Difference between revisions of "Expmint.m"
(Normalise sequential blank lines) |
m (Spot-update function page against current Spinach source) |
||
| Line 22: | Line 22: | ||
==Notes== | ==Notes== | ||
The auxiliary matrix method is massively faster than either commutator series or diagonalisation. | The auxiliary matrix method is massively faster than either commutator series or diagonalisation. | ||
| + | |||
| + | This is the most memory-intensive stage in many calculations; memory recycling is aggressive. | ||
==See also== | ==See also== | ||
Revision as of 15:49, 5 June 2026
Computes matrix exponential integrals of the following general type:
Integrate[expm(-i*A*t)*B*expm(i*C*t),{t,0,T}]
Matrix A must be Hermitian. For further info see the paper by Charles van Loan (http://dx.doi.org/10.1109/TAC.1978.1101743).
Contents
Syntax
R=expmint(spin_system,A,B,C,T)
Arguments
A,B,C - the three matrices involved in the integral
T - integration time
Outputs
R - the resulting integral
Notes
The auxiliary matrix method is massively faster than either commutator series or diagonalisation.
This is the most memory-intensive stage in many calculations; memory recycling is aggressive.
See also
intrep.m, propagator.m, average.m, rotframe.m, dirdiff.m
Version 2.2, authors: Luke Edwards, David Goodwin, Ilya Kuprov
