Abstract
This paper proposes an efficient neural-network-based digital predistortion (DPD), named as envelope time-delay neural network (ETDNN) DPD. The method complies with the physical characteristics of radio-frequency (RF) power amplifiers (PAs) and uses a more compact DPD model than the conventional neural-network-based DPD. Additionally, a structured pruning technique is presented and used to reduce the computational complexity. It is shown that the resulting ETDNN obtained after applying pruning becomes so sparse that its complexity is comparable to that of conventional DPDs such as memory polynomial(MP) and generalized memory polynomial (GMP), while the degradation in performance due to the pruning is negligible. In an experiment on a 3.5-GHz GaN Doherty power amplifier (PA), our method with the proposed pruning had only one-thirtieth the computational complexity of the conventional neural-network-based DPD for the same adjacent channel leakage ratio (ACLR). Moreover, compared with memory-polynomial-based digital predistortion, our method with the proposed pruning achieved a 7-dB improvement in ACLR, despite it having lower computational complexity.
Highlights
To cope with dramatic increases in data traffic, wireless systems are becoming complex and power hungry
The 3.5-GHz signal was fed to a 3.5-GHz-band gallium nitride (GaN) power amplifiers (PAs), which consisted of a driver amplifier and Doherty PA with a 63 dB total gain and 50 W average output power
The PA output was down-converted from 3.5 GHz to 230 MHz, and this down converted signal was digitalized with an ADC (M9202A, Keysight)
Summary
To cope with dramatic increases in data traffic, wireless systems are becoming complex and power hungry. [9] uses only the phase information in the phase filter, while [8] uses the complex-valued input signals (including the phase and the envelopes’ information) and complex-valued multiplication is used to recover the phase information Both methods have the good characteristic of satisfying the constraints of the physical modeling, which could make for a more compact structure than the I/Q separation type. It was reported in [9] that the vector-decomposition-based time-delay neural network (VDTDNN) [9] outperforms RVTDNN [4]. B. DECOMPOSITION OF ENVELOPE TIME-DELAY NEURAL NETWORK FOR REAL-VALUED CALCULATION ETDNN (1) has complex-valued parameters w(j,2m) , b(m2) and x(k − m) which are densely located around the phase filter. All parameters in (2) can be treated as real values, which in turn enables real-valued optimization
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
