Hybrid deep learning-AES framework for securing smart greenhouse sensor data under network sabotage scenarios
Abstract
Smart greenhouse systems for orchid acclimatization depend on continuous internet of things (IoT)-based environmental monitoring, but their bidirectional communication channels remain exposed to network sabotage threats such as packet injection, man-in-the-middle interception, and brute force attacks. This study introduces a hybrid deep learning-advanced encryption standard (DL-AES) framework that strengthens traditional AES encryption by incorporating an adaptive DL–based nonlinear ciphertext transformation layer applied after the AES encryption stage. This novelty enables cryptographic parameters to evolve in real time, increasing unpredictability and resilience against both cryptanalytic and network-level attacks while maintaining low computational overhead suitable for resource-constrained IoT nodes. Experimental results under simulated brute force and high-intensity adversarial traffic showed no successful breach was observed during the experimental evaluation, stable encryption latency (3.42–3.50 ms), consistent central processing unit (CPU) usage (~28–30%), and negligible memory overhead, demonstrating both security robustness and operational reliability. The key contribution lies in delivering a lightweight, scalable encryption solution that ensures end-to-end confidentiality and integrity of smart greenhouse sensor data under adversarial network conditions.
Keywords
Dynamic key scheduling; Hybrid deep learning–advanced encryption standard; Internet of things data confidentiality; Network sabotage resilience; Smart greenhouse security
Full Text:
PDFDOI: https://doi.org/10.11591/eei.v15i6.11332
Refbacks
- There are currently no refbacks.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Bulletin of Electrical Engineering and Informatics (BEEI)
ISSN: 2089-3191
,
e-ISSN: 2302-9285
This journal is published by the
Institute of Advanced Engineering and Science (IAES)
.