Algorithm

Pure mathematics

Statistics

Covariance matrix

Materials science

Toeplitz matrix

Signal-to-noise ratio (imaging)

Telecommunications

Computer science

Signal processing

Matrix (chemical analysis)

Mathematics

Estimator

Sample mean and sample covariance

Artificial intelligence

Composite material

Radar

Speech recognition

Preprocessor

IEEE Communications Letters

ESPRIT-like algorithm and its variants are widely used in preprocessing schemes that provide high performance of high-resolution parameter estimation algorithms for coherent signals. However, in the presence of certain phase differences between coherent signals and signal angles, this method suffers serious performance degradation and even fails. In this letter, a modified scheme based on Forward and Backward Partial Toeplitz Matrices Reconstruction named as FB-PTMR is proposed that exploits half rows of the sample covariance matrix (SCM) to reconstruct the data matrix to overcome this weakness. The new approach provides better performance on estimation and probability of resolution compared to the ESPRIT-like method by exploiting a small part of information of the sample covariance matrix. The simulation results verify the effectiveness of our proposed method.

An Improved ESPRIT-Like Algorithm for Coherent Signals DOA Estimation