A paper recently published in the Journal of Systems and Software proposed a novel hybrid quantum-classical architecture based on quantum machine learning (QML) and security information and event management (SIEM) for smart city security.
Study: Hybrid quantum architecture for smart city security. Image Credit: Studio_Nazh/Shutterstock.com
Smart City Security Challenges
In rapidly urbanizing environments, smart cities are crucial for addressing resource challenges, optimizing infrastructure, enhancing urban living, and harnessing technology for efficiency, improved quality of life, and sustainability. In recent years, technologies like mobile computing, cloud computing, the Internet of Things (IoT), and big data have become integral parts of smart cities.
Smart cities typically generate huge amounts of data due to the high usage of connected devices, sensors, and networks. The critical information exchanged or stored by technologies like IoT or cloud computing can be modified or accessed by attackers when the information is not properly secured.
Thus, IoT vulnerabilities, urban infrastructure risks, data privacy, and cyber threats are the major concerns for smart cities. These security concerns must be addressed by implementing robust solutions to protect critical services, citizens, and digital assets. Advanced and efficient techniques must be utilized to detect attacks and ensure higher availability, integrity, and confidentiality of the information exchange within smart cities.
Robust cybersecurity measures, public-private partnerships, community engagement, standardized security protocols, data encryption, and artificial intelligence (AI)-driven threat detection, can foster a secure and resilient smart city ecosystem. For instance, SIEM assists in real-time threat detection, incident response, and monitoring by analyzing and aggregating security data.
The Proposed Architecture
In this work, researchers proposed a hybrid quantum-classical architecture designed to enhance smart city security against cyberattacks by integrating Security Information and Event Management (SIEM) with Quantum Machine Learning (QML). The system leverages Quantum AI (QAI) alongside traditional rules and patterns to bolster threat detection and response. To validate this architecture, the researchers conducted experiments comparing QML algorithms with traditional AI algorithms, specifically random forest (RF).
The study aimed to improve the analysis of threats and vulnerabilities, enhancing incident response, detection, and overall privacy and security for smart city inhabitants. By incorporating quantum computing, the proposed system seeks to reduce the time required for attack identification in IoT environments.
The architecture combines IBM QRadar, a well-established SIEM system, with QBoost, a QML algorithm available through the D-Wave Leap Quantum Cloud. A proof-of-concept dashboard was developed to illustrate the integration of these components, enabling security analysts to monitor and respond to attacks in real-time without having to separately access QRadar and QBoost.
QRadar employs a rule-based approach to identify issues and intrusions, using a set of pre-configured rules to analyze data from various smart city interfaces. In this hybrid architecture, QML algorithms are crucial and are accessible as a service from external classical components.
The D-Wave quantum platform, specifically the D-Wave Leap Quantum Computing platform, is utilized for training classification models that detect intrusions and attacks within the system, leveraging data from the smart city infrastructure.
The proposed system features a three-layer architecture with six core components: QRadar (SIEM), a quantum cloud service, a database, the principal application, a dashboard, and a smart city data source. The architecture's data, application, and presentation tiers are developed as separate infrastructures to enhance modularity and security.
To evaluate the architecture, researchers used the CICIoT2023 dataset, which accurately reflects real-world smart city scenarios. This IoT dataset includes seven attack categories: Mirai, Spoofing, Brute-Force, Web-Based, Recon, DoS, and DDoS. Various attacks were performed for each category to assess the system’s effectiveness.
Importance of this Work
Researchers successfully demonstrated the validity of the hybrid quantum-classical architecture for smart city security. The comparative analysis of the QBoost and RF classifier performances showed the potential of quantum computing in improving predictive tasks and data analysis in smart cities.
Additionally, the comparison between RF and QBoost based on prediction and training times displayed the feasibility of implementing QBoost within a smart city framework. Specifically, QBoost reduced the training time and prediction time by 80 % and 71 %, respectively, when compared with RF, which showed the exceptional computing efficiency of QBoost.
This significant efficiency improvement was specifically evident in prediction times where QBoost attained a processing speed of 2.24 seconds. QBoost also showed its viability as an alternative to RF based on quality metrics. The accuracy achieved by RF was 99 %, while QBoost realized 96 % accuracy in attack classification, which was only 3 % lower than RF. The time between an attack identification and the offense alert varied from 2 to 3 minutes with IBM QRadar SIEM.
To summarize, the findings of this study effectively demonstrated the feasibility of using hybrid quantum-classical approaches for smart city security.
Journal Reference
Barletta, V. S., Caivano, D., De Vincentiis, M., Pal, A., Scalera, M. (2024). Hybrid quantum architecture for smart city security. Journal of Systems and Software, 217, 112161. DOI: 10.1016/j.jss.2024.112161, https://www.sciencedirect.com/science/article/pii/S0164121224002061
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.