Using example x9, with 2 parabolic tracks, a varying number of false
detections (0..3 per unit time, uniformly distributed over x,y 0..4),
missed target events, and a penalty for unused measurements in the cost
function, w(K) is the total cost of the best paths for K=1..4 (using
Kn=0.01, S2a=2, L=8):

  w =
       141.12        98.74       85.121       79.517

The cost is decreasing for K>2 due to low-cost missed target events
being choosen.  For example, with K=4, the measurement combinations
associated with the best paths are:

    [combos(:,path_index(:,1))', MM']

     1     2     3     5           5
     1     2     6     6           2
     1     2     6     6           4
     1     2     6     6           3
     1     2     3     6           5
     1     2     6     6           5
     1     2     3     6           3
     1     2     6     6           2
  *  1     2     6     6           5
  *  2     6     6     6           3
  *  2     3     6     6           4
  *  2     5     6     6           5
  *  2     3     6     6           5
     1     2     6     6           4
     1     2     6     6           2
     1     2     6     6           3
     1     2     6     6           5
     1     2     5     6           5
     1     2     3     6           3
     1     2     6     6           5
     1     2     6     6           2

The last column of the above table shows MM, the number of measurements
vs. time.  The first four columns show the measurement indices used in the
four tracks.  In these columns, the number 6 represents a missed target
event detection.  Numbers 1 to 5 represent selections of the measurements,
with 1 and 2 corresponding to real target measurements (except for rows
9-13, marked with *, where the first measurement was replaced with random
uniform values), and 3 to 5 corresponding to false detections.  Row 1 is
for time = 1, row 2 for time = 2, etc.