数字预失真器模型参数辨识算法研究

数字基带预失真技术是补偿功放非线性的最有效方法之一,而自适应算法的选择很大程度上影响着系统的预失真性能。其中,最小均方误差(LMS)算法和递归最小二乘法(RLS)是最常用的自适应收敛算法,在综合分析两种算法优缺点的基础上,提出了一种基于正弦函数的归一化LMS(NLMS)算法和RLS算法的组合算法。仿真结果表明,该组合算法在达到与RLS同样预失真精度的前提下,可大大减小计算量。

Parameters Identification Algorithm of Digital Pre-Distorter Model

Digital baseband pre-distortion is one of the most efficienct ways to compensate nonlinearity of the power amplifier, and the choice of adaptive algorithm directly affects,to a large extent,the pre-distortion performance of the system. LMS and RLS are the most frequently-used adaptive convergence algorithms,and based on the analysis of their advantages and disadvantages, a novel algorithm combined with sine function-based normalized LMS and RLS is proposed. Simulation results indicate that this combined algorithm could enormously lower computational complexity while maintaining the same pre-distortion precision as RLS.