Resource allocation for device-to-device (D2D) communications is a critical challenge in 5G and future wireless networks, as it requires balancing energy efficiency, dynamic traffic demand, and interference management. This paper proposes a hybrid framework that integrates metaheuristic optimization with machine learning-based prediction. An enhanced support vector machine combined with K-nearest neighbour (ESVKNN) is employed to forecast resource demand and reduce allocation uncertainty. To optimise channel allocation and transmit power, a hybrid Prairie Dog–Kepler optimization (HPDKO) approach is introduced. The method converts constrained optimisation problems into unconstrained ones through a self-adaptive penalty mechanism. Simulation results show that the proposed framework consistently outperforms several state-of-the-art baseline methods, including the Kepler optimization algorithm (KOA) and Prairie Dog optimization (PDO). In particular, HPDKO–ESVKNN achieves throughput improvements of 5–15%, energy efficiency gains of 2–5%, and fairness index values above 0.97, while maintaining superior signal-to-interference-plus-noise ratio (SINR) across diverse network conditions. These results demonstrate the robustness and scalability of the proposed approach, indicating its potential as an effective solution for efficient and reliable D2D communication in next-generation wireless networks.
Vinothkumar et al. (Mon,) studied this question.