From: Iterative unbiased FIR state estimation: a review of algorithms
Stage
Given:
K, N, m=n−N+1,
s=m+K−1, m+K⩽l⩽n.
Set:
x ̂ s by (69) and G s by (70)
Update:
F l by (100), H l by (101),
G l = [ H l T H l + ( F l G l − 1 F l T ) − 1 ] − 1 ,
K l = G l H l T ,
x ̂ l = f l ( x ̂ l − 1 )+ K l [ z l − h l ( x ̂ l − )].
Remark:
Use the estimate when l=n.