2025_programme: The Compressed Sensing Spatial Spectrum Synthesis Method Based on Guided Cross Spectral Matrix Cumulative Preprocessing

• Day: June 19, Thursday
    Location / Time: TBA at 15:30 - 16:30
• Last minutes changes: -
• Session: Poster session
  Organiser(s): N/A
  Chairperson(s): N/A
• Lecture: The Compressed Sensing Spatial Spectrum Synthesis Method Based on Guided Cross Spectral Matrix Cumulative Preprocessing
  Paper ID: 2140
  Author(s): Enming ZHENG, Chunnan SONG, Yi LI, Hua FANG
  Presenter: Enming ZHENG
  Abstract: For the case of broadband processing, in order to solve the performance degradation problem of frequency domain compressed sensing spatial spectrum synthesis under low signal-to-noise ratio(SNR), a compressed sensing spatial spectrum synthesis method based on cumulative preprocessing of guided cross spectral matrix was proposed. Firstly, the time-frequency transform for sensor array collecting data was done in frequency domain, the sensor array phase shift compensation matrix for compensating array cross spectral data was constructed on estimated azimuth angle, and the cross spectral matrix (referred to guided cross spectral matrix) after phase shift compensation of every frequency unit was obtained. Then, the sensing matrix and measurement values were cumulative via guided cross spectral matrix of every frequency unit, and the signal-to-noise ratio of the data using for convex optimization solution was enhanced. Finally, the spatial spectrum was obtained via convex optimization method. Compared with the spatial spectrum synthesis methods of conventional frequency domain compressed sensing(FCS method in this paper) and complex domain compressed sensing(CCS method in this paper), the processing results of numerical simulation and measured data show that the requirements of lowest input SNR and signal length have been decreased in different degrees(compared with FSC, the demand for lowest input SNR is reduced by 6dB. compared with CCS, the demand for signal length is reduced from 0.2s to 0.05s), and the accuracy and stability of spatial spectrum synthesis was enhanced.\nIn addition, it is necessary to note that the existing improvement methods of frequency domain and complex domain compression sensing are also applicable for this method.
• Corresponding author: Dr Enming ZHENG
  Affiliation: Institute of Acoustics, Chinese Academy of Sciences
  Country: China