Joint Task Offloading and Resource Orchestration in 6G SAGINs: A Hybrid Quantum-Inspired Meta-Heuristic Approach
Achieving global ubiquitous connectivity in 6G wireless systems necessitates the seamless integration of terrestrial networks with Space-Air-Ground Integrated Networks (SAGINs). However, joint task offloading and resource orchestration in such heterogeneous architectures involve large-scale combinatorial optimization challenges. This paper proposes a Hybrid Quantum-Inspired Meta-Heuristic (QIMH) approach to minimize system costs, effectively balancing latency and energy consumption. By utilizing quantum superposition principles and adaptive quantum rotation gates, the proposed algorithm enhances explorationexploitation capabilities. Furthermore, the optimization framework is implemented using vectorized tensor operations on an NVIDIA RTX 4090 GPU. Experimental results demonstrate that the proposed Hybrid QGA outperforms Particle Swarm Optimization (PSO) with up to 10.96% cost reduction. Notably, the algorithm maintains a near-constant execution time of approximately 0.25 seconds for up to 1000 users, proving its real-time feasibility for large-scale 6G non-terrestrial network applications.