6.1.3. The Number of Targets: 7

The measurement histories of the seven closely located targets are shown in Figure 12. In this scenario, considering that the number of targets was seven, the simulation was performed by extending the sparsity order to 7 in addition to the 1 and 5 used in the previous scenarios. As in the previous scenarios, multiple targets were gathered in the high density clutter region.

**Figure 12.** Simulation scenario with seven targets.

Figure 13 shows the CTTR over time. As shown in Figure 14, the estimation errors of with the sparsity order *n* = 7 were similar to the result using the true clutter measurement density. As the number of targets increased, increasing the sparsity order implied that better tracking results could be obtained, and the proposed MTT-SCMDE had better tracking performance compared to the existing SCMDE with the same sparsity order. Figure 15 represents the estimated clutter measurement density over time and shows that even with a large number of closely located targets, the proposed method had the best performance of estimating the clutter measurement density. In Table 3, the MTT-SCMDE with *n* = 7 showed more than 80% track retention performance, similar to the case with true clutter measurement density. It showed the best tracking performance among the adaptive estimation algorithms in comparison.

**Figure 13.** Confirmed true track rate.

**Figure 15.** True clutter measurement density and estimated clutter measurement density.


**Table 3.** Track retention statistics for Monte Carlo simulation.
