Development Of An Ultrasonic Doppler Current Meter

As an important parameter of marine ecological environment monitoring,ocean current plays an important role in monitoringglobal climate change,coastal erosion,destruction of marine engineering,marine organism migrationand diffusion of marine pollutants,etc.In all instruments that measure ocean currents,the one which utilizes ultrasonic Doppler principleto measure ocean currents is most widely used.Moreover,in the design and application of ultrasonic Doppler current meters,the key point is to detect ultrasonic echo signals and accurately measure the frequency of these signals,therefore,it is of great significance and use value to find out a high-precision frequency estimation algorithmsuitable for ultrasonic Dopplercurrent meters.At present,researchers at home and abroad mainly use the spectrum analysis method of Discrete Fourier Transform(DFT)to estimate the frequency,which has the advantage of fast calculation(uses Fast Fourier Transform).However,as DFT has a picket fence effect,the error of frequency estimation is comparatively large,especially when the data sampling length is shorter.In view of this situation,the paper will try to use sparse representation theory to carry out the frequency detection of ultrasonic Doppler current meter echo signals,because the signal sparse representation theory when applied to signal frequency estimation has the strong points of high frequency estimationresolution ratio and strong anti-interference ability,but its weak point is that the signal sparse decomposition speed is relatively slow and the calculated amountis large,which will result in a too slowsignal frequency detection speed.Therefore,in order to improve the speed of signal frequency detection,this paper,directing at the characteristics of ultrasonic Doppler current meter echo signals,will reasonably construct the atomic dictionary for sparse decomposition,and make some improvements to the sparse decomposition algorithm so as to improve the speed of sparse decomposition and make this algorithm be well applied to ultrasonic Doppler current meters.It then compares this frequency estimation algorithm with the spectrum analysis method of DFT,and uses Matlab software to conduct simulation verification in order to confirmthat this frequency estimation algorithm will effectively improve the accuracy of frequency estimation.Finally,it establishes a current meter systemhardware and software platform and embeds the newly researched frequency estimation algorithm to make practical verification.This paper,by using sparse representation theory to detect the frequency of echo signals of ultrasonic Doppler current meters,effectively improves the accuracy of frequency estimation and plays an important role in the design and application of ultrasonic Doppler current meters.