Machine Learning-Assisted Cryptographic Security: A Novel ECC-ANN Framework for MQTT-Based IoT Device Communication

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F60076658%3A12220%2F25%3A43909973" target="_blank" >RIV/60076658:12220/25:43909973 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.mdpi.com/2079-3197/13/10/227" target="_blank" >https://www.mdpi.com/2079-3197/13/10/227</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.3390/computation13100227" target="_blank" >10.3390/computation13100227</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Machine Learning-Assisted Cryptographic Security: A Novel ECC-ANN Framework for MQTT-Based IoT Device Communication

Popis výsledku v původním jazyce

The Internet of Things (IoT) has surfaced as a revolutionary technology, enabling ubiquitous connectivity between devices and revolutionizing traditional lifestyles through smart automation. As IoT systems proliferate, securing device-to-device communication and server–client data exchange has become crucial. This paper presents a novel security framework that integrates elliptic curve cryptography (ECC) with artificial neural networks (ANNs) to enhance the Message Queuing Telemetry Transport (MQTT) protocol. Our study evaluated multiple machine learning algorithms, with ANN demonstrating superior performance in anomaly detection and classification. The hybrid approach not only encrypts communications but also employs the optimized ANN model to detect and classify anomalous traffic patterns. The proposed model demonstrates robust security features, successfully identifying and categorizing various attack types with 90.38% accuracy while maintaining message confidentiality through ECC encryption. Notably, this framework retains the lightweight characteristics essential for IoT devices, making it especially relevant for environments where resources are constrained. To our knowledge, this represents the first implementation of an integrated ECC-ANN approach for securing MQTT-based IoT communications, offering a promising solution for next-generation IoT security requirements.

Název v anglickém jazyce

Machine Learning-Assisted Cryptographic Security: A Novel ECC-ANN Framework for MQTT-Based IoT Device Communication

Popis výsledku anglicky

The Internet of Things (IoT) has surfaced as a revolutionary technology, enabling ubiquitous connectivity between devices and revolutionizing traditional lifestyles through smart automation. As IoT systems proliferate, securing device-to-device communication and server–client data exchange has become crucial. This paper presents a novel security framework that integrates elliptic curve cryptography (ECC) with artificial neural networks (ANNs) to enhance the Message Queuing Telemetry Transport (MQTT) protocol. Our study evaluated multiple machine learning algorithms, with ANN demonstrating superior performance in anomaly detection and classification. The hybrid approach not only encrypts communications but also employs the optimized ANN model to detect and classify anomalous traffic patterns. The proposed model demonstrates robust security features, successfully identifying and categorizing various attack types with 90.38% accuracy while maintaining message confidentiality through ECC encryption. Notably, this framework retains the lightweight characteristics essential for IoT devices, making it especially relevant for environments where resources are constrained. To our knowledge, this represents the first implementation of an integrated ECC-ANN approach for securing MQTT-based IoT communications, offering a promising solution for next-generation IoT security requirements.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach<br>I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2025

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Computation

ISSN

—

e-ISSN

2079-3197

Svazek periodika

13

Číslo periodika v rámci svazku

10

Stát vydavatele periodika

CH - Švýcarská konfederace

Počet stran výsledku

23

Strana od-do

"1–23"

Kód UT WoS článku

001602348000001

EID výsledku v databázi Scopus

—