Abstract
This paper presents two novel neural network models for radio-frequency (RF) power amplifiers (PAs): vector decomposed time-delay neural network (VDTDNN) model and augmented vector decomposed time-delay neural network (AVDTDNN) model. In contrast to conventional neural network-based models, VDTDNN and AVDTDNN comply with the physical characteristics of RF PAs by employing carefully designed network structures. In particular, the nonlinear operations are conducted only on the magnitude of the input signals, while the phase information is recovered with the linear weighting. Linear terms with shortcut connection, as well as high-order terms, can be used to further boost the modeling performance. The complexity analysis shows that the proposed models have significantly lower complexity than the existing neural network models. A wideband GaN RF PA excited by the 40- and 60-MHz OFDM signals were employed to evaluate the performance. The extensive experimental results reveal that the proposed VDTDNN and AVDTDNN models can achieve better linearization performance with lower computational complexity compared with the existing neural network-based models.
Highlights
Driven by the consumer demand and wireless communication technology evolution, the transmitted signals in modern communication systems have the tendencies towards higher peak-to-average power ratios (PAPR) and wider bandwidth [1], [2], which deteriorates the bit error rate (BER) and intensifys the adjacent channel interference effects, due to the inherent nonlinear characteristics of radio frequency (RF) power amplifier (PA)
In this work, we propose two new neural network models: vector decomposed time-delay neural network (VDTDNN) model and augmented vector decomposed time-delay neural network (AVDTDNN) model and they conform more with the nonlinear physical mechanisms of RF PAs, where only the envelops of the input signal are conducted nonlinear operations while the phase information is recovered with linear weighting operations
COMPUTATIONAL COMPLEXITY COMPARISON ANALYSIS In this part, the complexity comparison analysis reveals that whether envelop-dependent terms are augmented into the input vector or not, the computational complexity of the proposed VDTDNN and AVDTDNN are superior to the existing in-phase and quadrature (IQ)-mapping-IQ based real-valued time-delay neural network (RVTDNN) and augmented real-valued time-delay neural network (ARVTDNN), respectively
Summary
Driven by the consumer demand and wireless communication technology evolution, the transmitted signals in modern communication systems have the tendencies towards higher peak-to-average power ratios (PAPR) and wider bandwidth [1], [2], which deteriorates the bit error rate (BER) and intensifys the adjacent channel interference effects, due to the inherent nonlinear characteristics of radio frequency (RF) power amplifier (PA). In this work, we propose two new neural network models: vector decomposed time-delay neural network (VDTDNN) model and augmented vector decomposed time-delay neural network (AVDTDNN) model and they conform more with the nonlinear physical mechanisms of RF PAs, where only the envelops of the input signal are conducted nonlinear operations while the phase information is recovered with linear weighting operations. Both theoretical analysis and experimental results reveal that the performance and computational complexity of the proposed models are superior to the existing neural network DPD models. Two other constraints should be satisfied: (i) the designed models should meet the ‘‘first-zone constraint’’, where odd-parity and unitary phase constraints must be satisfied [23], [24]. (ii) the models can handle complex-valued signals avoiding complex gradient operations
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.
