计算机应用 ›› 2012, Vol. 32 ›› Issue (11): 3054-3056.DOI: 10.3724/SP.J.1087.2012.03054

• 人工智能 • 上一篇    下一篇

采用粒子群算法的空时二维参数估计

邱新建1,2,山拜?达拉拜1,薛凤凤3   

  1. 1. 新疆大学 信息科学与工程学院, 乌鲁木齐 830046
    2. 中国人民解放军 68203部队,甘肃 酒泉 735000
    3. 空军工程大学 电讯工程学院, 西安 710077
  • 收稿日期:2012-05-13 修回日期:2012-06-26 发布日期:2012-11-12 出版日期:2012-11-01
  • 通讯作者: 邱新建
  • 作者简介:邱新建(1984-),男,陕西南郑人,博士研究生,主要研究方向:智能信号处理;山拜·达拉拜(1959-),男(哈萨克族),新疆乌鲁木齐人,教授,博士,主要研究方向:阵列信号处理、智能信号处理;薛凤凤(1985-),女,陕西西安人,博士研究生,主要研究方向:智能信号处理。
  • 基金资助:
    国家自然科学基金资助项目(60971130)

Application of particle swarm optimization to spacetime twodimensional parameter estimation

QIU Xin-jian1,2,SHANBAI Dalabaev1,XUE Feng-feng3   

  1. 1. College of Information Science and Engineering, Xinjiang University,Urumqi Xinjiang 830046,China
    2. Unit 68203 of PLA, Jiuquan Gansu 735000, China
    3. Telecommunications Engineering Institute, Air Force Engineering University, Xi’an Shaanxi 710077,China
  • Received:2012-05-13 Revised:2012-06-26 Online:2012-11-12 Published:2012-11-01
  • Contact: QIU Xin-jian

摘要: 针对传统的空时二维参数估计计算复杂、鲁棒性及通用性差、收敛速度慢等缺点,根据空时具有等效性,以空域和时域处理算法可以相互转化为基础,推导出合适的适应度函数,运用改进的粒子群算法同时搜索信号的到达角和频率,用Kmeans聚类算法对搜索结果进行分类,利用粒子群算法计算简单、全局收敛、可并行性等特点提高算法的搜索能力。计算机仿真表明,与传统的方法相比该算法具有较好的统计和收敛性能。

关键词: 空时二维参数估计, 粒子群算法, 谱估计

Abstract: The traditional spacetime twodimensional parameter estimation has many shortcomings, such as high computational complexity, poor robustness and generalization, and slow convergence speed. According to the spacetime equivalence and that the spatial and time domain processing algorithms can be transformed into each other, a suitable fitness function was derived, the improved particle swarm algorithm was used to search the arrival angle and frequency of signal, and the search results were classified with Kmeans clustering algorithm. Using particle swarm algorithms feature, such as global convergence, parallelism, can improve the algorithms searching capabilities. The computer simulation shows that the proposed method has better statistics and convergence performance than traditional methods.

Key words: spacetime twodimensional parameter estimation, Particle Swarm Optimization (PSO), spectrum estimation

中图分类号: