GA-optimized JT-CoMP F-NOMA in HetNets: Adaptive resource allocation with imperfect SIC

Power domain non-orthogonal multiple access has emerged as a critical solution to address the spectral inefficiency of orthogonal multiple access and to meet the demands of next-generation multiple access in 5G and beyond 5G and 6G internet of things networks. Coordinated multi-point-non-orthogonal multiple access can effectively manage interference and enhance spectral efficiency in densely populated heterogeneous networks particularly for cell-edge users. However, relying solely on non-orthogonal multiple access for all users across varying and practical channel conditions, especially under imperfect successive interference cancellation, is often impractical. An adaptable approach is required to serve a broader range of users. To address this need, we propose a joint transmission-coordinated multi-point enabled flexible-non-orthogonal multiple access framework that maximizes the network's sum-rate and dynamically optimizes power allocation. Our approach mitigates challenges posed by paired non-orthogonal multiple access users who fail to meet minimum quality of service requirements due to practical issues like successive interference cancellation imperfections or by presence of unpaired users. The study focuses on jointly optimizing Coordinated multi-point and non-coordinated multi-point user selection, non-orthogonal multiple access user pairing, orthogonal multiple access/non-orthogonal multiple access mode selection, and power allocation using a Genetic algorithm to enhance sum-rate performance across the network. We introduce a Genetic Algorithm-based flexible non-orthogonal multiple access with adaptive quality of service and orthogonal multiple access switching algorithm to solve this joint optimization problem, ensuring reliable service provision and reducing disruptions caused by suboptimal channel conditions and interference. By combining joint transmission-coordinated multi-point with flexible-non-orthogonal multiple access, our proposed framework significantly boosts network capacity and spectral efficiency while addressing key challenges in next-generation multiple access-enabled internet of things networks. Furthermore, we evaluate the impact of coordinated multi-point user detection limitations and imperfect successive interference cancellation on non-orthogonal multiple access performance. We conduct an asymptotic analysis of user rates under extreme signal-to-noise ratio conditions to evaluate the long-term performance of the proposed framework. Additionally, we analyze the computational complexity of the proposed Genetic Algorithm-based solution, which offers significant improvements over the exponentially increasing complexity of exhaustive search methods. Simulation results demonstrate that Genetic Algorithm-based flexible non-orthogonal multiple access with adaptive quality of service and orthogonal multiple access switching algorithm outperforms traditional coordinated multi-point non-orthogonal multiple access and coordinated multi-point orthogonal multiple access approaches, achieving sum-rate improvements of up to 20% and 30% respectively under diverse network conditions.