基于时间序列分析的预测控制算法

Algorithm of Prediction Control Based on Time Series Analysis

在分析Turbo码的译码结构以及外部信息传递特性的基础上,从时间序列的角度分析分量译码器外部信息的传递规律,基于预测算法对Turbo码的迭代译码结构进行了改进,提出了单次预测和复次预测的改进算法。建立分量译码器的n阶多项式线性模型,将前n次迭代信息保存至线性模型中,采用多项式曲线拟合或指数拟合算法预测分量译码器第n+1次迭代的外部信息值。利用预测的外部信息值代替分量译码器第n+1次迭代的译码过程。在保证外部信息准确预测、误码率性能相近的前提下,减少了迭代次数和迭代时延,简化了译码结构。

In the analysis of Turbo codes decoding structure and transmission characteristics of exterior information,based on the time series and then from the point of view of an exterior decoder component of the transmission of information laws,an improved iterative decoding structure based on prediction algorithm is presented.The n-order polynomial linear model of the component decoder is established,and the values of the former n-iterations exterior information is saved in to the linear model.The polynomial fitting curve or the index component algorithm is adopted to predict the（n＋1） iterative exterior information.The value of predicted exterior information is taken instead of the（n＋1） iterative exterior information.Under the premise of accurately predict the exterior information and the closely bit error rate performance,it simplifies the decoding structure and reduces the number of iterations and iteration delay.