一种支持向量机的相干解调系统误码率模拟预测算法  

A bit error rate simulation prediction algorithm for coherent demodulation system based on support vector machine

作　　者：孙培刚[1] 李美丽[1] 张全禹[1] SUN Peigang;LI Meili;Zhang Quanyu(College of Electrical Engineering,Suihua College,Suihua 152061,China)

机构地区：[1]绥化学院电气工程学院,黑龙江绥化152061

出　　处：《电子设计工程》2024年第10期74-77,82,共5页Electronic Design Engineering

基　　金：黑龙江省省属高校基本科研业务费科研项目(2017-KYYWF-0720)。

摘　　要：针对信道中高斯加性噪声引起接收端相干解调系统的误码问题,提出了一种支持向量机的相干解调系统误码率模拟预测算法,为提高通信系统的抗噪特性提供研究基础。文中采用差分进化算法与支持向量机SVM算法相结合的方法,通过差分进化算法优化SVM算法中的惩罚因子与高斯径向基的核函数参数,并对模拟环境下相干解调系统受高斯加性噪声影响所产生的误码数据进行了预测。实验结果表明,基于差分进化的SVM算法预测模型相比经典遗传算法预测精度提高了3.7%,预测精度基本满足误码率数据的预测要求,并具有较强的泛化能力。Aiming at the bit error problem of dry demodulation system of receiving terminal caused by Gaussian additive noise in channel,a bit error rate simulation prediction algorithm of coherent demodulation system based on support vector machine is proposed,which provides a research basis for improving the anti⁃noise characteristics of communication system.In this paper,differential evolution algorithm and SVM algorithm are combined to optimize the penalty factor and kernel function parameters of Gaussian radial basis in SVM algorithm by differential evolution algorithm,and the error data of coherent demodulation system under the influence of Gaussian additive noise is predicted.The experimental results show that the prediction accuracy of the SVM algorithm based on differential evolution is 3.7%higher than that of the classical genetic algorithm.The prediction accuracy basically meets the prediction requirements of bit error rate data and has strong generalization ability.

关 键 词：误码率 支持向量机 差分进化算法 缺一个 相干解调系统 

分 类 号：TN99[电子电信—信号与信息处理]