Kalman Filter Equations
§
Measurement update step at time
k
:
§
§
§
§
Dynamic propagation step from times
k
to
k+1
:
§
§
§
§
§
Put multiple months of flux in state vector
x
k
, method becomes
effectively a fixed-lag Kalman smoother
Error
Time
2
1
1
1
1
1
1
2
2
2
2
2
F
=
tangent
linear model
Great for time-dependent problems.
But what if