多数动物的生理动作,都是在大脑的参与下完成的.了解大脑信号的组成具有重要的意义.而针对脑电这样的盲信号,对其有效信息的提取变的尤为重要.为此,本文以呼吸下的脑电信号为例,研究呼吸曲线有效信息的提取方法.通过差分,阈值法和预估频法得到有效峰谷点,建立峰谷模型.并进一步根据呼吸曲线始末点对呼吸曲线进行分类,建立了呼吸周期计算模型,最后用实例验证了模型的有效性.此外,本文提出的算法,也可用于除呼吸之外的其他生理信号的处理. Most animal’s physical movement is completed with the participation of the brain.It is important to know the composition of brain signals.But the blind signal against EEG is particularly important for the extraction of effective information.Therefore, Taking the EEG signals under the breath as an example, the effective information extraction method of the respiratory curve was studied, the effective peak point and the peak-valley model were obtained by difference, threshold method and pre-estimation frequency method, and the respiratory curve Then the model of respiratory cycle is established, and the validity of the model is validated by an example.Furthermore, the algorithm proposed in this paper can also be applied to the treatment of other physiological signals besides the respiration.