Robust Inference and Communication: Theory, Algorithms and Performance Analysis

随着传感器和无线通信的日益普及，以及网络拓扑结构的日益复杂，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。这项调查的目标是通过遵循两个互补的轨道来回应这些需求。第一个问题是基于围绕kullbackleibler （K-L）散度的几何结构和研究者在其先前研究中开发的相关方法，为可能存在建模不确定性的复杂模型构建推理和解码算法的方法。第二个轨道处理绩效评估和绩效改进。性能评估和算法选择都是使用蒙特卡罗或相关的样本路径学习技术进行的。选择这些方法主要是因为与确定性数值技术相比，它们易于应用。蒙特卡罗技术的应用是以高方差的形式为代价的。有效的模拟和学习技术与假设检验和交流的研究相结合，以构建更快的性能评估和适应算法。除了对这些主题进行理论研究外，研究人员还将把技术转移到工业和社区组织，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织CUWiN。研究生和本科生都将从事应用和理论研究。

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.