摘要
针对伪测量卡尔曼滤波压缩感知算法应用于时变信道时精确度和实时性的不足,提出了改进的卡尔曼滤波压缩感知信道估计算法。该算法进一步优化了扩展滤波过程的范数框架,利用Levenberg-Marquardt方法实现了方差矩阵的自优化,解决了估计误差不能一致减少的问题,保证了全局收敛性;根据卡尔曼增益设置了迭代收敛条件,解决了伪测量过程的自适应收敛问题。仿真分析表明,该算法的估计精度和收敛速度有较大程度提高,在SUI-3信道条件下性能明显优于传统信道估计方法。
In order to improve the accuracy and efficiency of Pseudo measurement process,an improved channel estimation algorithm based on Kalman filtered compressed sensing was proposed. This algorithm optimized the PM process, and the Levenberg-Marquardt method was used to optimize the covariance matrix in order to main-tain the reduction of the evaluated error, which ensured the global convergence of the algorithm. The conver-gence condition was defined according to the value of Kalman filter gain The simulation results showed that, under the same simulation condition, the estimation accuracy and convergence speed were improved greatly com-pared to the traditional algorithm, which was also shown under the SUI-3 channel condition
出处
《探测与控制学报》
CSCD
北大核心
2016年第6期99-103,共5页
Journal of Detection & Control
基金
国家自然科学基金项目资助(61201209)
陕西省电子信息系统集成重点实验室基金项目资助(2011ZD09)
