基于二阶锥规划的自适应均衡算法  

Adaptive equalization algorithm based on second-order cone programming

作　　者：陈阳 张艺朦 赵安邦 余赟 冯烨旻[2] 

机构地区：[1]哈尔滨工程大学水声技术重点实验室,黑龙江哈尔滨150001 [2]中兴通讯股份有限公司,江苏南京210012

出　　处：《系统工程与电子技术》2011年第4期904-907,共4页Systems Engineering and Electronics

基　　金：国家自然科学基金(60802060);水声技术重点实验室基金(9140C200201080C20);中央高校基本科研业务费专项资金(F030102);海洋公益性行业科研专项经费专项(gz2010001)资助课题

摘　　要：提出一种改进的均衡器算法。该方法基于最小均方误差(minimum mean square error,MMSE)准则,使均衡器的输出与训练码的均方误差最小,并且将信道均衡的最小均方误差目标函数转化为二阶锥形式,利用内点法求最优解。与传统基于最小均方误差(least mean squares,LMS)和递归最小二乘(recursive leastsquares,RLS)自适应算法的均衡器相比,由于不需要迭代收敛过程,不存在收敛速度与精度的矛盾,克服了基于LMS和RLS的自适应均衡器参数设置的困难,而且利用更短的训练序列长度即可获得相同的均衡效果,对于改善通信效率具有参考价值。An improved equalization is proposed.Based on the principle of minimum mean square error（MMSE）,the algorithm minimums the mean square error between the output of the equalization and the training codes.The objective function is transformed to the second-order cone programming（SOC）,which is solved by the standard interior point method.Comparing with traditional adaptive algorithms equalization based on least mean squares（LMS） and recursive least squares（RLS）,the algorithm which does not need iterative process is not affected by the contradiction between convergent speed and precision.Therefore it overcomes the difficulty of parameters setting in adaptive equalization.Furthermore,the algorithm only needs shorter training codes to achieve the same equilibrium effect and is a guide to improve the communication efficiency.

关 键 词：自适应均衡 二阶锥规划 最小均方误差 递归最小二乘 

分 类 号：TN914[电子电信—通信与信息系统]