Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{208667,
        author = {Nicchanametla Bindu Maneesha and Dr. I. Kullayamma},
        title = {CNN–Attention–XGBoost Based Cognitive Radio Spectrum Sensing and Modulation Classification with Explainable AI},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {4},
        pages = {2442-2449},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=208667},
        abstract = {To enable efficient utilization of the radio-frequency spectrum, especially in Cognitive Radio systems, spectrum awareness is crucial for the ability to dynamically utilize the available frequency band. In this paper, a novel wireless signal classification approach, which is based on CNN–Attention–XGBoost, is proposed. The proposed approach leverages a 1-D Convolutional Neural Network to learn discriminative representations directly from a 2-channels I/Q signal samples with an attention mechanism for feature refinement. The deep features are classified with XGBoost in 5 signal classes: BPSK, QPSK, QAM16, QAM64 and FSK. In order to gain more insight into features contribution and necessary signal region, SHAP and Grad-CAM are added to improve the model interpretability. This implemented system was tested with 500,000 signal samples, split into the train and test set in 80:20 ratio using stratified sampling method. The overall accuracy of the experimental results was found to be 92% with the class-wise ROC-AUC ranging from 0.94 to 0.97. Practical examples of uploading streams, categorizing them, assessing confidence, and visualizing streams further illustrate an interactive Streamlit interface.},
        keywords = {Cognitive Radio; Spectrum Sensing; CNN; XGBoost; Attention Mechanism; Explainable AI; Modulation Classification.},
        month = {September},
        }