Robust Inference and Communication: Theory, Algorithms and Performance Analysis

稳健的推理和交流：理论、算法和性能分析

• 批准号: 0729031
• 负责人: Sean Meyn
• 金额: $ 38万
• 依托单位: University of Illinois at Urbana-Champaign
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2007
• 资助国家: 美国
• 起止时间: 2007-10-01 至 2011-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0729031&HistoricalAwards=false
• 关键词: Robust; Inference; Communication; Theory; Algorithms

Robust Inference and Communication: Theory, Algorithms and Performance Analysis Sean P. Meyn and Venugopal V. VeeravalliAs sensors and wireless communication become increasingly pervasive, and network topologies increasingly complex, there is an urgent need for new techniques for inference and communication in complex environments, as well as techniques to evaluate their performance. The goal of this investigation is to respond to these needs by following two complementary tracks. The first concerns methods for constructing inference and decoding algorithms for complex models, with possible modeling uncertainty, based on the geometry surrounding Kullback-Leibler (K-L) divergence and related methods developed by the investigators in their prior research. The second track treats performance evaluation and performance improvement. Both performance evaluation and algorithm selection are performed using Monte-Carlo or related sample path learning techniques. These approaches are chosen primarily because of their ease of application when compared to deterministic numerical techniques. The application of Monte-Carlo techniques comes at a price in the form of high variance. Efficient simulation and learning techniques are developed in concert with research on hypothesis testing and communication to construct faster algorithms for performance evaluation and adaptation. In addition to theoretical research on these topics, the investigators will transfer technology to industry and community organizations, including the Motorola Communications Center at Illinois, United Technologies Research Center, Vodafone, and the community wireless group CUWiN. Both graduate and undergraduate students will be engaged in applied and theoretical research.

鲁棒推理和通信：理论，算法和性能分析Sean P. Meyn和Venugopal V. Veeravalli随着传感器和无线通信变得越来越普遍，网络拓扑结构越来越复杂，迫切需要在复杂环境中进行推理和通信的新技术，以及评估其性能的技术。 本次调查的目标是通过两个相辅相成的轨道来满足这些需求。第一个关注的方法，为复杂的模型，可能建模的不确定性，周围的Kullback-Leibler（KL）发散和相关的方法，研究人员在他们以前的研究中开发的几何基础上，构建推理和解码算法。第二轨道处理业绩评价和业绩改进。 性能评估和算法选择都使用蒙特-卡罗或相关的样本路径学习技术来执行。 选择这些方法主要是因为它们与确定性数值技术相比易于应用。蒙特-卡罗技术的应用是以高方差形式为代价的。 有效的模拟和学习技术的发展与假设检验和通信的研究，以构建更快的算法的性能评估和适应。除了对这些主题进行理论研究外，研究人员还将向工业和社区组织转让技术，包括伊利诺伊州的摩托罗拉通信中心、联合技术研究中心、沃达丰和社区无线组CUWiN。 研究生和本科生都将从事应用和理论研究。