基于EM算法的自适应离子单通道信号参数估计技术研究

Study of an adaptive technology for estimating the parameters of single ion channel signal based on EM algorithm

许多离子通道的电流信号特别微弱,运用标准的阈值检测器方法不能估计出通道信号的各种参数。基于隐Mark-ov模型（HMM）的各种算法运算量大,不具有自适应功能。本文借用增广最小二乘法（ELS）参数辩识算法中的运算技巧,应用随机逼近原理,在EM算法的基础上推导出一种具有自适应能力的离子通道信号参数估计技术,仿真证明其估计精度较高,稳健性强,而且易于实现。

Many kinds of ion channels have particularly weak signal, and their parameters could not be estimated satisfactorily by the standard method of threshold detector. Those algorithms based on hidden Markov models (HMM) have the limitation that they require very complicated computations and they are not adaptive. Enlightened by the technique used in extended least squares (ELS) algorithm for parameters identifying, based on stochastic approximation principle a new adaptive algorithm was concluded from EM algorithm. Simulations suggest that it estimated parameters satisfactorily from low signal-noise ratio patch-clamp recordings, and had high precision and quality. Moreover the algorithm performs very easily.