Non linear filters and hidden Markov models

非线性滤波器和隐马尔可夫模型

• 批准号: 7964-2009
• 负责人: Elliott, Robert
• 金额: $ 3.35万
• 依托单位: University of Calgary
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2013
• 资助国家: 加拿大
• 起止时间: 2013-01-01 至 2014-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=521066
• 关键词: Non; linear; filters; hidden; Markov

The central theme of the research is the optimal estimation of signals observed in noise. The most successful algorithm for this is the Kalman filter. Extensions of the Kalman filter when the system jumps around a set of nominal operating modes have been developed, but they have been ad hoc procedures. Indeed, the IMM technology appears to be the current standard in the United States. The research will develop an exact filter obtained in previous work which provides improvements on the IMM, investigate Viterbi modifications and derive related smoothers. The results potentially have applications in many areas of communications, tracking and signal detection.

研究的中心主题是在噪声中观察到的信号的最佳估计。最成功的算法是卡尔曼滤波器。扩展的卡尔曼滤波器时，系统跳跃周围的一组标称操作模式已经开发出来，但他们一直特设的程序。事实上，IMM技术似乎是美国目前的标准。这项研究将开发一个精确的过滤器，在以前的工作中，提供了改进的IMM，调查维特比修改，并获得相关的平滑。这些结果可能在通信、跟踪和信号检测的许多领域中有应用。