| Autor: |
Borijindargoon, Narong, Boon Poh Ng |
| Předmět: |
|
| Zdroj: |
Progress in Electromagnetics Research Letters; 2019, Vol. 81, p21-28, 8p |
| Abstrakt: |
An algorithm called MUSIC-like algorithm was originally proposed as an alternative method to theMUltiple SIgnal Classification (MUSIC) algorithm in order to circumvent requirement on subspace segregation. The relaxation parameter β, which was introduced into the formulation of the MUSIClike algorithm, has enabled the algorithm to achieve high resolution performance comparable to the MUSIC algorithm without requiring explicit estimation of the signal and noise subspaces. An adaptive framework for the MUSIC-like algorithm was later developed under the α-stable distributed noise environment. In spite of great improvement on target's resolvability performance, a trade-off between such improvement and the estimation bias is inherent. In this letter, two novel directional adaptive β-selection methods for MUSIC-like algorithm under α-stable distributed noise are proposed. The proposed methods aim at reducing estimation bias and noise sensitivity which are inherent in prior adaptive β framework. Simulation results highlight noticeable reduction in the estimation bias as well as the noise sensitivity of the proposed methods without excessive compromise on target's resolvability performance compared with the original adaptive β framework. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
