Adaptive Modulation and Coding Scheme Based on Multi-source Indicator Fusion

Underwater acoustic communication systems play a crucial role in ocean development and management. However, the complexity and variability of ocean communication channels pose challenges for ensuring reliable and stable communication quality using single-modulation coding systems. Consequently, the...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:IEEE International Conference on Electronic Information and Communication Technology (Online) s. 129 - 134
Hlavní autori: Sun, Zheng, Shi, Wei, Wang, Jingjing, Niu, Qiuna
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 21.07.2023
Predmet:
adaptive modulation and coding
Adaptive systems
canonical correlation analysis
Correlation
Encoding
Feature extraction
feature fusion
hydroacoustic communication
Neural networks
Oceans
Throughput
ISSN:2836-7782
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Underwater acoustic communication systems play a crucial role in ocean development and management. However, the complexity and variability of ocean communication channels pose challenges for ensuring reliable and stable communication quality using single-modulation coding systems. Consequently, the effectiveness of underwater acoustic communication technology is significantly impacted. To enhance the throughput of ocean communication systems, researchers have explored and implemented adaptive modulation and coding (AMC) techniques in underwater acoustic communication systems. However, traditional AMC methods that rely on fixed thresholds suffer from issues such as inaccurate channel state assessment and imprecise Modulation and Coding Scheme (MCS) selection. In this paper, we propose the integration of neural network algorithms into adaptive modulation and coding for ocean channels. Specifically, we present a novel approach called the multi-source index fusion underwater adaptive modulation and coding method, which leverages neural networks. We also devise a neural network-based solution to address the Canonical Correlation Analysis (CCA) feature matrix, thereby enhancing system classification accuracy and throughput. Experimental results demonstrate that our proposed adaptive modulation and coding method, based on multi-source index fusion, outperforms the traditional method that relies on fixed threshold MCS judgment. It achieves a significant increase in system throughput.
ISSN:2836-7782
DOI:10.1109/ICEICT57916.2023.10245368