next up previous contents
Next: Future Plan Up: SciFi tracker : its Previous: 2nd track finding :   Contents

Goodness of the tracks

The goodness of the tracks are defined in various way and at present, the optimum variable is not found yet. In track common( fgscitrk.ch, various variables exist. You can choose one of them to discriminate the good tracks. The contribution from anybody to find the optimum goodness variable is WELCOME.

The variables defined in fgscitrk.ch are as follows(In addition to these, the $\chi^2$ from Kalman filtering/smoothing method will be prepared soon[3]):

1. fggdtrksci : $\chi^2$ for linear fitting
2. fgtrnovlyrsci : The overlap length of X and Y fragments in terms of layer
3. fgtrnhmlyrsci : The number of overlapped layer of X and Y fragments
4. fgtreffsci : ${\mathrm nint(100\cdot
[N_{hit}(x)+N_{hit}(y)]/[Track \; length \; in \; layer])}$
5. fgtrwgtsci : ${\mathrm fgtrnhmlyrsci\cdot 10^5 + fgtreffsci\cdot
10^2 + [n_{hit}(x)+n_{hit}(y)]}$( recommened by H.Park)
Here, ${\mathrm n_{hit}(x)}$ = the number of hits in X fragment, ${\mathrm n_{hit}(y)}$ = the number of hits in Y fragment


    Example:
                                1 1 1 1 1
    layer#    1 2 3 4 5 6 7 8 9 0 1 2 3 4 .....
    X-fragm       o o o o x o o o o
    Y-fragm     o o o x o o o x o o
                +-----------------+ : Track length ( = 10 )
                  +++++++++++++++++ : Overlap length ( = 9 )
                  + +   +   +   + + : # of overlapped layer ( = 6 )

    fgtrnlyrsci   = 10 : track length
    fgtrnovlyrsci =  9
    fgtrnhmlyrsci =  6
    fgtreffsci = nint[100*{(8+8)/10}] = 160
    fgtrwgtsci = $6*10^5 + 160*100 + (8+8)$ = 616016



K2K experimant account
2000-02-09