Introduction - If you have any usage issues, please Google them yourself
A new approach to improved filter design is presented for
the radar tracking problem. An idealized version of the
extended Kalman filter, which is unrealizable in practice,
is constructed using a universal linearization concept, and
then it is utilized for the development of a practical tracking
filter. The resulting filter tunes the measurement error variance
in an adaptive manner to account for the measurement
nonlinearities effectively.