4.2.1. Experiment One

Aim: The first set of experiments was designed to evaluate the performance of the suggested recovery model in terms of actual execution time as a function of log file size. In general, the cost of recovery is very low for any scheme, since the full log information is stored at the current base station. When the MH travels a grea<sup>t</sup> distance from the initial BS, the difficulty of the recovery method used to locate and transmit the log file rises. Increasing the file size results in an increase in the transmission cost.

Main Results: A As shown in Table 5, the suggested game theory-based recovery model delivers a faster execution time while increasing file size in a variety of mobile settings, depending on the simulation's changing nature.

Discussion: One reason for these results is that, since the proposed model is based on a knowledge base that was developed after the pre-implementation of each chosen recovery protocol in various simulated settings, it chooses the best appropriate recovery protocol for the present circumstance (variation of log file size). As a result, the proposed model requires less time to execute for numerous simulation runs.

#### 4.2.2. Experiment Two

Aim: The second set of experiments evaluated the suggested recovery model's performance as a function of mobility rate in terms of actual execution time. The suggested recovery model is based on the creation of a knowledge base that is created once and used to choose the best suitable procedure based on the reward matrix and dominant equilibrium method. The decision is done here on the basis of the integration of three utility functions that serve as performance and evaluation benchmarks for the candidate protocols.


**Table 4.** Simulation Parameters.

**Table 5.** Execution time (Seconds) as a function of log file size.


Main Results: Figure 5 shows that when mobility rate rises, recovery cost increases. In addition, the current base station has all the log information, lowering the recovery cost. Therecoverycostisgreaterwhenthemobile nodeisrecoveredinthesamebasestation.

Discussion: The results in Figure 5 show that despite their success in limited regions, log managemen<sup>t</sup> and agent-based methods usually suffer from increased implementation time in the long term. On the contrary, the hybrid approach may be more successful in a vast environment than in a small area since it only requires a checkpoint once before MH moves across regions, thus saving time.

**Figure 5.** Execution time as a function of mobility rate.
