I had no idea this was a technique! From a British Antarctic Survey report on these devices[1]:

    Accuracy
    In field trials at Bird Island, South Georgia, the mean
    error (great-circle distance) in position estimation
    of static devices was 85 km, with standard deviations
    in latitudinal and  longitudinal errors of 0.61° and
    0.99°, respectively (BAS, unpublished data).

[1] http://polaris.nipr.ac.jp/~penguin/oogataHP/pdfarticles/23p2...

That's a really clever solution to a unique set of domain specific constraints!