Scholarlink

The increasing reliance on Internet of Things (IoT) infrastructures has ampli-fied the need for robust intrusion detection and prevention systems (IDS/IPS). Artificial intelligence (AI), particularly deep learning, has been widely adopted to enhance detection accuracy and adaptiveness in these environments. Existing AI-based approaches demonstrate strong perfor-mance, yet challenges remain in terms of scalability, resilience to novel at-tacks, and the handling of highly complex feature spaces. Quantum Neural Networks (QNNs) represent a new computational paradigm with the poten-tial to address some of these limitations by exploiting properties such as superposition and entanglement to capture correlations beyond classical models. This paper examines the current state of AI-driven IDS/IPS for IoT networks and analyzes the potential impact of introducing QNN capabili-ties. The discussion considers architectural implications, possible perfor-mance gains, and the constraints of present-day noisy intermediate-scale quantum (NISQ) devices. Rather than proposing a concrete hybrid imple-mentation, the paper focuses on mapping how QNN concepts could com-plement or extend existing AI-based frameworks for IoT security. The anal-ysis highlights both the opportunities and the open challenges of integrat-ing quantum-inspired methods into intrusion detection, contributing to the broader discourse on next-generation approaches to securing IoT systems.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Hybrid Deep Learning and QNN for Detecting Attacks Within IoT Networks

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Teodor Cervinski,
  • Cristian Toma,
  • Marius Popa,
  • Catalin Boja,
  • Claudiu Brandas,
  • Andrei Cazacu

摘要

The increasing reliance on Internet of Things (IoT) infrastructures has ampli-fied the need for robust intrusion detection and prevention systems (IDS/IPS). Artificial intelligence (AI), particularly deep learning, has been widely adopted to enhance detection accuracy and adaptiveness in these environments. Existing AI-based approaches demonstrate strong perfor-mance, yet challenges remain in terms of scalability, resilience to novel at-tacks, and the handling of highly complex feature spaces. Quantum Neural Networks (QNNs) represent a new computational paradigm with the poten-tial to address some of these limitations by exploiting properties such as superposition and entanglement to capture correlations beyond classical models. This paper examines the current state of AI-driven IDS/IPS for IoT networks and analyzes the potential impact of introducing QNN capabili-ties. The discussion considers architectural implications, possible perfor-mance gains, and the constraints of present-day noisy intermediate-scale quantum (NISQ) devices. Rather than proposing a concrete hybrid imple-mentation, the paper focuses on mapping how QNN concepts could com-plement or extend existing AI-based frameworks for IoT security. The anal-ysis highlights both the opportunities and the open challenges of integrat-ing quantum-inspired methods into intrusion detection, contributing to the broader discourse on next-generation approaches to securing IoT systems.