A lightweight hybrid attention network with multi-scale feature integration for intelligent recognition of underwater acoustic targets

Nahid-Al Mahmud, Tao Zhang, Yasir Iqbal, Farhana Bari Sumona, Anjum Iqbal, Yanzhang Geng, Wajid Khan, Basem Abu Zneid, Ievgen Zaitsev · 2026

Read A lightweight hybrid attention network with multi-scale feature integration for intelligent recognition of underwater acoustic targets by Nahid-Al Mahmud, Tao Zhang, Yasir Iqbal, Farhana Bari Sumona, Anjum Iqbal, Yanzhang Geng, Wajid Khan, Basem Abu Zneid, Ievgen Zaitsev (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13212563.txt.

Description

Nahid-Al Mahmud and colleagues introduce in 2026 a compact neural network for recognising underwater sounds, such as ships picked up by sonar. Noise, fading signals, and shifting water conditions make the task hard, while many accurate models are too heavy for real-time use.

Their design mixes efficient local feature extraction with attention modules that track longer patterns, joining the two through an adaptive fusion step. Tested on two public ship-sound collections, the system scored above 98 percent accuracy on both. It did so with only about half a million parameters and very low delay. That balance could suit monitoring equipment with limited computing power.

Catalog record

CREATORNahid-Al Mahmud, Tao Zhang, Yasir Iqbal, Farhana Bari Sumona, Anjum Iqbal, Yanzhang Geng, Wajid Khan, Basem Abu Zneid, Ievgen Zaitsev
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE68 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13212563

Browse the full library · About the Tower of Babel