Enhancing resource allocation for power sharing in cognitive radio communication networks using ensemble moth-flame optimized dynamic recurrent neural networks

Wireless data transmission networks are radio communication networks (RCN). These networks allow radios as well as phones to communicate, exchange data and calls across radio waves. Broadcasting, emergency services and mobile telecommunications are among the numerous industries that use RCN's flexibility and ease. Our proposed method, Ensemble Moth-Flame Optimized Dynamic Recurrent Neural Network (EMFO-DRNN), addresses the challenges inherent in DNN-based systems, maximizing power distribution, minimizing interference and optimizing spectrum utilization. This solution enables adaptive channel selection in real-time, catering to dynamic conditions and user demands in cognitive radio communication networks. We use the radio wave dataset in our research. We preprocessed using z-score normalization and employed the Fast Fourier Transform (FFT) for feature extraction to develop the EMFO-DRNN approach for improving the allocation of resources for power sharing in Cognitive Radio Communication Networks (CRCN). The EMFO-DRNN method has better performance metrics, such as high accuracy (92%), precision (95%), f1-score (96%) and recall (94%) in enhancing resource allocation for power sharing in CRCN. The proposed approach outperforms traditional methods in resource allocation due to its ensemble nature, which improves stability along with generalization and its Moth-Flame optimization aids in finding optimal solutions promptly.