Robust regression and outlier detection
Robust regression and outlier detection
Signal processing with alpha-stable distributions and applications
Signal processing with alpha-stable distributions and applications
Least lp-norm impulsive noise cancellation with polynomial filters
Signal Processing
Robust techniques for wireless communications in non-gaussian environments
Robust techniques for wireless communications in non-gaussian environments
Median power and median correlation theory
IEEE Transactions on Signal Processing
IEEE Transactions on Signal Processing
IEEE Transactions on Signal Processing
IEEE Transactions on Signal Processing
An ℓp-norm minimization approach to time delay estimation in impulsive noise
Digital Signal Processing
Hi-index | 0.08 |
Sinusoidal frequency estimation has been studied for many years. The MUSIC method represents a class of super-resolution methods based on subspace decomposition. However, the MUSIC method has poor performance in impulsive noise environments due to the prevalence of outliers and very large noise variance. A more robust method called trimmed correlation based-MUSIC (TR-MUSIC) method is proposed in this paper. Through a trimming operation, outliers in the samples participating in the correlation calculation are discarded, yielding a correlation sequence that is closer to the true underlying correlation. The amount of trimming is determined by the Mahalanobis distance in which robust estimates of location and scale are utilized to compensate for outlier effects. Frequency estimation results from the eigendecomposition of the trimmed correlation matrix. In the simulations, we take α-stable noise (α 1) as an example of impulsive noise. The proposed method is very robust and performs better than the conventional MUSIC and other robust methods. Furthermore, it can be applied to real signals as well as complex signals.