A new TOPS algorithm Based on Optimal Reference Frequency Optimization

The Test of Orthogonality of Projected Subspaces (TOPS) algorithm estimates the direction of arrival (DOA) by projecting the signal subspace at a reference frequency onto multiple frequencies using a transformation matrix, and exploiting the orthogonality between the projected signal subspace and the noise subspace. However, this algorithm often exhibits spurious peaks, and its performance is highly sensitive to the choice of reference frequency. To address these issues, this paper proposes a novel improvement to the TOPS algorithm. By incorporating multi-dimensional feature differences of the signal subspace and adopting a multi-subband weighted strategy to optimize frequency selection, the proposed method significantly enhances robustness under complex conditions. Additionally, the signal subspace projection is refined to reduce spurious peaks caused by subspace estimation errors. Simulation results demonstrate that, compared to the original TOPS algorithm, the proposed approach effectively suppresses spurious peaks and improves estimation accuracy and resolution through optimized reference frequency selection.