D2D communication plays a vital role in the Internet of Things within the framework of 5G technology. Within D2D communication, resource allocation stands out as a critical research area aimed at enhancing the capabilities of 5G communication for IoT applications. In our proposed work, we begin by establishing a mathematical framework designed to minimize energy consumption in the allocation of frequency bands, utilizing a novel algorithm. Subsequently, we harness the core principles of this algorithm through the application of three distinct optimization methods: (i) the genetic algorithm, (ii) the simulated annealing approach, and (iii) the coalition game theory technique. The analysis and results derived from these methodologies shed light on the individual contributions of these optimization techniques and allow us to conduct a comparative analysis, focusing on convergence patterns. The proposed work provides insights for researchers in domain of the D2D system to facilitate decision-making regarding optimization strategies.