多新息辨识算法是估计系统参数的一种有效方法 ,其特点是可以克服坏数据对参数估计的影响 ,具有很强的鲁棒性。该文给出了衰减激励信号的定义 ,并在衰减激励条件下 ,利用随机过程理论 ,研究了随机系统多新息辨识算法的性能 ,给出了参数估计误差收敛时 ,衰减指数应满足的条件 ,以及算法中设计参变量的选择方法。分析表明 :若设计参变量选择为 r(t) =O(t2ε(lnt) c) ,(c>0 ) ,衰减指数满足 0≤ ε<1/ 4,则参数估计均方误差以 O 1(lnt) c 速度收敛于零 Multi-innovation identification algorithm is an effective method to estimate system parameters, which is characterized by overcoming the influence of bad data on parameter estimation and having strong robustness. In this paper, the definition of attenuated excitation signal is given, and under the condition of attenuated excitation, the stochastic process theory is used to study the performance of stochastic multi-renewal identification algorithm. The condition that the attenuation index should meet when the error of parameter estimation converges is given , As well as the choice of design parameters in the algorithm. The analysis shows that if the design parameters are selected as r (t) = O (t2ε (lnt) c), (c> 0) and the attenuation index satisfies 0 ≤ ε <1/4, lnt) c The speed converges to zero