Next: Statistical signification of
Up: THE LSL ALGORITHM
Previous: Recursions: time updating
The weighting consists of using a tapering coefficient
to minimize the weighted norm of the residuals:
It forces the algorithm to forget the ``old data'' in the minimization
problem: it extends the adaptive properties of the algorithm.
We can use the recursion results derived up to now, simply adapted to
another choice of scalar product:
Actually, only the covariances and correlation recursions are modified:
| |
(5) |
| |
Next: Statistical signification of
Up: THE LSL ALGORITHM
Previous: Recursions: time updating
Stanford Exploration Project
1/13/1998