【摘要】 针对多通道乘性噪声系统问题的实际需要 ,推广 Rajasekaran滤波算法 ,利用线性最小方差的概念和投影公式 ,导出含有乘性噪声统计特性参数阵特殊乘法运算的新息协方差矩阵 ,并利用这个中间矩阵 ,在观测为多通道 ,且各个通道的乘性噪声不同 ,以及系统的动态噪声和观测噪声同时刻相关的情况下 ,导出状态递推滤波算法 ,该算法在线性最小方差意义下是最优的。并对该算法进行仿真研究 ,仿真结果表明了该算法的有效性更多还原

【Abstract】 Owing to the practical requirement of mult i-channel system with multiplicative noise, the Rajasekaran filtering algorithm is broadened. Use of the concept of linear minimum variance and projection form ula deduce the covariance matrix of the innovation sequence, which contains spec ial multiplicative operation of the statisticly characteristic parameter matrix with multiplicative noise. The state recursive-filtering algorithm is daduced b y applying the covariance and according to the multi-channel observation and th e different multiplicative noise of each observation channel and the relevance o f system and observaed noise in the same time. This algorithm is optimal under t he linear Aminimum variance. A number of simulations are performed.更多还原