MP-DPD: Low-Complexity Mixed-Precision Neural Networks for Energy-Efficient Digital Predistortion of Wideband Power Amplifiers

IEEE Microwave and Wireless Technology Letters, vol. 34, pp. 817–820

Abstract

Digital predistortion (DPD) enhances signal quality in wideband radio frequency (RF) power amplifiers (PAs). As signal bandwidths expand in modern radio systems, DPD’s energy consumption increasingly impacts overall system efficiency. Deep neural networks (DNNs) offer promising advancements in DPD, yet their high complexity hinders their practical deployment. This article introduces open-source mixed-precision (MP) neural networks that employ quantized low-precision fixed-point parameters for energy-efficient DPD. This approach reduces computational complexity and memory footprint, thereby lowering power consumption without compromising linearization efficacy. Applied to a 160-MHz-BW 1024-QAM OFDM signal from a digital RF PA, MP-DPD gives no performance loss against 32-bit floating-point precision DPDs, while achieving$-$43.75 (L)/$-$45.27 (R) dBc in the adjacent channel power ratio (ACPR) and$-$38.72 dB in error vector magnitude (EVM). A 16-bit fixed-point-precision MP-DPD enables a 2.8$\times$reduction in estimated inference power. The DPD code in PyTorch is publicly available on GitHub.

Authors 8

1. Yizhuo Wu

   Delft University of Technology

   Affiliation as printed

   Department of Microelectronics, Delft University of Technology, Delft, The Netherlands

2. Ang Li

   Delft University of Technology

   Affiliation as printed

   Department of Microelectronics, Delft University of Technology, Delft, The Netherlands

3. Mohammadreza Beikmirza

   Delft University of Technology

   Affiliation as printed

   Department of Microelectronics, Delft University of Technology, Delft, The Netherlands

4. Gagan Deep Singh

   Delft University of Technology

   Affiliation as printed

   Department of Microelectronics, Delft University of Technology, Delft, The Netherlands

5. Qinyu Chen Aachen Leiden Institute of Advanced Computer Science

   Leiden University

   Affiliation as printed

   Leiden Institute of Advanced Computer Science (LIACS), Leiden University, Leiden, The Netherlands

6. Leo C. N. de Vreede

   Delft University of Technology

   Affiliation as printed

   Department of Microelectronics, Delft University of Technology, Delft, The Netherlands

7. Morteza S. Alavi

   Delft University of Technology

   Affiliation as printed

   Department of Microelectronics, Delft University of Technology, Delft, The Netherlands

8. Chang Gao

   Delft University of Technology

   Affiliation as printed

   Department of Microelectronics, Delft University of Technology, Delft, The Netherlands