摘要
提出一种基于高阶累积量和支撑矢量机的数字信号自动调制识别新方法 ,即将接收信号的四阶、六阶累积量作为分类特征向量 ,利用支持矢量机把分类特征向量映射到一个高维空间 ,并在高维空间中构造最优分类超平面以实现信号分类。这种方法对高斯噪声和星座图由于信号初始相位而引入的旋转具有良好的稳健性 ,并避免了神经网络中的过学习和局部极小点等缺陷。计算仿真结果表明 。
This paper presents a new method for modulation and recognition of digital communication signals based on higher order cumulants (HOC) and support vector machines (SVM). The fourth and sixth order cumulants of the received signal are used as the classification vectors. SVM maps input vectors nonlinearly into a high dimensional feature space and constructs the optimum separating hyperplane in space to realize signal recognition. This method is robust to Gaussian noise and constellation rotation due to initial phase of signal and avoids overfitting and local minimum in neural networks. The high performance and robustness of the algorithm are proved by computer simulation.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2003年第8期1007-1011,共5页
Systems Engineering and Electronics
基金
国家"8 63"高技术计划重大项目 (2 0 0 1AA12 3 0 3 1)
高等学校优秀青年教师教学科研奖励计划
教育部科学技术研究重点项目资助课题
关键词
调制识别
高阶累积量
支撑矢量机
Modulation and recognition
Higher order cumulants
Support vector machines
