基于分布式贝叶斯分层先验模型的稀疏估计及其在协作频谱感知中的应用

项目名称： 基于分布式贝叶斯分层先验模型的稀疏估计及其在协作频谱感知中的应用

项目编号： No.11301413

项目类型： 青年科学基金项目

立项/批准年度： 2014

项目学科： 数理科学和化学

项目作者： 李锋

作者单位： 西安交通大学

项目金额： 22万元

中文摘要： 压缩感知理论是当前应用数学学科研究的一个热门领域，其在宽带认知无线电网络频谱感知中的应用也受到了极大关注。为了解决无中心控制节点网络中的稀疏估计问题，必须突破高效分布式稀疏估计理论与信号处理技术。本项目拟从数学理论及其应用角度研究高效分布式稀疏估计理论，为中低信噪比下实数域及复数域上的稀疏估计问题提供较低复杂度的算法；基于此理论研究协作频谱感知算法，能有效对抗频谱空穴的快时变特性及共享信息的非理想特性。本项目主要创新包括：尝试将贝叶斯分层先验模型应用在分布式稀疏估计和协作压缩频谱感知中；尝试从L1/2正则子角度研究基于贝叶斯分层先验模型稀疏估计高效算法，以期获得更具典型意义的算法。项目的实施对稀疏估计理论的发展及其在信号处理领域中的应用有积极意义，并为认知无线电系统研发提供技术支持。

中文关键词： 认知无线电；压缩感知；频谱感知；信道估计；正交频分复用

英文摘要： The theory of compressed sensing has become popular in current research field of applied mathematics, and its application in spectrum sensing in wideband cognitive radio networks has also attracted a great deal of attention. In order to solve the problem of sparse estimation in decentralized network, breakthroughs must be made in the field of efficient distributed sparse estimation and signal processing theory. This project will study efficient distributed sparse estimation theory which is of low computational complexity for sparse signal representation in both real and complex number fields in low and moderate signal-to-noise ratio regimes. Based on the theory, the algorithm of cooperative spectrum sensing, which can effectively combat both the fast time variation of the spectrum holes and the imperfections of sharing information, will be studied. The innovation of this project is twofold. Firstly, Bayesian hierarchical prior modeling is utilized to distributed sparse estimation and cooperative compressed spectrum sensing. Secondly, efficient Bayesian hierarchical prior modeling based sparse estimation is studied from the angle of L1/2 regularization to obtain more typical algorithms. The implementation of this project will have a positive effect for the development of the theory of sparse estimation and its ap

英文关键词： Cognitive radio；Compressed sensing；Spectrum sensing；Channel estimation；Orthogonal Frequency Division Multiplexing