The success of deep learning has renewed interest in applying neural networks and other machine learning techniques to most fields of data and signal processing, including communications. Advances in architecture and training lead us to consider new modem architectures that allow flexibility in design, continued learning in the field, and improved waveform coding. This dissertation examines neural network architectures and training methods suitable for demodulation in power-limited communication systems, such as those found in wireless sensor networks. Such networks will provide greater connection to the world around us and are expected to contain orders of magnitude more devices than cellular networks. A number of standard and proprietary protocols span this space, with modulations such as frequency-shift-keying (FSK), Gaussian FSK (GFSK), minimum shift keying (MSK), on-off-keying (OOK), and M-ary orthogonal modulation (M-orth). These modulations enable low-cost radio hardware with efficient nonlinear amplification in the transmitter and noncoherent demodulation in the receiver.
The dissertation proposes a novel complex neural network demodulator architecture suitable for such systems. Mathematical derivation of the backpropagation equations for the proposed architecture provides insights into its applicability to either coherent or noncoherent demodulation, and experimental results demonstrate its performance and suitability for a variety of modulation formats. When trained in nominal AWGN channel conditions, the proposed architecture learns comparable, and in some cases better, performance than traditional demodulators. It is also flexible enough to learn improved tolerance to radio or channel impairments that are difficult or intractable to include in mathematical derivations. When combined with a trainable modulator, the proposed demodulator can also learn new, spectrally efficient waveform coding tailored to a specific channel. An added benefit of the proposed architecture is continued learning in the field, and we examine the challenges of incremental learning and identify areas of future research. It is hoped that this work will lead to a future common demodulator architecture that can support the wide range of modulation formats in this space and realize the additional benefits offered by machine learning.
|Commitee:||Zhuang, Hanqi, Aalo, Valentine, Zhu, Xingquan|
|School:||Florida Atlantic University|
|School Location:||United States -- Florida|
|Source:||DAI-B 82/2(E), Dissertation Abstracts International|
|Subjects:||Electrical engineering, Operations research, Computer science|
|Keywords:||Communication theory, Complex-valued, Demodulation, Machine learning, Neural networks, Signal processing|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be