基于记忆非线性多项式的星地一体预失真算法  

Pre-distortion Algorithm Based on Memory Polynomial for Integrated Satellite and Ground-station HPA

作　　者：杨茂强[1] 鲁璐[1] 郭道省[2] 潘小飞[2] YANG Mao-qiang;LU Lu;GUO Dao-xing;PAN Xiao-fei(Postgraduate Team 3 CCE,PLAUST,Nanjing 210007,China;College of Communications Engineering,PLAUST)

机构地区：[1]解放军理工大学通信工程学院研究生3队,江苏南京210007 [2]解放军理工大学通信工程学院

出　　处：《军事通信技术》2013年第1期1-6,共6页Journal of Military Communications Technology

基　　金：国防预研基金资助项目

摘　　要：传统的基于间接学习结构的预失真技术通常仅考虑对单个地面功放或卫星功放进行补偿,线性化性能有限;文中联合地面功放与卫星功放的非线性失真和卫星信道的实际传输特性,提出了一种适合透明卫星功放的星地一体间接学习预失真算法。该算法利用记忆深度为5、非线性阶数为9的奇偶项多项式作为预失真器,同时采用最小均方(LMS)与递归最小二乘(RLS)联合的算法自适应更新预失真器的系数,以兼顾运算量和收敛速度。仿真结果表明,经过星地一体预失真后,卫星高功放输出端APSK星座图误差矢量幅度改善值达到91.52%,带外功率谱抑制平均提升了11.41dB,系统线性化性能非常理想。The pre-distortion techniques based on the traditional Indirect Learning Architecture(IDLA)are commonly used to linearize High Power Amplifiers(HPA)settled in ground-station or on board satellite merely,which canr t achieve a satisfying linearization.In this paper,an improved IDLA,which takes into considertation the satellite channel features with two HPAs in the loop,was put forward for bent-pipe satellite HPA.Nonlinear polynomial with 9 order and 5 memory depth was utilized as pre-distorter and Recursive Least Square(RLS)in conjunction with the Least Mean Square(LMS)adaptive algorithm adopted for its coefficients updating to make a tradeoff between the convergence speed and the computational complex.Simulation results demonstrate that with pre-distortion the Error Vector Magnitude(EVM)at the output of satellite HPA of the signal constellation is greatly improved up to 91.52%,that the out-band Power Spectrum Density(PSD)of signal achieves 11.41 dB constraint in average,and that the desirable system linearization performance is acquired.

关 键 词：卫星通信 非线性失真 预失真 记忆非线性多项式 线性化 

分 类 号：TN911[电子电信—通信与信息系统]