Research On Sub-nyquist Sonar Sensing With Finite Rate Of Innovation Principle

The trend of sonar system is systematic and comprehensive,which means it will be more and more complicated,effective sampling and data transmission is an important foundation.The amount of sampled data by traditional Nyquist sampling is too large for data transmission underwater and wireless.Therefore,the research of high efficient sampling scheme for sonar echo is necessary.In recent years,Finite Rate of Innovation(FRI)sampling theory is proffered to sample and recover the unknown parameters of FRI signals far below the Nyquist sampling rate.In this paper,the sonar echo is modeled as a parameterized signal,the FRI sampling theory can be used to subsample sonar echo.Analog preprocessing is first done for the sonar echo,and then implement a low-rate sampling,a set of Fourier series coefficients of sonar echo can be obtained from the sampled data,various recovery algorithms can be used to estimate the unknown parameters according to the characteristics of sonar echo.The main work in this paper is:1.Research on the modeling of sonar echo and FRI Sampling Theory.The modeling of the sonar echo is described in detail,including the modeling of the sonar target echo by the bright spot model method,and the reverberation signal is modeled by the combination of the point scattering method and the unit scattering method.The sampling and recovery of FRI sampling theory is analyzed.The performance of different sampling structures and recovery algorithms is compared,which provides the theoretical basis for processing the sonar echo using the FRI principle.2.Focusing on the detection of stationary point targets in the non-reverberation background.The sonar echo is modeled as a parameterized signal determined by the delays and amplitudes,and the FRI sampling theory can be applied.The Fourier series coefficients of the sonar echo can be obtained using the single channel modulation lowpass filter structure and the multi-channel modulation integral structure.The two structures have their own advantages and disadvantages,the sampling structure is chosen according to the specific situation.The influencing factors of the two sampling structures are analyzed.3.Focusing on the detection of moving targets in the background of reverberation,the sonar target echo is modeled as a parameterized signal determined by the delays,amplitudes and Doppler frequencies,sub-Nyquist sampling can be realized using the FRI sampling theory.Since the Doppler frequency of the sonar target echo different from the reverberation signal,the Doppler focusing technique is proffered to deal with the sonar echo,which can eliminate the interference of reverberation.Recovering the Doppler frequencies of the target at first,and then the delays and amplitudes can be recovered,the advantages of Doppler focusing processing are illustrated and simulated.4.Hardware experimental verification of FRI sampling of sonar echo.The design of the hardware module is completed,and the analog preprocessing structure is composed of the multiplier module,the amplifier module and the analog low-pass filter module.The NI PXIe device is controlled by LabVIEW to implement the signal output,data sampling,data storage and PC recovery.This experimental prototype proves that the sub-Nyquist methodology for sonar echo described in this paper is actually feasible in practice,the compression ration can achieve to 4.72%.