Convolutional neural network based SNR-modulation classification using constellation diagrams in V2V communication

Abstract In vehicle-to-vehicle (V2V) communication systems, the signal-to-noise ratio (SNR) at the receiver vehicle varies significantly owing to dynamic channel conditions. These variations affect the reliability of data transmission, making it essential to adapt the modulation scheme according to the prevailing SNR. Therefore, accurate SNR-Modulation classification has become a promising solution for enabling adaptive modulation in a vehicular communication environment. To address this problem, we propose a deep learning-based approach for the joint classification of SNR ranges and modulation schemes. The proposed framework employs a pre-defined convolutional neural network (CNN), Visual Geometry Group (VGG19) architecture with additional layers to extract discriminative features from constellation diagrams. A CNN-based framework for SNR-Modulation classification was developed, considering digital modulation techniques such as binary phase-shift keying (BPSK) and quadrature PSK (QPSK) under SNR levels ranging from 0 to 10 dB. The proposed CNN model was built using a dataset comprising 8400 constellation diagrams that incorporated various amplitude and phase imbalances to replicate the real-world conditions. The model classified four distinct SNR-Modulation classes: 0 ≤ SNR < 5 dB:BPSK (Class 1), 0 ≤ SNR < 5 dB:QPSK (Class 2), 5 ≤ SNR ≤ 10 dB:BPSK (Class 3), and 5 ≤ SNR ≤ 10 dB:QPSK (Class 4). The proposed model achieved an accuracy of 98.21%, with precision, recall, and F1-score values of 98% and an AUC of 98%.

Authors

Institutions

Publication Details

Journal
Physica Scripta
Published
2026-11-17
DOI
https://doi.org/10.1088/1402-4896/aec336
Primary Topic
Wireless Signal Modulation Classification
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Convolutional neural network based SNR-modulation classification using constellation diagrams in V2V communication

M. V. S. Sairam, Jerzy Ryszard Szymanski, Mithileysh Sathiyanarayanan, Marta Żurek-Mortka
1 links
Physica Scripta
Wireless Signal Modulation Classification
article

Convolutional neural network based SNR-modulation classification using constellation diagrams in V2V communication

M. V. S. Sairam, Jerzy Ryszard Szymanski, Mithileysh Sathiyanarayanan, Marta Żurek-Mortka
article en

Abstract

Abstract In vehicle-to-vehicle (V2V) communication systems, the signal-to-noise ratio (SNR) at the receiver vehicle varies significantly owing to dynamic channel conditions. These variations affect the reliability of data transmission, making it essential to adapt the modulation scheme according to the prevailing SNR. Therefore, accurate SNR-Modulation classification has become a promising solution for enabling adaptive modulation in a vehicular communication environment. To address this problem, we propose a deep learning-based approach for the joint classification of SNR ranges and modulation schemes. The proposed framework employs a pre-defined convolutional neural network (CNN), Visual Geometry Group (VGG19) architecture with additional layers to extract discriminative features from constellation diagrams. A CNN-based framework for SNR-Modulation classification was developed, considering digital modulation techniques such as binary phase-shift keying (BPSK) and quadrature PSK (QPSK) under SNR levels ranging from 0 to 10 dB. The proposed CNN model was built using a dataset comprising 8400 constellation diagrams that incorporated various amplitude and phase imbalances to replicate the real-world conditions. The model classified four distinct SNR-Modulation classes: 0 ≤ SNR < 5 dB:BPSK (Class 1), 0 ≤ SNR < 5 dB:QPSK (Class 2), 5 ≤ SNR ≤ 10 dB:BPSK (Class 3), and 5 ≤ SNR ≤ 10 dB:QPSK (Class 4). The proposed model achieved an accuracy of 98.21%, with precision, recall, and F1-score values of 98% and an AUC of 98%.

Physica Scripta
University of London (GB), Kazimierz Pułaski University of Technology and Humanities in Radom (PL), Sopot University of Applied Science (PL)
Industry, innovation and infrastructure
Wireless Signal Modulation Classification
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon