Low Complexity Joint Impairment Mitigation of I/Q Modulator and PA Using Neural Networks
Journal article, 2021

neural networks (NNs) for multiple hardware impairments mitigation of a realistic direct conversion transmitter are impractical due to high computational complexity. We propose two methods to reduce the complexity without significant performance penalty. First, propose a novel NN with shortcut connections, referred to as shortcut real-valued time-delay neural network (SVDEN), where trainable neuron-wise shortcut connections are added between the input and output layers. Second, we implement a NN pruning algorithm that gradually removes connections corresponding to minimal weight magnitudes in each layer. Simulation and experimental results show that SVDEN with pruning achieves better performance for compensating frequency-dependent quadrature imbalance and power amplifier nonlinearity than other NN-based and Volterra-based models, while requiring less or similar complexity.

hardware impairment mitigation

in-phase (I) and quadrature (Q) imbalance

Neural networks (NNs)

digital predistortion (DPD)

power amplifier

Author

Yibo Wu

Chalmers, Electrical Engineering, Communication, Antennas and Optical Networks, Communication Systems

Ericsson

Ulf Gustavsson

Ericsson

Alexandre Graell I Amat

Chalmers, Electrical Engineering, Communication, Antennas and Optical Networks, Communication Systems

Henk Wymeersch

Chalmers, Electrical Engineering, Communication, Antennas and Optical Networks, Communication Systems

IEEE Journal on Selected Areas in Communications

0733-8716 (ISSN)

Vol. 40 1 54-64

Deep RF

Swedish Foundation for Strategic Research (SSF) (DnrID19-0021), 2020-01-01 -- 2024-12-31.

Ericsson, 2020-01-01 -- 2024-12-31.

Areas of Advance

Information and Communication Technology

Subject Categories

Telecommunications

Communication Systems

Signal Processing

DOI

10.1109/JSAC.2021.3126024

More information

Latest update

1/20/2022