Coati optimisation algorithm based hyperparameter tuned attention B-BiLTF model for spectrum prediction Online publication date: Fri, 26-Jul-2024
by Avani Vithalani
International Journal of Intelligent Engineering Informatics (IJIEI), Vol. 12, No. 3, 2024
Abstract: The burgeoning demand for spectrum in the 5G era and internet of things underscores the critical need for accurate spectrum prediction models. Existing methods grapple with challenges, particularly the inability to capture frequency band features at specific times. This research introduces the coati optimisation algorithm-based attention B-BiLTF model, addressing the pervasive issue of gradient disappearance in spectrum prediction. Combining bidirectional long short-term memory (BiLSTM) and backpropagation (BP) neural networks, the B-BiLTF algorithm achieves enhanced convergence speed and overall prediction accuracy. The attention B-BiLTF mechanism mitigates the impact of sequence length changes on performance. Leveraging the coati optimisation algorithm ensures systematic hyperparameter optimisation, outperforming existing approaches across diverse signal-to-noise ratio conditions and sequence lengths. Experimental results on RML2016.10a dataset demonstrate superior accuracy, RMSE, MAE, and Haversine distance, affirming the model's reliability and robustness modulation modes and signal to noise ratio (SNR) levels. This research contributes an efficient approach to spectrum prediction, advancing cognitive radio systems, and optimising spectrum utilisation.
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Intelligent Engineering Informatics (IJIEI):
Login with your Inderscience username and password:
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.
If you still need assistance, please email subs@inderscience.com