In many applications extraction of source signals of interest from observed signals is a more feasible approach than simultaneous separation of all the source signals, since the latter often costs lots of computing time and often is not necessary.
However, the increase in power and precision achieved through the use of mixed models comes with considerable costs in terms of computing time.
It is also important to realize that 1 simulation (thus one run for one parameter set) costs about 6-7 hofrs of computing time for one core meaning that it is not feasible to explore the entire parameter space with equal detail.
The method can further be easily modified to incorporate the effects of angle diversity reception with an imaging receiver, but at the cost of several tens times of computing time.
In consideration of computing time cost, we just did cross validation to FPT and the logistic regression method that demonstrated the most similar performance.
In consideration of computing time cost, we chose the minimum support as 2, the minimum hit-rate as 0.5, and the rules below them were considered to be insignificant.
Although the VPG will cost approximately the same amount of computing time as a single PG, its results are fully deterministic and therefore void of statistical sampling errors.
Thus the original electro-mechanical coupling geometric nonlinear problem could be solved iteratively on the macroscopic scale, which will save a tremendous amount of computing time and cost.
These codes require hundreds of hours of computing time that would have cost a fortune to run just a few years ago.
Secondly, Boinc would allow fuller use of computing time.
