摘要
盲均衡可以看作模式分类问题,每一类由信源字符表的可能输出定义。由于支持向量回归机具有优良的泛化性能,提出了一种基于v-支持向量回归机的盲均衡算法,并且利用加权最小二乘方法求解v-支持向量回归机。计算机仿真结果表明提出的盲均衡算法具有计算复杂度低、适于实际应用的特点。
The blind equalization can be seen as a pattern recognition problem, where each class is defined by the possible outcomes of the symbol alphabet. A new blind equalization algorithm using a v-support vector regressor (v-SVR) was developed because of the good generalization properties the SVR presents. A weighted least square procedure was proposed for attaining the v-SVR solutions. The performance of the proposed algorithm was evaluated via simulations and shown to have low computation complexity and implement easily.
出处
《系统仿真学报》
CAS
CSCD
北大核心
2009年第10期2921-2924,共4页
Journal of System Simulation
基金
国家部级基金资助项目
关键词
支持向量回归机
盲均衡
加权最小二乘
泛化性能
support vector regressor
blind equalization
weighted least square
generalization properties