稀疏假设下双频功率放大器的建模与预失真

Modeling and predistortion of dual-frequency power amplifierunder sparse hypothesis

提出一种简单而灵活的功放建模与预失真改进方法。该方法基于Volterra级数核系数的稀疏性假设,结合最大似然估计和最佳模型确定的信息准则,并且在选择活跃系数时结合了贪婪算法。将该技术应用于输入信号中心频率为2.61 GHz的宽带码分多址(wideband code division multiple access,WCDMA)信号和中心频率在2.69 GHz处的长期演进(long term evolution,LTE)信号的双频功率放大器预失真中,将输入信号放到欠采样双频预失真结构中,分别得到2个频带的预失真仿真测试结果。与Volterra级数和记忆多项式模型进行对比,结果表明,双频欠采样方法在预失真时减少了Volterra级数的参数,与传统Volterra预失真器相比,所提出的欠采样双频预失真器满足参考标准的频谱掩模和误差矢量幅度限制,在达到线性化补偿相同效果下,所提模型参数量减少了40%以上,减少了模型的复杂度。

In this paper,a simple and flexible method of power amplifier modeling and predistortion is proposed.Based on the sparsity assumption of the kernel coefficients of Volterra series,this mothod combines the maximum likelihood estimation with the information criterion determined by the best model,and combines the greedy algorithm when selecting the active coefficients.Finally,the technique is applied to the predistortion of WCDMA signal with a center frequency of 2.61 GHz and LTE signal with a center frequency of 2.69 GHz.The input signal is put into the under sampling dual frequency predistortion structure,and the predistortion simulation test results of the two frequency bands are obtained respectively,which are compared with the Volterra series and the memory polynomial model.The results show that the dual frequency Under sampling method reduces the parameters of Volterra Series in predistortion.Compared with the traditional Volterra predistorter,the proposed under sampling dual frequency predistorter meets the spectrum mask and error vector amplitude limit of reference standard.Under the same effect of linearization compensation,the number of parameters of the proposed model is reduced by more than 40%and the complexity of the model is reduced.