Learning to Communicate: Channel Auto-encoders, Domain Specific Regularizers, and Attention

Published: 2016 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)
Open Access: arXiv:1608.06409

Abstract: We address the problem of learning efficient and adaptive ways to communicate binary information over an impaired channel. We treat the problem as reconstruction optimization through impairment layers in a channel autoencoder and introduce several new domain-specific regularizing layers to emulate common channel impairments. We also apply a radio transformer network based attention model on the input of the decoder to help recover canonical signal representations. We demonstrate some promising initial capacity results from this architecture and address several remaining challenges before such a system could become practical.

An Introduction to Deep Learning for the Physical Layer

Published: IEEE Transactions on Cognitive Communications and Networking, Vol. 3 Issue 4, Dec 2017
Open Access: arXiv:1702.00832

Abstract: We present and discuss several novel applications of deep learning for the physical layer. By interpreting a communications system as an autoencoder, we develop 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. We show how this idea can be extended to networks of multiple transmitters and receivers and present the concept of radio transformer networks as a means to incorporate expert domain knowledge in the machine learning model. Lastly, we demonstrate the application of convolutional neural networks on raw IQ samples for modulation classification which achieves competitive accuracy with respect to traditional schemes relying on expert features. The paper is concluded with a discussion of open challenges and areas for future investigation.

Radio Transformer Networks: Attention MOdels for Learning to Synchronize in Wireless Systems

Published: 2016 50th Asilomar Conference on Signals, Systems, and Computers
Open Access: arXiv:1605.00716

Abstract: We introduce learned attention models into the radio machine learning domain for the task of modulation recognition by leveraging spatial transformer networks and introducing new radio domain appropriate transformations. This attention model allows the network to learn a localization network capable of synchronizing and normalizing a radio signal blindly with zero knowledge of the signals structure based on optimization of the network for classification accuracy, sparse representation, and regularization. Using this architecture we are able to outperform our prior results in accuracy vs signal to noise ratio against an identical system without attention, however we believe such an attention model has implication far beyond the task of modulation recognition.

Convolutional Radio Modulation Recognition Networks

Published: 2016 Engineering Applications of Neural Networks
Open Access: arXiv:1605.00716

Abstract: We study the adaptation of convolutional neural networks to the complex temporal radio signal domain. We compare the efficacy of radio modulation classification using naively learned features against using expert features which are widely used in the field today and we show significant performance improvements. We show that blind temporal learning on large and densely encoded time series using deep convolutional neural networks is viable and a strong candidate approach for this task especially at low signal to noise ratio.

Unsupervised Representation Learning of Structured Radio Communication Signals

Published: 2016 First International Workshop on Sensing, Processing and Learning for Intelligent Machines (SPLINE)
Open Access: arXiv:1604.07078

Abstract: We explore unsupervised representation learning of radio communication signals in raw sampled time series representation. We demonstrate that we can learn modulation basis functions using convolutional autoencoders and visually recognize their relationship to the analytic bases used in digital communications. We also propose and evaluate quantitative metrics for quality of encoding using domain relevant performance metrics.

Deep Architectures for Modulation Recognition

Published: 2017 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN)
Open Access: arXiv:1703.09197

Abstract: We survey the latest advances in machine learning with deep neural networks by applying them to the task of radio modulation recognition. Results show that radio modulation recognition is not limited by network depth and further work should focus on improving learned synchronization and equalization. Advances in these areas will likely come from novel architectures designed for these tasks or through novel training methods.

Semi-Supervised Radio Signal Identification

Published: 2017 19th International Conference on Advanced Communication Technology (ICACT)
Open Access: arXiv:1611.00303

Abstract: Radio emitter recognition in dense multi-user environments is an important tool for optimizing spectrum utilization, identifying and minimizing interference, and enforcing spectrum policy. Radio data is readily available and easy to obtain from an antenna, but labeled and curated data is often scarce making supervised learning strategies difficult and time consuming in practice. We demonstrate that semi-supervised learning techniques can be used to scale learning beyond supervised datasets, allowing for discerning and recalling new radio signals by using sparse signal representations based on both unsupervised and supervised methods for nonlinear feature learning and clustering methods.

