AI enabled: a novel IoT-based fake currency detection using millimeter wave (mmWave) sensor

Niaz, Fahim; Zhang, Jian; Khalid, Muhammad; Qureshi, Kashif Naseer; Zheng, Yang; Younas, Muhammad; Imran, Naveed

Abstract

In recent years, the significance of millimeter wave sensors has achieved a paramount role, especially in the non-invasive and ubiquitous analysis of various materials and objects. This paper introduces a novel IoT-based fake currency detection using millimeter wave (mmWave) that leverages machine and deep learning algorithms for the detection of fake and genuine currency based on their distinct sensor reflections. To gather these reflections or signatures from different currency notes, we utilize multiple receiving (RX) antennae of the radar sensor module. Our proposed framework encompasses three different approaches for genuine and fake currency detection, Convolutional Neural Network (CNN), k-nearest Neighbor (k-NN), and Transfer Learning Technique (TLT). After extensive experiments, the proposed framework exhibits impressive accuracy and obtained classification accuracy of 96%, 94%, and 98% for CNN, k-NN, and TLT in distinguishing 10 different currency notes using radar signals.

Citation

Niaz, F., Zhang, J., Khalid, M., Qureshi, K. N., Zheng, Y., Younas, M., & Imran, N. (in press). AI enabled: a novel IoT-based fake currency detection using millimeter wave (mmWave) sensor. Computing, https://doi.org/10.1007/s00607-024-01300-2

• Journal Article Type: Article
• Acceptance Date: May 26, 2024
• Online Publication Date: Jun 27, 2024
• Deposit Date: Jun 28, 2024
• Publicly Available Date: Jun 28, 2025
• Journal: Computing
• Print ISSN: 0010-485X
• Electronic ISSN: 1436-5057
• Publisher: Springer
• Peer Reviewed: Peer Reviewed
• DOI: https://doi.org/10.1007/s00607-024-01300-2
• Keywords: Millimeter wave; Fake currency; Machine learning; Deep learning; Signal processing
• Public URL: https://hull-repository.worktribe.com/output/4720905