paperKB
coga / coga-kb
Help
Sign in

Chunk #17 — 2. Materials and Methods — 2.2. Preprocessing Algorithms — 2.2.2. Orthogonal Empirical Model Decomposition

Source
Identifying patients with poststroke mild cognitive impairment by pattern recognition of working memory load-related ERP.
Embedded
yes

Text

In particular, once the first IMF is derived, define g 1(t) = g o1(t), which is the smallest temporal scale in f(t). To determine the rest of the IMFs, generate the residue r 1(t) = f(t) − g 1(t). r 1(t) can be treated as the new signal and the EMD decomposing is performed for the second IMF g o2(t). In order to achieve the orthogonal component, g 1(t) has to be subtracted from g o2(t); that is, (4)g2(t)=go2(t)−β21g1(t), where g 2(t) is the second orthogonal IMF and β 21 is the orthogonal parameter between g o2(t) and g 1(t). With the orthogonality between g 2(t) and g 1(t), β 21 can be obtained as (5)∫0Tg1(t)g2(t)=∫0Tg1(t)go2(t)−β21∫0Tg12(t)=0,β21=∫0Tg1(t)go2(t)∫0Tg12(t).