Radio Machine Learning Dataset Generation with GNU Radio

Published: 2016 Proceedings of the 6th GNU Radio Conference (GRCon)

Abstract: This paper surveys emerging applications of Machine Learning (ML) to the Radio Signal Processing domain.  Provides some brief background on enabling methods and discusses some of the potential advancements for the field.  It discusses the critical importance of good datasets for model learning, testing, and evaluation and introduces several public open source synthetic datasets for various radio machine learning tasks.  These are intended to provide a robust common baselines for those working in the field and to provide a benchmark measure against which many techniques can be rapidly evaluated and compared.

Deep Learning Based MIMO Communications

Published: 2017 55th Annual Allerton Conference on Communication, Control, and Computing
Open Access: arXiv:1707.07980

Abstract: We introduce a novel physical layer scheme for single user Multiple-Input Multiple-Output (MIMO) communications based on unsupervised deep learning using an autoencoder. This method extends prior work on the joint optimization of physical layer representation and encoding and decoding processes as a single end-to-end task by expanding transmitter and receivers to the multi-antenna case. We introduce a widely used domain appropriate wireless channel impairment model (Rayleigh fading channel), into the autoencoder optimization problem in order to directly learn a system which optimizes for it. We considered both spatial diversity and spatial multiplexing techniques in our implementation. Our deep learning-based approach demonstrates significant potential for learning schemes which approach and exceed the performance of the methods which are widely used in existing wireless MIMO systems. We discuss how the proposed scheme can be easily adapted for open-loop and closed-loop operation in spatial diversity and multiplexing modes and extended use with only compact binary channel state information (CSI) as feedback.

Approximating the Void: Learning Stochastic Channel Models from Observation with Variational Generative Adversarial Networks

Published: (Pending)
Open Access: arXiv:1805.06350

Abstract: Channel modeling is a critical topic when considering designing, learning, or evaluating the performance of any communications system. Prior work in manual modulation system design or learned modulation system design has largely focused on simplified analytic channel models such as additive white Gaussian noise (AWGN) or Rayleigh fading channels, in more recent work we consider the usage of generative adversarial networks (GANs) to jointly approximate of a wireless channel response and design an efficient encoding and decoding of information to robustly survive it. In this paper, we focus more specifically on characterizing how well a GAN can capture the stochastic nature of a typical wireless channel response, and the topic of effectively designing the network and loss function to accurately capture its stochastic behavior in a probabilistic sense. We illustrate the problems with certain approaches and share results capturing the performance of such as system over a range channel distributions.

Physical Layer Communications System Design Over-the-Air Using Adversarial Networks

Published: (Pending)
Open Access: arXiv:1803.03145

Abstract: This paper presents a novel method for synthesizing new physical layer modulation and coding schemes for communications systems using a learning-based approach which does not require an analytic model of the impairments in the channel. It extends prior work published on the channel autoencoder to consider the case where the channel response is not known or can not be easily modeled in a closed form analytic expression. By adopting an adversarial approach for channel response approximation and information encoding, we can jointly learn a good solution to both tasks over a wide range of channel environments. We describe the operation of the proposed adversarial system, share results for its training and validation over-the-air, and discuss implications and future work in the area.

Over the Air Deep Learning Based Radio Signal Classification

Published: IEEE Journal of Selected Topics in Signal Processing, Vol. 12, Issue 1, Feb 2018
Open Access: arXiv:1712.04578

Abstract: We conduct an in depth study on the performance of deep learning based radio signal classification for radio communications signals. We consider a rigorous baseline method using higher order moments and strong boosted gradient tree classification and compare performance between the two approaches across a range of configurations and channel impairments. We consider the effects of carrier frequency offset, symbol rate, and multi-path fading in simulation and conduct over-the-air measurement of radio classification performance in the lab using software radios and compare performance and training strategies for both. Finally we conclude with a discussion of remaining problems, and design considerations for using such techniques.