Learning to Detect

@article{Samuel2019LearningTD,
  title={Learning to Detect},
  author={Neev Samuel and Tzvi Diskin and Ami Wiesel},
  journal={IEEE Transactions on Signal Processing},
  year={2019},
  volume={67},
  pages={2554-2564}
}
In this paper, we consider multiple-input-multiple-output detection using deep neural networks. We introduce two different deep architectures: a standard fully connected multi-layer network, and a detection network (DetNet), which is specifically designed for the task. The structure of DetNet is obtained by unfolding the iterations of a projected gradient descent algorithm into a network. We compare the accuracy and runtime complexity of the proposed approaches and achieve state-of-the-art… 
Massive MIMO Data Detection Using 1-dimensional Convolutional Neural Network
TLDR
This work explores the use of an adaptive one-dimensional convolutional neural network (1d-CNN) for the massive multiple-input multiple-output (MIMO) data detection and employs a data augmentation approach based on the existing computationally cheaper detectors.
Deep Signal Recovery with One-bit Quantization
TLDR
A model-based machine learning method is proposed and unfold the iterations of an inference optimization algorithm into the layers of a deep neural network for one-bit signal recovery, which can efficiently handle the recovery of high-dimensional signals from acquired one- bit noisy measurements.
MetaSSD: Meta-Learned Self-Supervised Detection
TLDR
A meta-learning-based self-supervised symbol detector named MetaSSD, which helps the model adapt to a new channel environment based on experience with various meta-training environments, and help the model to use relatively less supervision than the previously suggested learning-based detectors.
VPNet: Variable Projection Networks
TLDR
VPNet, a novel model-driven neural network architecture based on variable projection (VP), offers fast learning ability and good accuracy at a low computational cost of both training and inference and is anticipated to have a profound impact on the broader field of signal processing.
Online Meta-Learning For Hybrid Model-Based Deep Receivers
TLDR
A data-efficient two-stage training method that facilitates rapid online adaptation that allows receivers to outperform previous techniques based on self-supervision and joint-learning by a margin of up to 2.5 dB in coded bit error rate in rapidly-varying scenarios.
Deep MIMO detection
  • N. Samuel, Tzvi Diskin, A. Wiesel
  • Computer Science
    2017 IEEE 18th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC)
  • 2017
TLDR
The results show that deep networks can achieve state of the art accuracy with significantly lower complexity while providing robustness against ill conditioned channels and mis-specified noise variance.
LoRD-Net: Unfolded Deep Detection Network With Low-Resolution Receivers
TLDR
Numerically evaluate the proposed receiver architecture for one-bit signal recovery in wireless communications and demonstrate that the proposed hybrid methodology outperforms both data-driven and model-based state-of-the-art methods, while utilizing small datasets, on the order of merely $\sim 500$ samples, for training.
CNN-Based Learning System in a Generalized Fading Environment
In this paper, we investigate the Block Error Rate (BLER) performance of a Convolutional Neural Network (CNN)based autoencoder as a self-learning communication system under a generalized fading
CMDNet: Learning a Probabilistic Relaxation of Discrete Variables for Soft Detection With Low Complexity
TLDR
Numerical simulation results in MIMO systems reveal CMD net to feature a promising accuracy complexity trade-off compared to State of the Art, and it is demonstrated CMDNet’s soft outputs to be reliable for decoders.
Complexity-Scalable Neural-Network-Based MIMO Detection With Learnable Weight Scaling
TLDR
This paper introduces a framework for systematic complexity scaling of deep neural network (DNN) based MIMO detectors that uses a fraction of the DNN inputs by scaling their values through weights that follow monotonically non-increasing functions to achieve accuracy-vs-complexity scalability during inference.
...
...

References

SHOWING 1-10 OF 42 REFERENCES
Going deeper with convolutions
We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition
An Introduction to Deep Learning for the Physical Layer
  • Tim O'Shea, J. Hoydis
  • Computer Science
    IEEE Transactions on Cognitive Communications and Networking
  • 2017
TLDR
A fundamental new way to think about communications system design as an end-to-end reconstruction task that seeks to jointly optimize transmitter and receiver components in a single process is developed.
Neural Network Detection of Data Sequences in Communication Systems
TLDR
This work considers detection based on deep learning, and shows it is possible to train detectors that perform well without any knowledge of the underlying channel models, and demonstrates that the bit error rate performance of the proposed SBRNN detector is better than that of a Viterbi detector with imperfect CSI.
Deep robust regression
TLDR
This work builds upon Huber's robust regression and the classical least trimmed squares estimator, and proposes a deep neural network that generalizes both and provides high accuracy with low computational complexity.
Deep Unfolding: Model-Based Inspiration of Novel Deep Architectures
TLDR
This work starts with a model-based approach and an associated inference algorithm, and folds the inference iterations as layers in a deep network, and shows how this framework allows to interpret conventional networks as mean-field inference in Markov random fields, and to obtain new architectures by instead using belief propagation as the inference algorithm.
Deep MIMO detection
  • N. Samuel, Tzvi Diskin, A. Wiesel
  • Computer Science
    2017 IEEE 18th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC)
  • 2017
TLDR
The results show that deep networks can achieve state of the art accuracy with significantly lower complexity while providing robustness against ill conditioned channels and mis-specified noise variance.
Deep Residual Learning for Image Recognition
TLDR
This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.
An Introduction to Machine Learning Communications Systems
TLDR
By interpreting a communications system as an autoencoder, this work develops a fundamental new way to think about radio communications system design as an end-to-end reconstruction optimization task that seeks to jointly optimize transmitter and receiver components in a single process.
Learning to invert: Signal recovery via Deep Convolutional Networks
  • A. Mousavi, Richard Baraniuk
  • Computer Science
    2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2017
TLDR
A new signal recovery framework is developed that learns the inverse transformation from measurement vectors to signals using a deep convolutional network and closely approximates the solution produced by state-of-the-art CS recovery algorithms yet is hundreds of times faster in run time.
Learning representations by back-propagating errors
TLDR
Back-propagation repeatedly adjusts the weights of the connections in the network so as to minimize a measure of the difference between the actual output vector of the net and the desired output vector, which helps to represent important features of the task domain.
...
...