Cross-Spectral TLS-ESPRIT Estimating Method in Colored Stochastic Noise
Huaxun Zhang
Abstract
Huaxun Zhang
Abstract
The modern cross-spectral estimation methods are effective to restrain colored observed noises.Based on further analysis of signal estimation for signal via rotational invariance techniques,a novel cross-spectral TLS-ESPRIT method in colored stochastic noise is proposed.It does not need any pre-knowledge about the colored noises and can work in very low SNR.The peak searching procedure inherent in previous cross-spectral estimation methods is eliminated and signal parameters can be obtained by eigenvalue.Two times SVD decomposition makes the computation greatly deduced.The simulation results show this method exhibits good spectral estimation resolution and stability.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The modern cross-spectral estimation methods are effective to restrain colored observed noises.Based on further analysis of signal estimation for signal via rotational invariance techniques,a novel cross-spectral TLS-ESPRIT method in colored stochastic noise is proposed.It does not need any pre-knowledge about the colored noises and can work in very low SNR.The peak searching procedure inherent in previous cross-spectral estimation methods is eliminated and signal parameters can be obtained by eigenvalue.Two times SVD decomposition makes the computation greatly deduced.The simulation results show this method exhibits good spectral estimation resolution and stability.
Key concepts: Colors of noise, Colored, Rotational invariance, Spectral density estimation, Noise (video), SIGNAL (programming language), Computation, Computer science