be further investigated and applied to a broad range of topics in signal processing and communications.  ...  Performance results indicate that the proposal outperforms the competing algorithms for binary and multi-class configuration learning in underwater optical communication datasets.  ...  On the other hand, this conjugate-symmetric OFDM modulation carries the same symbols in both sides of the baseband, therefore the receiver signal recovery SNR can be improved by combining the symbols and  ... 
arXiv:2008.01221v1 fatcat:eijpod7dsve2lgmwncq2z5nq7i
« Previous Showing results 1 — 15 out of 5,683 results