Abstract:

To solve the high computational cost problem of the current algorithm to achieve multi-dimensional parameter estimation of bistatic electromagnetic vector sensors multiple input multiple output (EMVS MIMO) radar, the dimensionality-reduction transformation technology is utilized to achieve low complexity angle and polarization parameter solutions. To address the issue of large dimensionality in array received data, a corresponding beamspace transformation matrix is designed to achieve dimensionality-reduction processing of array received data. To address the high computational complexity of the algorithm itself, a parallel factor decomposition algorithm with low computational complexity is adopted. The proposed algorithm can accurately solve the transmission factor matrix and reception factor matrix. Moreover, by constructing a new estimating signal parameter via rotational invariance techniques (ESPRIT) relationship, the solution of the transmit/receive pitch angle can be achieved. Furthermore, the estimation of the transmit/receive azimuth angle, transmit/receive polarization angle and transmit/receive polarization phase difference can be achieved through the reconstruction of the transmit/receive spatial response matrix. Simulation experiments show that the proposed algorithm can maintain superior multidimensional parameter estimation performance while reducing computational complexity.

Key words: bistatic electromagnetic vector sensors multiple input multiple output (EMVS-MIMO) radar, multi-dimensional parameter estimation, beamspace spatial transformation, parallel factor decomposition algorithm