4.1.2. Parameter Sensitivity Analysis

As explained in the previous case, the objective value of the optimal scheduling problem depends on changes in the value of a particular parameter. Therefore, we performed a sensitivity analysis to determine how the parameters affect the outcome of the objective function. Three parameters were determined to affect the results. Sensitivity analysis was performed by fixing the remaining parameters while adjusting one target parameter, as shown in Table 5. The target parameter for the first sensitivity analysis was the simultaneous tracking capability (*n*capa) of the radar, as shown in Table 5.


**Table 5.** Parameter setting for sensitivity analysis.

Figure 8 shows how the results varied with the simultaneous tracking capability. As shown in Table 4, the result of the objective function initially increased when *n*capa increased. However, it can be seen that, after a certain level, the result of the objective function was not significantly affected. This tendency was only the result of a given condition, and, when the condition changed, a point that was not affected by the change of *n*capa could be changed. Specifically, if the simultaneous tracking capability of the radar covered the number of targets, the influence of *n*capa would be insignificant. If the number of targets to be tracked were greater than the simultaneous tracking capability of the radar, when the value of *n*capa is high, the objective value would also rise, as shown in Figure 8, until the radars can cover all the targets.

**Figure 8.** Objective value with respect to number of simultaneous tracking capability.

The second parameter that affects the value of the objective function is the minimum tracking assignment time. This parameter is the minimum time required for a radar to track a target, and physically refers to the time it takes for the radar filter system to stabilize the target tracking. Figure 9 shows the objective value according to the change of the minimum tracking assignment time. As can be seen in the figure, smaller minimum tracking assignment times led to higher levels of assignment, but the assignment was not affected after a certain level of minimum tracking assignment time.

**Figure 9.** Objective value with respect to minimum tracking assignment time.

The third parameter for the sensitivity analysis was the time required to hand over the target between networked radars. In fact, this is a simulation parameter, and in a real environment it is very likely to be affected by the physical state of the network and filtering system. However, since it is a parameter that has an important influence on the simulation results, it was selected as a parameter in the sensitivity analysis to grasp its influence and overall tendency. As shown in Figure 10, the objective value decreased linearly as the handover time increased. This result shows that longer handover times led to less efficient overall target tracking. Therefore, a shorter handover time is better. In other words, the network system actually should be constructed so as to minimize the time required for target transmission, as well as the time required for convergence between a transmitted target and a self-detected target.

**Figure 10.** Objective value with respect to handover time.
