Advancing Machine Learning Strategies for Power Consumption-Based IoT Botnet Detection
Almustapha A. Wakili, Saugat Guni, Sabbir Ahmed Khan, Wei Yu, Woosub Jung · 2025
Read Advancing Machine Learning Strategies for Power Consumption-Based IoT Botnet Detection by Almustapha A. Wakili, Saugat Guni, Sabbir Ahmed Khan, Wei Yu, Woosub Jung (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12736941.txt.
Description
Almustapha Wakili and four co-authors benchmark learning models that spot hacked smart-home devices by watching their power use, in a 2025 study. Swarms of infected gadgets form botnets, and network watchers struggle when traffic is encrypted or sparse. Power draw offers a side channel that betrays a device's state.
The team compared three model families under one testing setup: classic methods, deep networks, and hybrids mixing the two. They used an existing dataset plus a new three-class one they curated, testing on single devices and across devices. A random forest won on single devices, reaching 99.43 percent accuracy on a voice assistant. A hybrid network paired the best accuracy with high speed across devices. The authors present the results as practical guidance for choosing a detector under accuracy, delay, and throughput limits.
Catalog record
| CREATOR | Almustapha A. Wakili, Saugat Guni, Sabbir Ahmed Khan, Wei Yu, Woosub Jung |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 72 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |