A Kernel-based TDOA Localization Algorithm

摘要：

A Kernel-based Weighted Total Least Square (KWTLS) method is proposed at first. Based on this method, an effective TDOA localization algorithm is developed. The results of simulation show that using KWTLS method instead of the first WLS in the Chans algorithm can significantly improve its performance in distant source cases. A dynamic weight coefficient adjusting method is also proposed, which can effectively reduce the effect of noise variance.

关键词： localization TDOA kernel method weighted total least square

作者: Li Zhang XiaoDong Yu

作者单位: Shandong Institute of Light Industry Jinan China

会议类型: 国际会议

会议名称: The 2010 International Conference on Computer Application and System Modeling(2010计算机应用与系统建模国际会议 ICCASM 2010)

会议地点: 太原

会议语种:英文

页码: 412-415

在线出版日期: 2010-10-22（万方平台首次上网日期，不代表论文的发表时间）