Journal of Radio Electronics. eISSN 1684-1719. 2025. ¹12

Contents

Full text in Russian (pdf)

Russian page

 

 

DOI: https://doi.org/10.30898/1684-1719.2025.12.6

 

 

 

sparse-learning methods

for bidirectional lstm model

of digital predistorter

 

L.I. Averina, N.E. Guterman, A.A. Khaidarov

 

Voronezh State University, 394018, Voronezh, Universitetskaya sq., 1

 

The paper was received November 17, 2025.

 

Abstract. Two sparse-learning methods for recurrent neural network used as a digital predistorter for microwave power amplifier linearization are presented. Comparative analysis of investigated approaches with unstructured pruning is carried out. Described methods make it possible to build a sparse digital predistorter without using additional hyperparameters and fine-tuning. Sparse neural network, in turn, is a compact implementation of digital predistorter with reduced computational complexity. Such attractive feature allows effective digital predistortion in the transmitter path of both user equipment and multichannel base stations.

Key words: digital predistorter, nonlinear power amplifier, recurrent neural network, sparse variational dropout, regularization.

Financing: The research was carried out at the expense of grant from Russian Science Foundation ¹ 24-19-00891: https://rscf.ru/project/24-19-00891/.

Corresponding author: Guterman Nickita Evgen'evich, n.guterman@internet.ru

 

 

References

1. Nazarov L.E., Zudilin A.S. Influence of power amplifier nonlinearities on OFDM signals // Journal of Radio Electronics. – 2010. – ¹. 12. – Ñ. 5-5.

2. Averina L.I., Shutov V.D., Rybalkin R.A. Structureless Modeling of Power Amplifiers Accounting for Inertial Properties // Radioelectronics and Communications Systems. – 2013. – Ò. 56. – ¹. 1. – Ñ. 50-57.

3. Ghoniem A.A. et al. An error vector magnitude performance modeling and analysis methodology for 5G mm-Wave Transmitters // IEEE Microwave and Wireless Technology Letters. – 2023. – Ò. 33. – ¹. 6. – Ñ. 771-774.

4. Haider M.F. et al. Predistortion-based linearization for 5G and beyond millimeter-wave transceiver systems: A comprehensive survey // IEEE Communications Surveys & Tutorials. – 2022. – Ò. 24. – ¹. 4. – Ñ. 2029-2072.

5. Averina L.I., Guterman N.E. Comparative Analysis of Learning Architectures for Digital Predistortion in Wideband Power Amplifier Linearization Task // Radiotekhnika. – 2025. – ¹. 9. – Ñ. 45-53.

6. Chen P. et al. Behavioral modeling of GaN power amplifiers using long short-term memory networks // 2018 International Workshop on Integrated Nonlinear Microwave and Millimeter-wave Circuits (INMMIC). – IEEE, 2018.  – Ñ. 1-3.

7. Averina L.I., Bugrov O.V. Digital Predistorters Based on Neural Networks for Linearization of Power Amplifiers // Proceedings of Voronezh State University. Series: Physics. Mathematics. – 2017. – ¹. 1. – Ñ. 5-14.

8. Xu G. et al. GRU-Attention Model for Linearizing Millimeter-Wave Transmitters in Vehicle to Satellite Communication Systems // IEEE Transactions on Vehicular Technology. – 2024. – Ò. 73. – ¹. 11. – Ñ. 15992-16000.

9. He Z., Tong F. Residual RNN models with pruning for digital predistortion of RF power amplifiers // IEEE Transactions on Vehicular Technology. – 2022. – Ò. 71. – ¹. 9. – Ñ. 9735-9750.

10. Tanio M., Ishii N., Kamiya N. A sparse neural network-based power adaptive DPD design and its hardware implementation // IEEE Access. – 2022. – Ò. 10. – Ñ. 114673-114682.

11. Watanabe T., Ohseki T., Kanno I. Hardware-efficient Neural Network Digital Predistortion for Terahertz Power Amplifiers Using DeepShift and Pruning // IEEE Access. – 2025.

12. Li W. et al. GPU-Based Implementation of Pruned Artificial Neural Networks for Digital Predistortion Linearization of Wideband Power Amplifiers // IEEE Journal of Microwaves. – 2025. – Ò. 5. – ¹. 3. – Ñ. 726-738.

13. Liu Z. et al. Low computational complexity digital predistortion based on convolutional neural network for wideband power amplifiers // IEEE Transactions on Circuits and Systems II: Express Briefs. – 2021. – Ò. 69. – ¹. 3. – Ñ. 1702-1706.

14. Molchanov D., Ashukha A., Vetrov D. Variational dropout sparsifies deep neural networks // International conference on machine learning. – PMLR, 2017. – Ñ. 2498-2507.

15. Kingma D.P., Salimans T., Welling M. Variational dropout and the local reparameterization trick // Advances in neural information processing systems. – 2015. – Ò. 28.

16. Louizos C., Welling M., Kingma D.P. Learning sparse neural networks through L0 regularization // arXiv preprint arXiv:1712.01312. – 2017.

17. Sun J. et al. Behavioral modeling and linearization of wideband RF power amplifiers using BiLSTM networks for 5G wireless systems // IEEE Transactions on Vehicular Technology. – 2019. – Ò. 68. – ¹. 11. – Ñ. 10348-10356.

18. Lobacheva E. et al. Structured sparsification of gated recurrent neural networks // Proceedings of the AAAI Conference on Artificial Intelligence. – 2020. – Ò. 34. – ¹. 04. – Ñ. 4989-4996.

19. Lobacheva E., Chirkova N., Vetrov D. Bayesian sparsification of recurrent neural networks // arXiv preprint arXiv:1708.00077. – 2017.

20. Wen L. et al. Structured pruning of recurrent neural networks through neuron selection // Neural Networks. – 2020. – Ò. 123. – Ñ. 134-141.

21. Ngoya E., Quindroit C., Nebus J.M. On the continuous-time model for nonlinear-memory modeling of RF power amplifiers // IEEE Transactions on Microwave Theory and Techniques. – 2009. – Ò. 57. – ¹. 12. – Ñ. 3278-3292.

22. Kingma D.P., Welling M. Auto-encoding variational bayes // arXiv preprint arXiv:1312.6114. – 2013.

23. Chirkova N., Lobacheva E., Vetrov D. Bayesian compression for natural language processing // arXiv preprint arXiv:1810.10927.  – 2018.

24. Louizos C., Ullrich K., Welling M. Bayesian compression for deep learning // Advances in neural information processing systems.  – 2017.  – Ò. 30.

25. Morgan D.R. et al. A generalized memory polynomial model for digital predistortion of RF power amplifiers // IEEE Transactions on signal processing.  – 2006. – Ò. 54. – ¹. 10. – Ñ. 3852-3860.

For citation:

Averina L.I., Guterman N.E., Khaidarov A.A. Sparse-learning methods for bidirectional LSTM model of digital predistorter // Journal of Radio Electronics. – 2025. – ¹. 12. https://doi.org/10.30898/1684-1719.2025.12.6 (In Russian)