Adaptive Bayesian Receivers in Fading Channels: A Sequential Monte Carlo Filtering Design Paradigm

点击翻译按钮获取中文摘要

AbstractThe coming generation of tetherless communication technology promises a giant leap forward in informationaccessibility. Advanced features of the so-called fourth-generation wireless systems and beyond, such as data rates compatible with multimedia applications, will enable many emerging applications not possible with current wireless systems. However, it is not at all clear how wireless receivers should be optimally designed to meet the technical challenges introduced by the wider bandwidths and higher data rates inherent in the future wireless systems. It is generally believed that the real niche for future wireless receivers lies in the development of adaptive systems to perform sophisticated signal processing functions. But, at this time there is a lack of concrete principles that can be used to design these futuristic receivers. It is important at this stage to acquire the insights and theoretical tools that may help spark revolutionary breakthroughs in this field.Investigation of design methodologies of adaptive Bayesian receivers in single-user and multiuser fading channels is proposed. The approach is to formulate the problems of signal reception in unknown time-varying channels as multivariate Bayesian inference problems. Sequential Monte Carlo filtering methods, the relatively simple but extremely powerful numerical techniques recently developed in the field of statistics, will be employed to develop adaptive systems for computing the Bayesian estimates of the channels and data. An array of receiver design problems found in wireless communications, such as mitigation of various types of radio-frequency interference (including multiple-access interference, narrowband interference, impulsive noise), tracking of fading channels, resolving multipath channel dispersion, space-time processing by multiple antennas, exploiting coded signal structures, etc., will be treated under the unified framework of sequential Monte Carlo Bayesian estimation. The theoretical effort in this project is expected to culminate in the formulation of novel receiver design concepts applicable for future wireless systems.