High-Quality MIMO Sonar Imaging Using Cross Correlation Suppression

Multiple-Input Multiple-Output (MIMO) sonar systems enhance imaging resolution by synthesizing a virtual array produced by orthogonal transmitted waveforms and a bank of matched filters. Up- and down-chirp pulses are widely used due to their approximative orthogonality and robustness for underwater remote sensing application. However, the cross-correlation function terms between them always exist and thereby, degrading the quality of the MIMO sonar image. In this letter, we propose a novel cross-correlation suppression method for improving the image quality of MIMO sonar. The proposed method leverages the frequency-domain differences between the envelopes of auto-correlation functions (ACFs) and cross-correlation functions (CCFs) of up- and down-chirp pulses. After Fourier transformed, the envelopes of ACFs approximately to be square wave function, while CCFs approximately to be sinc function. Therefore, CCFs energy are concentrated and easier to be suppressed. By setting the main components of the CCFs to zero in the frequency domain of the multi-beam outputs’ envelopes, the method effectively suppresses range sidelobes caused by cross-correlation, thereby enhancing the quality of the imaging results. Through numerical simulations and lake experiments, we quantitatively analyze the performance of the proposed method and demonstrate its effectiveness in MIMO sonar imaging.