广义最小二乘的算法结构，广义最小二乘递推的每一步用两步递推算法，先把噪声参数和模型参数均给个初值，然后第一步先假设噪声参数已知，用最小二乘估计出模型参数，第二步再用最新的模型参数来估计噪声参数，如此循环直到辨识精度达到指定要求或者可用数据用完为止-Generalized least squares algorithm structure, generalized recursive least squares every step of the two-step recursive algorithm, first noise parameters and model parameters to the initial value, then the first step is to assume that the noise parameters are known, with the smallestsquares to estimate the model parameters, and the second step, and then the latest model parameters to estimate the noise parameters, and so on until the identification accuracy specified requirements or available data is used up