Madhi, Hadi Hussein and Al-Saedi, BAli Mohammed and Albazoone, Mustafa Challoob (2026) SDN-Controlled Adaptive Authenticated Encryption for Resource-Constrained IoT Networks: A Multi-Load Experimental Study. American Journal of Technology Advancement, 3 (10). ISSN 2997-9382

Text AJTA_SDN-Controlled+Adaptive+Authenticated.pdf Download (597kB)

Abstract

Resource-constrained Internet of Things (IoT) deployments require confidentiality and integrity protection, although the most suitable authenticated encryption policy may differ according to traffic load and device pressure. The above study describes the working of an SDN controlled adaptive policy to select a suitable authenticated encryption configuration from a selection of packet encryption algorithms namely AES-128-GCM, AES-256-GCM and ChaCha20 Poly1305. This is a Linux/WSL2, Mininet, Open vSwitch, FAUCET controller, MQTT, and Python implementation. A full factor experiment involving three loads, three counts of devices, three sizes of payload, four policies and 30 repetitions yielded 3,240 completed runs. All of the policies achieved delivery of the 756,000 planned messages with zero aggregate loss, zero authentications failures and zero selection errors. For the entire matrix, the end-to-end median latency for the adaptive policy was 12.015 ms as compared to 12.092 ms for AES-128-GCM, 12.180 ms for AES-256-GCM and 12.027 ms for ChaCha20-Poly1305. The adaptive policy matched the behavior of its chosen fixed baseline at low and medium loads. Under high load, it was shown that the median latency vs AES-128-GCM was reduced by 2.118%. However, this impact was not statistically significant. According to the findings, relying on their own laboratory SDN policy allows for authentic delivery of data while explicitly permitting encryption choice depending on circumstances. We argue that our contribution is a repeatable framework for involving policy-management and evaluation, not that one cipher dominates all workloads.

Item Type:	Article
Uncontrolled Keywords:	software-defined networking, Internet of Things, authenticated encryption, adaptive security policy, AES-GCM, ChaCha20-Poly1305
Subjects:	H Social Sciences
Depositing User:	admin eprints
Date Deposited:	04 Nov 2026 07:28
Last Modified:	04 Nov 2026 07:28
URI:	http://eprints.umsida.ac.id/id/eprint/17413

Actions (login required)

View Item