时序自适应滤波技术用于听觉诱发电位的跟踪

本文针对诱发电位是具有重复出现特性的非平稳信号这一特点，利用时间序列自适应滤波技术对听觉诱发电位进行动态跟踪，时间序列自适应滤波技术将非平稳的听觉诱发电位进行分段处理，每段用一个专门的ＬＭＳ自适应信号增强滤波器来实现，实验表明，利用这一新方法不但能够得到优于常规平均技术的结果，更重要的是它能不断地跟踪诱发电位的变化轨迹，提供变化的细节，为临床诊断和神经生理学的研究提供有力的手段。

TIME SEQUENCED ADAPTIVE FILTERING APPROACH FOR TRACKING BRAINSTEM AUDITORY EVOKED POTENTIAL

it is well known that evoked potentials are nonstationary signal with acharacteristic of recurring statistic. With the help of this feature, we presented a newmethod to track the variation of brainstem auditory evoked potential (BAEP) trial to trial using time-sequenced adaptive filtering (TSAF) technique. In TSAF, the nonstationaryBAEP are segmented into many subsets and every subset is processed as stationary signal using LMS adaptive signal enhancer. It is reported that, by using this method, themeasurement time can be largely reduced comparing with the traditional ensembleaveraging method and the signal-to-noise ratio(SNR) can be improved in a largedegree. Furthermore, single BAEP signal can be obtained which makes it possible forclinicians to analyze the variation of signals across this.