Research The Respiratory Signal In Sleep With The NSP Algorithm

Sleep is an activity that accounts for nearly a third of human lifespan,which is also one of the most critical parts of human self-regulation and one of the activities that everyone has to do every day.But trapped in the shackles of equipment and technology,there is already a long time that humans can not carry out comprehensive and detailed research.With the development of technology and big data,researchers in this field are starting to work on their own sleep.At present,the researchers have successfully extracted the respiratory signal during the state of sleep which can be divided into W(wake stage),N1,N2,N3(N1,N2 and N3 are called No Rapid Eyes Movement Stage)and R(Rapid Eyes Movement Stage).NSP algorithm is a zero-space tracking algorithm based on operator,which was proposed by two professors,Peng Silong and Huang Wenliang,in 2008.It is a new self-adaptive signal decomposition method based on the signal decomposition algorithm.It has more advantages like this algorithm defines signal decomposition as an optimization problem and defined operator to decompose the sum of multiple sub-signals from the signal.By defining some parameterized integrals or differential operators,the algorithm extracts local narrowband signals that can be removed by the defined operator from a complex signal.In this article,firstly,we will introduce the respiration signal processing and the development of that.Then,after learning algorithm of NSP,we are going to apply it on the respiration signal to find the reasonable parameters analysis,denoising and extraction.Then we look for different measures to classify the respiratory signals,and explore them with the classification standards given by the experts.The experimental process and results of three methods of energy belt,scatter method and frequency and depth variance method are given in the text.Finally,reasonable distinguishing criteria are found to interpret the respiratory signals.