Open AccessCombining MUSIC Spatial Sampling and Bootstrap to Estimate Closed Space DOA for Few Samples
Author(s) -
Sidi Mohammed Hadj Irid,
Samir Kameche
Publication year - 2018
Publication title -
algerian journal of signals and systems
Language(s) - English
Resource type - Journals
eISSN - 2676-1548
pISSN - 2543-3792
DOI - 10.51485/ajss.v3i3.68
Subject(s) - sampling (signal processing) , simplicity , computer science , multiple signal classification , algorithm , sample (material) , signal (programming language) , space (punctuation) , pattern recognition (psychology) , signal processing , data mining , artificial intelligence , computer vision , telecommunications , antenna (radio) , filter (signal processing) , operating system , philosophy , chemistry , radar , epistemology , chromatography , programming language
DOA estimation in array processing uses MUSIC (Multiple Signal Classification) algorithm, mainly. It’s the most investigated technique and is very attractive because of its simplicity. However, it meets drawbacks and fails when only very few samples are available and the sources are very close or highly correlated. In these conditions, the problem is more intricate and the detection of targets becomes arduous. To overcome these problems, a new algorithm is developed in this paper. We combine bootstrap technique to increase sample size, spatial sampling and MUSIC method to improve resolution. Through different simulations, the performance and the effectiveness of the proposed approach, referred as spatial Sampling and Bootstrapped technique ‘’SSBoot’’, are demonstrated.