Abstract:

To address the issue of degraded positioning performance in non-uniform noise environments for terahertz signals in the near-field region, a localization algorithm based on the reconstruction of the received signal data covariance matrix is proposed, utilizing a array-of-subarrays structured hybrid beamforming architecture for a terahertz massive multiple input multiple output system. Firstly, by increasing the number of time slots, full-array information is obtained from a limited number of radio frequency chains. Secondly, the data covariance matrix of the full-array information is decomposed to obtain a diagonal matrix containing the noise power. Then, the minimum value among the diagonal elements is selected and used to replace the remaining diagonal elements, resulting in a reconstructed data covariance matrix. Finally, the source parameters are estimated using a subspace algorithm. In addition, the Cramer-Rao lower bound is derived to validate the performance of the proposed algorithm. Theoretical analysis and simulation results demonstrate that, compared to existing algorithms, the proposed algorithm can achieve effective suppression of non-uniform noise interference with lower computational complexity, improving the localization accuracy. It also validates that hardware costs can be reduced by increasing the number of time slots.

Key words: terahertz, near-field, non-uniform noise, localization