中文摘要：

提出了一种新的未知信号状态模型——多项式预测模型，并给出其滤波算法。分析表明，采用该方法建立的状态方程不需要已知信号本身的参数信息，都能准确描述运动或信号的动态。因此，提出的多项式信号的最优滤波算法适用于任何可以用多项式描述的信号的状态估计问题。计算机仿真验证了该方法的正确性、有效性及实用性。

英文摘要：

A novel state model of unknown signal-polynomial prediction model （PPM） and its optimal filtering algorithm are proposed in this paper. The analysis shows that, the state equation set up with the proposed method can accurately describe the motion or the state of a signal, without requiring any prior knowledge of the signal parameters. As a result, the optimal filtering algorithm corre-sponding to our PPM can be used to any state estimation problem of polynomial signals. The simulation results verify the effectiveness, correctness, and the practicality of the proposed method.