Joint Optimization of Detection and Tracking in Adaptive Radar Systems - Advances in Heuristic Signal Processing and Applications - page 135

Fig. 6.10 The performance
of the algorithms in 5 dB
case. At some portions of the
performance trajectory of
heuristic approaches, there
are sharp bends, resulting in a
significant change in the
performance. Therefore, in
general one cannot guarantee
a reliable performance with a
heuristic approach
Initial motivation to formulate the static and dynamic optimization methods
based on HYCA is due to its promise in modeling and therefore improving tran-
sient behavior of the overall system better than MRE. In experimental results, how-
ever, we could not observe this improvement to the extent hoped for. Formulating
and solving the detector threshold optimization problem based on either MRE or
HYCA does not result in a big difference in system performance.
The variation of the average operating P FA values suggested by the algorithms
over the considered SNR range is shown in Fig. 6.11 . A common observation is that,
on the average, optimization algorithm suggested P FA values are all decreased when
SNR increases. This is consistent with the expectation and with the track-before-
Fig. 6.11 ( a ) The average operating P FA values suggested by the static and dynamic optimization
schemes and ( b ) those of the DTOP schemes as a function of SNR