Ensemble deep learning approach with hybrid optimisation for enhanced underwater acoustic OFDM communication systems
by S. Pradeep; Subba Reddy Borra; C.V.P.R. Prasad; N. Sreekanth; Sudhakar Kallur
International Journal of Ad Hoc and Ubiquitous Computing (IJAHUC), Vol. 47, No. 2, 2024

Abstract: Underwater acoustic (UWA) communication involves transmitting information through sound waves in aquatic environments, which presents challenges due to signal attenuation, multi-path propagation, and background noise. This research presents a novel approach using ensemble deep learning (EDL) combined with hybrid optimisation for UWA-orthogonal-frequency-division-multiplexing (OFDM) systems. Contrary to the traditional receiver dependence on channel estimation and equalisation for symbol detection, the proposed EDL model directly retrieves transmitted symbols after sufficient training. It employs convolutional neural networks (CNNs), bi-directional long-short-term memory (bi-LSTM) networks, and recurrent neural networks (RNNs) to capture spatial, temporal, and sequence-based dependencies in the signal. To optimise the training process, a hybrid strategy, OppTalO, which integrates driving training-based optimisation (DTBO) and osprey optimisation algorithm (OOA), is utilised. The effectiveness of this EDL approach with hybrid optimisation is assessed across various system parameters: cyclic prefix length and pilot symbol count; and found to have less error rate than existing methods.

Online publication date: Thu, 03-Oct-2024

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
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 Ad Hoc and Ubiquitous Computing (IJAHUC):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your 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