| Title |
Multi-target SCORE Algorithm using Dominant Mode Prediction |
| Authors |
김지현(Ji-Hyeon Kim) ; 서영광(Young-Kwang Seo) ; 권순영(Soon-Young Kwon) ; 박진오(Jin-Oh Park) ; 강현진(Hyun Jin Kang) ; 김재윤(Jae Yun Kim) ; 문병호(Byung Ho Mun) ; 김형남(Hyoung-Nam Kim) |
| DOI |
https://doi.org/10.5573/ieie.2019.56.4.79 |
| Keywords |
Cyclostationarity ; MT-SCORE ; DMP algorithm |
| Abstract |
A self-coherence restoral (SCORE) algorithm is used in extracting a signal of interest (SOI) by utilizing the cyclostationarity of the signal even in the absence of a prior information on the incidence angle, the structure of the array antenna, and the statistical properties of the noise when multi-interference signals arrive in one channel. In the SCORE algorithm, a cross-SCORE algorithm that updates both the beamforming vector and the control vector shows better performance than the least-squares (LS)-SCORE algorithm. In this paper, we propose a multi-target (MT)-SCORE algorithm to separate multiple co-channel signals based on the cross-SCORE algorithm. The proposed MT-SCORE algorithm constructs cross-SCORE in parallel to separate multi-interference signals. In addition, we guarantee the convergence of each SCORE output to different signals and improve the convergence speed by estimating the initial beamforming vector with the dominant mode prediction (DMP) algorithm. We simulate the co-channel interference environment in which a new signal appears at any instants. By using the initial beamforming vector in the signal direction of arrival and the weight vector gain change in the SOI's incidence angle, the proposed algorithm has better performance of interference elimination and fast convergence than the conventional cross-SCORE algorithm. |