Comparative Analysis of Attention Mechanisms for Automatic Modulation Classification in Radio Frequency Signals
Ferhat Ozgur Catak 1 , Murat Kuzlu 2 , Umit Cali 3 · 2025
Read Comparative Analysis of Attention Mechanisms for Automatic Modulation Classification in Radio Frequency Signals by Ferhat Ozgur Catak 1 , Murat Kuzlu 2 , Umit Cali 3 (2025) 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/arxiv-2508-09996.txt.
Description
Ferhat Ozgur Catak, Murat Kuzlu, and Umit Cali compared attention designs for reading radio signals in a 2025 paper. Radios can encode data in several modulation schemes, and a receiver that can tell them apart on its own is key for smart radio systems. The team paired neural networks with three flavors of attention, the mechanism that lets a model weigh which parts of a signal matter.
Their hybrid model was tested on a standard public set of radio recordings. The plain multi-head attention scored best, at about eighty-five percent correct. Two leaner designs, causal and sparse attention, ran far faster, cutting inference time by about four-fifths and three-quarters, while staying above eighty-four percent. Each design also favored different modulation types. The trade-off the paper maps is practical: a small accuracy cost buys the speed that live radio work demands.
Catalog record
| CREATOR | Ferhat Ozgur Catak 1 , Murat Kuzlu 2 , Umit Cali 3 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 23 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-09996