论文部分内容阅读
针对传统固定步长LMS算法预失真收敛速度较慢和非线性改善不足以及RLS算法预失真复杂度较高,对一种基于变步长LMS算法的多项式预失真技术进行了研究。通过仿真分析该算法的性能,发现该算法在不较大增加算法复杂度情况下收敛速度较快,能较好地改善带内非线性失真。构建了基于DSP和FPGA的短波发射机硬件平台以验证算法的有效性,通过双音信号测试功放输出,表明基于该算法的多项式预失真系统对三阶互调降低23.3 dB,五阶互调降低15 dB,有效地改善了功放的非线性。
Aiming at the problem that the traditional fixed-step LMS algorithm has the disadvantages of slow convergence and non-linear improvement of pre-distortion, and high complexity of pre-distortion of RLS algorithm, a polynomial predistortion technique based on variable-length LMS algorithm is studied. Through the simulation analysis of the performance of the algorithm, it is found that the algorithm converges faster without increasing the complexity of the algorithm, and can better improve the in-band nonlinear distortion. The shortwave transmitter hardware platform based on DSP and FPGA was constructed to verify the validity of the algorithm. The output of the amplifier was tested by two-tone signal. The polynomial predistortion system based on the proposed algorithm reduced the third-order intermodulation by 23.3 dB and decreased the fifth-order intermodulation 15 dB, effectively improving the amplifier’s nonlinearity.