2.2.2. Target Priority

The target importance needs to be assessed using a priority-based metric that reflects the relative distance and remaining time between the radar and the target. Therefore, even if the same target is tracked by two or more radars, the target importance is different for each radar.

In this paper, since it is difficult to quantify the degree of threat according to the type of target, the target importance is calculated using the time remaining until the target hits the surface and the distance between the radar and the target. Here, the impact time of the target reflects the urgency to engage the target. Thus, it sets a higher priority when the remaining time becomes smaller. For a fast target, the remaining time will decrease very quickly, and thus the increasing rate of the tracking value over time will be higher than those of other targets. The distance between the target and the radar is a factor that reflects the Signal-to-Noise Ratio (SNR) and hence the expected tracking performance. Therefore, the target priority used in this study reflects the expected performance and urgency. The tracking value (*vt*), determined by remaining time to impact (*τ*) and the distance from the radar (*dist*), is calculated as follows [25].

$$v\_t = \left(1 - \frac{1}{1 + e^{-\left(\tau - \eta\_0\right)/a\_{\tau}}}\right) + \left(1 - \frac{\beta\_{dist}}{1 + e^{-\left(dlist - dlist\_0\right)/a\_{\text{dist}}}}\right) \tag{1}$$

where *τ*0, *ατ*, *dist*0, *αdist*, and *βdist* are parameters to determine the shape of the sigmoid function. *τ*<sup>0</sup> = 100, *ατ* = 15, *dist*<sup>0</sup> = 500, *αdist* = 100, and *βdist* = 0.8 are used in this work.

Equation (1) decreases non-linearly (sigmoid) as the distance increases and reflects the change in average tracking performance according to SNR when the target is tracked in a single radar with a fixed resource.
