电子学报Issue(6)：1055-1060,6.DOI:10.3969/j.issn.0372-2112.2014.06.003
基于时间序列分析的频谱异常自主检测和稳健估计方法
An Autonomous Detection and Robust Estimation Method of Spectrum Anomaly Based on Time Series Analysis
摘要
Abstract
How to precisely detect electromagnetic spectrum anomaly is a major challenge for radio monitoring and electro-magnetic environment evaluation ,especially in the condition of complex electromagnetic environment and lack of pre-knowledge in-formation about frequency use .Based on time series analysis theory ,a timing model which presents the correlation between the pre-vious and the following sequence of spectrum occupancy ,is built to help us realize the autonomous identification and robust estima-tion of typical spectrum anomaly .The analysis results indicate that ,without actually requiring pre-knowledge of frequency database and radio monitoring historical data support ,this method can effectively identify the types of spectrum anomaly ,occurrence time , anomaly effect intension and other relative information .Furthermore ,through the robust estimation of spectrum occupancy model ,we can significantly improve the model′s fitting performance and raise the adaptability and robustness of the model to external interfer-ences .关键词
频谱异常/自主检测/稳健估计/电磁环境Key words
spectrum anomaly/autonomous detection/robust estimation/electromagnetic environment分类
信息技术与安全科学引用本文复制引用
王磊,谢树果,苏东林,王国玉..基于时间序列分析的频谱异常自主检测和稳健估计方法[J].电子学报,2014,(6):1055-1060,6.基金项目
国家重点基础研究发展规划（973计划）项目（No ．2010CB731801）；国家自然科学基金 （）
