*6.1. Scenario 1: The Number of Targets Is Not Time-Varying*

Consider a linear multi-source scenario with two sources. Since the PASTD algorithm cannot track the time-varying target, all the target survival time are 1–50 s. The initial source state are *<sup>x</sup>*<sup>1</sup> <sup>=</sup> [−30;−0.5], and *<sup>x</sup>*<sup>2</sup> <sup>=</sup> [5; 0.5].

Figure 1a shows the RMSE of angles for four algorithms when running 100 MC at α = 2, GSNR = 10 dB, and Figure 1b shows two source trajectories for a single MC. It can be seen from Figure 1a that the UT-MB-FLOM-MUSIC algorithm proposed in this paper is obviously better than the traditional PASTD and has the highest accuracy when the number of targets is constant. It can be seen in Figure 1b that the algorithm can effectively track the target trajectory, while the PASTD algorithm deviates from the real trajectory at several times.

**Figure 1.** Root mean square error (RMSE) of angle under α = 2, *L* = 100 and Generalized Signal to Noise Ratio (GSNR) = 10 dB: (**a**) The RMSE of 100 MC; (**b**) source trajectory of Single MC.

We show the RMSE for tracking the multi-source motion when α = 1.3, GSNR = 10 dB, MC = 100, and *L* = 100 in Figure 2a. It can be seen from Figure 2a that the RMSE of the UT-MB-FLOM-MUSIC algorithm is smaller than that of the other three algorithms. The accuracy of the MB-MUSIC algorithm is significantly reduced in impulse noise, and the PAST algorithm is more accurate than MB-MUSIC. It can be seen from Figure 2b that the MB-MUSIC algorithm cannot effectively track the target trajectory in impulse noise, and the PASTD algorithm also has the problem of inaccurate target tracking. Based on the fact that the above target numbers are unchanged, we will analyze the target time-varying DOA tracking.
