The challenge of limited spectrum in the Ultra-High Frequency (UHF) band has resulted in the creation of intelligent Dynamic Spectrum Access (DSA) methods. To address multi-objective spectrum allocation problem in Cognitive Radio Networks (CRNs), this study proposes an Artificial Intelligence Enhanced Genetic Algorithm (AI-GA). Conventional Genetic Algorithms due to static control parameters experience premature convergence; to resolve this, a FeedForward Neural Network (FNN) was incorporated that uses real-time generation fitness data to dynamically adjust the crossover and mutation rates. The optimization framework considered five users and seven frequency bands, aimed to maximize throughput, spectral efficiency and fairness, while minimizing interference. The simulation results showed that the AI-GA achieved a Throughput of 3.78 x 105 bps, Spectrum Efficiency of 5 bps/Hz, Fairness Index of 0.98 and consistent interference minimization of 90 dB. The proposed model outperformed conventional GA and other AI-enhanced methodologies by dynamically balancing exploration and exploitation, providing a reliable solution for fair and effective spectrum management. The results demonstrated that using neural networks to dynamically adjust GA parameters significantly improves Genetic Algorithms’ search capabilities, which makes the proposed AI-GA framework suitable for dynamic interference sensitive and fairness critical cognitive radio environments.
Ekoko et al. (Mon,) studied this question.