Abhay Bansal

The evolution of sixth‐generation (6G) wireless networks demands ultra‐reliable low‐latency communication (URLLC), massive connectivity, and high‐capacity data transmission in highly dynamic environments. Space–Air–Ground Integrated Networks (SAGINs) have emerged as a promising architecture by seamlessly integrating satellites, unmanned aerial vehicles (UAVs), and terrestrial infrastructure to provide ubiquitous connectivity. However, stochastic traffic arrivals, UAV mobility, time‐varying wireless channels, and the coexistence of enhanced Mobile Broadband (eMBB) and URLLC services make traffic offloading and resource allocation highly challenging. These factors transform the optimization task into a stochastic mixed‐integer nonlinear programming (MINLP) problem. To address this challenge, this paper proposes a Quantum Federated Reinforcement Learning (QFRL)‐based traffic offloading framework for RSMA‐enabled SAGINs. The optimization problem is formulated as a constrained Markov decision process (CMDP), allowing distributed small cells to jointly optimize traffic offloading ratios, bandwidth allocation, RSMA power distribution, and UAV trajectory planning while satisfying stringent delay and reliability requirements. A variational quantum circuit (VQC)‐based actor‐critic architecture is developed to improve learning efficiency and policy representation in high‐dimensional continuous action spaces. In addition, a federated aggregation mechanism enables privacy‐preserving distributed learning and scalable coordination across the space, air, and ground segments. The proposed framework employs temporal‐difference learning and parameter‐shift gradient optimization to ensure stable convergence under stochastic network dynamics. Simulation results demonstrated that the proposed QFRL framework reduces traffic dropping probability by 28%–35%, decreases URLLC delay by 22%–30%, improves network availability by 18%–25%, and enhances traffic offloading efficiency by 20%–27% compared with Differentiated Federated Soft Actor‐Critic (DFSAC), Double Q‐Learning delay sensitive replay memory (DSRPM), and Nash Equilibrium Iteration Offloading (NEIO‐G) schemes, respectively.

!Vehicle-to-everything (V2X) communication has emerged as a key paradigm that has garnered proliferating interest by researchers providing seamless connectivity among vehicles, roadside infrastructures and aerial platforms. However, its dynamic nature, privacy and poor V2X channel quality can be a challenge in offloading tasks and resource optimization. This paper proposes a data-preserving, energy-efficient task offloading strategy in an aerial autonomous vehicle (AAV)-assisted V2X network by deploying multiple intelligent reflecting surfaces (IRSs) devices and by jointly optimizing the IRS reflection coefficient matrix, task vehicle offloading mode, and beamforming vector. It helps in minimizing the average offloading delay in an energy-efficient manner. The article proposes a feasible joint optimization algorithm based on twin-delayed multi-agent deep deterministic policy gradient with a federated learning framework (MAF-TD3), designing a practical partially observable Markov Decision Process (POMDP). Diverse, distributed agents, enabled by federated learning (FL), can collaborate to generate secure, energy-efficient global models without exchanging raw data, thereby protecting privacy and reducing communication overhead. Whereas the twin-delayed deep deterministic policy gradient (TD3) provides faster learning for continuous actions under partially observable dynamic scenarios. Unlike centralized TD3, MAF-TD3 enables scalable and privacy-preserving learning by aggregating only actor policies via federated coordination, while retaining local twin critics to ensure stability under partial observability in dense multi-IRS V2X networks. Numerical simulation results show a reduction in the latency by approximately 61.5%, 48.7%, and 28.5%, respectively, when compared to other MARL-based conventional DRL algorithms MAF-DDPG, MATD3 and MADDPG. It also minimizes the average energy consumption by 52%, 39.5%, and 28%, providing a performance boost in convergence, highlighting the importance of IRS and V2X task offloading in improving system performance.