研究目的
To propose a novel 2-D square-root-based memory polynomial behavioral model for concurrent dual-band digital predistortion to reduce the number of coefficients and computational complexity while maintaining or improving linearization accuracy.
研究成果
The proposed 2D-SRBMP model significantly reduces the number of coefficients and computational complexity while achieving better or similar linearization performance compared to existing models. It offers improved NMSE and ACPR with fewer FLOPs and shorter running times, making it suitable for efficient dual-band DPD applications.
研究不足
The model may have limitations in handling higher-order nonlinearities or very deep memory depths beyond the tested ranges. The experimental setup is specific to the used signals and PA, and generalization to other systems might require further validation.
1：Experimental Design and Method Selection:
The study involves designing and testing a new behavioral model for digital predistortion in dual-band systems. The proposed model uses square-root-based basic functions to reduce complexity. Performance is compared with existing models (2D-DPD, 2D-SOC, 2D-MMP) using metrics like NMSE and ACPR.
2：Sample Selection and Data Sources:
Two types of signals are used: a dual-band 20-MHz single-carrier long-term evolution signal with a PAPR of 7.5 dB and a dual-band 20-MHz two-carrier wideband code-division multiple access signal with a PAPR of 8.1 dB, centered at 2.0 and 2.2 GHz.
3：5 dB and a dual-band 20-MHz two-carrier wideband code-division multiple access signal with a PAPR of 1 dB, centered at 0 and 2 GHz.
List of Experimental Equipment and Materials:
3. List of Experimental Equipment and Materials: Vector signal generator (VSG, Agilent MXG N5182A), power amplifier under test, vector signal analyzer (VSA, Agilent PXA N9030A), and a computer for data processing.
4：Experimental Procedures and Operational Workflow:
The test bench setup includes generating signals with the VSG, amplifying them through the PA, and analyzing the output with the VSA. DPD models are applied, and performance is evaluated based on NMSE and ACPR measurements.
5：Data Analysis Methods:
Data is analyzed using normalized mean-square error (NMSE) and adjacent channel power ratio (ACPR) as evaluation criteria. The number of coefficients and floating point operations (FLOPs) are also compared.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
