Abstract: For the instability problem of multiple signal classification (MUSIC) beam-forming estimating subspace in frequency domain, a multiple signal classification beam-forming method based on time-domain analysis (TAMUSIC) was proposed. Firstly, the complex analysis data were obtained from the time domain real data by the Hilbert transformation. Secondly, the covariance matrix was constructed in time domain after the time delay, and the noise subspace was statistically obtained by eigen-decomposition. Finally, the beam was obtained by the orthogonal properties of the noise subspace in direction of arrival. The processed results of numerical simulation and measured data show that under the case of fast moving target, compared to the MUSIC beam-forming method, TAMUSIC beam-forming method can statistically get the noise subspace and the beam in direction of arrival for fast moving target, improve the side-lobe level more than 3dB. It effectively detects the fast moving target, and has no phenomenon of false target and split beam. It also improves the stability of MUSIC beam-forming method in practical project.

Key words: multiple signal classification (MUSIC) beam-forming; fast moving target; time domain analysis