Device-to-device (D2D) communication produces a new dimension in the mobile environment, easing the data exchange process between physically neighboring devices. To achieve an effective utilization of available resources, reduce latency, improve data rates, and increase system capacity, D2D communication utilizes nearby communicating devices. The mobile operator’s action to collect the short-range communications for maintenance of the proximity-based services and improve the performance of networks drives the development of D2D. This paper presents an extensive review of proposed solutions aiming to enhance the security in D2D communication. The main goal of the research is to present an extensive review of the recent advances in various D2D domains such as the discovery process, mode selection schemes, interference management, power control techniques and finally the mode selection for D2D applications for 5G technologies. Additionally, we highlight the open problems and identify the challenges with regard to the D2D communication problem.

Unmanned aerial vehicle (UAV)-assisted multi-access edge computing (MEC) has emerged as a promising solution for providing flexible and low-latency computation services in dynamic and infrastructure-limited environments. However, the performance of such systems is fundamentally constrained by UAV energy limitations, complex path planning requirements, and the need for continuous adaptation to heterogeneous user demands. This paper proposes an integrated reinforcement learning–based framework for joint path planning and wireless charging optimization in multi-UAV-assisted MEC networks. The problem is formulated as a Markov Decision Process, in which UAVs act as learning agents that autonomously optimize their trajectories, charging decisions, and service provisioning policies based on environmental risk, terminal user demand, and energy state information. A cooperative architecture with mobile charging UAVs is introduced to enable in-flight wireless energy replenishment and sustained mission execution. Since the battery-limited UAVs cannot accomplish all tasks and maintain mission flight without recharging, we integrate the mobile charging UAVs (CUAV) into the environment map. Decentralized mobile charging stations are deployed to sustain UAV operations and enable long-term data offloading, where the UAVs can recharge their energy and prolong the mission flight. Furthermore, a QoS-aware sigmoid demand model is incorporated into the reward design to improve service prioritization. Extensive simulation results demonstrate that the proposed framework significantly reduces mission completion time, minimizes unnecessary charging events, maintains stable energy safety margins, and scales effectively with increasing operational area, outperforming baseline strategies in both efficiency and robustness.