previous up next contents Previous: 10.1.3 Carrier Density Up: 10. Evaluation of Quantum Next: 10.2.1 Coherence Length


10.2 Discussion of Discrepancy

We are aware that several authors have implemented both methods. However, we know of no published comparison which discusses the discrepancy between the two methods in the case of coherent transport.

First we want to stress that the results of the QTBM have been checked by refining the mesh and repeating the simulations. The difference in the simulation results is negligible. In contrast when checking the results from the Wigner simulation by refinement of the grid we run into problems with numerical costs as discussed in Section 8.5.

By our deliberate choice of the mesh size, the short electrode length and the unsuitable discretization we are in the worst case for the finite difference Wigner method. However, the chosen numerical simulation parameters are comparable to those used for many similar problems in the literature.



Subsections previous up next contents Previous: 10.1.3 Carrier Density Up: 10. Evaluation of Quantum Next: 10.2.1 Coherence Length

R. Kosik: Numerical Challenges on the Road to NanoTCAD