Title: SPNet: Siamese pyramid network for malicious node detection in mobile networks with blockchain-based authentication scheme

Authors: Gotte Ranjith Kumar; K. Suresh Babu

Addresses: Department of Computer Science and Engineering, Jawaharlal Nehru Technological University, Kukatpally, Hyderabad, Telangana, 500085, India; School of CS&AI, SR University, Warangal – 506371, Telangana, India ' Department of Computer Science and Engineering, Jawaharlal Nehru Technological University, Kukatpally, Hyderabad, Telangana, 500085, India

Abstract: This paper proposes an approach for malicious node detection in mobile networks utilising hybrid networks named Siamese pyramid network (SPNet). The entities involved in this potent framework are the user, network manager, mobile service provider (MSP) and blockchain (BC). The malicious nodes are detected by employing SPNet, which is the integration of the Siamese convolutional neural network (SCNN) and deep pyramidal residual network (PyramidNet). After that, the encryption, hashing function, one-time password (OTP), and X-OR functions are employed. In this research, the evaluation is done based on the number of users and transaction. The developed SPNet model achieved the minimum communication time, memory and time cost of 37.05 ms, 434.36 bytes and 46.78 sec, for user 20. Additionally, the proposed model attained the maximum accuracy, sensitivity, and specificity of 90.880%, 90.090% and 90.520%, based on transaction 100.

Keywords: mobile network; malicious nodes; authentication; PyramidNet; Siamese convolutional neural network; SCNN.

DOI: 10.1504/IJSCC.2026.152984

International Journal of Systems, Control and Communications, 2026 Vol.17 No.2, pp.143 - 175

Received: 20 Dec 2024
Accepted: 13 Jun 2025
Published online: 17 Apr 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article