计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (16): 141-143.

• 数据库、信号与信息处理 • 上一篇    下一篇

改进的OPTICS算法在调制识别中的应用

王 品,黄 焱   

  1. 解放军信息工程大学 信息工程学院,郑州 450002
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-06-01 发布日期:2011-06-01

Application of modified OPTICS algorithm in modulation identification

WANG Pin,HUANG Yan   

  1. Institute of Information Engineering,Information Engineering University of PLA,Zhengzhou 450002,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-06-01 Published:2011-06-01

摘要: 提出了一种基于星座聚类的调制识别新方法。该算法利用改进的OPTICS算法能准确发现数据集分布的特性,不仅克服了模糊C-均值聚类算法对初始聚类中心、样本输入次序敏感等不足,而且能够同时发现不同密度的聚类,结合聚类有效性分析实现了聚类中心目的自适应调整,同基于DBSCAN的聚类算法相比,降低了时间复杂度。将该算法用于对MPSK/MAPSK信号星座重构和识别,实验结果表明该方法是实际有效的。

关键词: 星座图, 调制识别, 通过对象排序识别聚类结构(OPTICS), 聚类有效性分析

Abstract: A novel modulation recognition method is proposed.With the ability of modified OPTICS to find the distribution of database accurately,the presented clustering algorithm overcomes the shortage that the C-means clustering algorithm is sensitive to initial clustering center and sample input order.It can also find out all the meaningful clusters for datasets with varied densities.It determines the number of clusters adaptively via validity analysis.This method is applied to modulation recognition of MPSK/MAPSK signals and experiments show that the recognition method is effective.

Key words: constellation, modulation identification, Ordering Points To Identify the Clustering Structure(OPTICS), cluster-validity analysis