司法消噪算法改进及关键参数优化

Improvement of Judicature De-noising Algorithm and Optimization of Key Parameters

对智能人工呼吸床系统的司法消噪算法进行改进,将司法处理思想引入时间序列消噪中,提出一种新的呼吸周期预测算法。在此基础上完善司法消噪算法的法规综合处理函数,优化影响消噪效果的关键参数(违法因子、制定噪声法规所需的依据数据数目、第i个立法数据权值及最大举证数据数目)。实验结果证明,改进的算法对所处理的信号序列要求较低,适用范围更广,预测效果更好。

This paper improves Judicature De-noising Algorithm（JDNA） in intelligent mechanical ventilation bed system.The judicature process is induced into the de-noising disposal of time series in algorithm,which builds a new de-nosing algorithm.On this basis,this paper optimizes key parameters that affect de-noising effects： illegal factor,the number of data used to formulate noise regulation,the first i legislative data weight,the number of largest evidence data.Through improvements upon the algorithm,it is relatively loose for requirements applicable scope of handling signal sequences,wide for applicable scope,the forecasting effect is obvious,to achieve the ideal of the treatment effect.