| Weak signal detection is the most important subject in signal processing front-end.Weak optical signal and weak sound signal produced by human body are typical small signals.These signals usually contain rich useful information.The effective detection of them,namely collection,amplification,denoising,and extraction of useful information has always been the focus and difficulty in the research field of weak signals.In this paper,single photon measurement technique is used to measure weak optical signal.The technology has been widely used in radioactive detection,analysis of non-destructive materials,three-dimensional imaging,quantum communication and medical imaging.In this study,we first analyze the shortcomings of the existing single-photon measurement technology in noise removal,and then propose a new accurate and effective noise removal method.The details of the new scheme are as follows :(1)in the hardware circuit,an AD sampling unit and a DA voltage output unit are added on the basis of the original preamplifier unit,voltage comparison unit and pulse counting unit.The AD sampling unit outputs the pre-amplified signal to FPGA.The DA voltage output unit changes the threshold voltage to be set by FPGA from digital to analog.(2)FPGA programming,there are two parts.The first part is waveform peak detection,the dynamic single threshold peak detection method is proposed,which can effectively detect the peak value of each photoelectronic pulse and noise,so that the threshold voltage of voltage comparator can be set according to the peak value.The second is the photoelectric pulse counting part,which counts the rectangular pulse output by the voltage comparator.Then the simulation and testing :(1)the hardware circuit part is simulated with Multisim,and the subroutines are simulated with Modelsim software,respectively verifying the correctness of the corresponding design.(2)Waveform generator is used to generate corresponding waveform,and the whole design is tested.The maximum relative deviation between the measured threshold voltage and the theoretical value is 3.778%.Compared with the previous methods,the noise in photoelectronic signal can be filtered more effectively and accurately.This scheme is helpful to improve the existing photon measurement and photon imaging technology.The weak sound signal,namely the popping sound generated during the cervical vertebra movement in this study,is processed by wavelet transform.Wavelet transform has been widely used in the analysis of non-stationary signals such as lung sound signal,heart sound signal,ECG signal and epileptic signal.In this paper,firstly,the method of obtaining cervical spine popping sound and denoising scheme based on wavelet transform are introduced.Secondly,a Wavelet Transform Based Stationary-nonstationary(WTST＿NST)filter was used to reduce the noise of cervical spine popping sound.The frequency component analysis of the denoised signal is also carried out.Finally,in the test part,the image of cervical spine popping sound after wavelet denoising was compared and analyzed.The denoising percentage is 91.02%under suitable conditions.This study is helpful for the daily monitoring of cervical vertebra status. |
