TV系统中基于Kalman滤波器的自回归频谱空洞预测  

Autoregressive Spectrum Hole Prediction Model Based on Kalman Filter for Broadcast-TV Systems

作　　者：鲜永菊[1] 曹华孝[1] 

机构地区：[1]重庆邮电大学移动通信技术重点实验室,重庆400065

出　　处：《电视技术》2011年第2期71-75,共5页Video Engineering

基　　金：重庆邮电大学自然科学基金项目(A2008-28);重庆大学研究生创新重点项目(200904B1A0010306);重庆市教委项目(KJ090502)

摘　　要：研究了基于Kalman滤波器的自回归的频谱空洞预测模型。首先讨论了广播电视系统中授权用户和认知用户共存的系统模型;其次讨论了自回归预测原理和Kalman滤波预测的无偏性,并针对不同的通信模式提出不同的预测方法。理论分析和仿真结果均表明,基于Kalman滤波器的自回归预测值能很好地与理论值近似匹配,采用预测模型可以大大减少授权用户和认知用户的冲突,提高系统利用率。In this paper, an autnregressive channel prediction model based on Kahnan filter is presented for broadcast-TV systems to estimate spectrmn holes. Firstly, this paper discusses the spectrum hole of the coexistence with rental users and licensed users for broadeast-TV systems; Next we investigate the autoregressive prediction principle, and prediction based on Kahnan filter＇s deviation. Different prediction rules apply to periodic and stochastic ON-OFF patterns. Theoretical analysis and simulations show that autoregressive channel prediction model based on Kalman filter can better match the theoretical value, and CR systems based on this scheme can reduce the number of collisions between licensed users and rental users, and improve utilization of system.

关 键 词：广播电视系统 自回归 KALMAN滤波 频谱空洞 认知无线电 

分 类 号：TN914[电子电信—通信与信息系统]