Source-linked AI summary
Security Aspects of Internet of Things aided Smart Grids: a Bibliometric Survey
Jacob Sakhnini, Hadis Karimipour, Ali Dehghantanha, Reza M. Parizi, Gautam Srivastava
TL;DR
Smart-grid communication and sensing create substantial economic and social benefits but also expose increasingly complex systems to diverse cyber threats. This paper identifies, classifies, and reviews smart-grid cybersecurity publications through bibliometric analysis, finding exponential field growth and emphasizing unresolved mitigation and deception-strategy gaps.
Problem
The variety and complexity of cyber threats in smart grids create a need to understand the cybersecurity research landscape.
Method
The paper identifies, classifies, and reviews publications on smart-grid cybersecurity using bibliometric analysis organized by authors, dates, and publication trends.
Results
The findings demonstrate exponential growth in smart-grid security research over the last decade and reflect diverse threats and complex smart-grid systems.
Takeaways & Limitations
The survey consolidates existing cybersecurity knowledge while highlighting the need to address the variety and complexity of threats and smart-grid security research gaps.
Takeaways & Limitations
Research gaps remain in efficient deception strategies for smart grids and in mitigating cyber threats that have already occurred.
Abstract
from arXiv · showhide
The integration of sensors and communication technology in power systems, known as the smart grid, is an emerging topic in science and technology. One of the critical issues in the smart grid is its increased vulnerability to cyber threats. As such, various types of threats and defense mechanisms are proposed in literature. This paper offers a bibliometric survey of research papers focused on the security aspects of Internet of Things (IoT) aided smart grids. To the best of the authors' knowledge, this is the very first bibliometric survey paper in this specific field. A bibliometric analysis of all journal articles is performed and the findings are sorted by dates, authorship, and key concepts. Furthermore, this paper also summarizes the types of cyber threats facing the smart grid, the various security mechanisms proposed in literature, as well as the research gaps in the field of smart grid security.
1. Introduction
Smart grids improve power-system coordination and efficiency through extensive digital, communication, sensor, and IoT integration, but this interconnected complexity increases exposure to cyber threats. The paper surveys this expanding security literature and its research directions.
- Smart-grid integration: Smart grids integrate smart meters, SCADA, controllers, and energy sources to support more efficient distribution and broader power-system coordination.IoT devices additionally support network functions in generation and storage and connectivity between suppliers and consumers.
- Security challenges: Digital integration and system complexity increase the possibility of cyber attacks and failures propagating between interconnected systems.The paper identifies modeling nonlinearities, stochasticity, and diverse cyber attacks as security challenges.
- IoT security: IoT devices enhance smart-grid capabilities but introduce additional vulnerabilities to cyber attacks.The paper describes IoT as networks of internet-connected physical devices used across power generation, storage, and supplier–consumer connectivity.
- Prior defenses: Existing defenses include model-based estimation, Kalman filters, supervised and semi-supervised learning, reinforcement learning, and deep-learning approaches.These methods target threats including false-data injection and use measurements, nonlinear features, and temporal patterns for detection or mitigation.
- Research gap: The variety and complexity of smart-grid cyber threats have generated many proposed solutions, while existing reviews lack an up-to-date bibliometric analysis and attack-detection-method inquiry.Earlier reviews were published before 2016, and more recent threat surveys did not provide bibliometric analysis or sufficient coverage of detection methods.
- Paper purpose: This paper identifies, classifies, and reviews smart-grid cybersecurity publications to characterize research patterns, threats, defenses, and future directions.Its questions address publication trends, future cybersecurity direction, studied threat types, and defense mechanisms in IoT-integrated smart grids.
2. Methodology
The study applies a bibliometric workflow to journal literature on smart-grid cybersecurity, combining database searches, relevance screening, categorization, and visualization.
- Bibliometric analysis: The bibliometric analysis measures research output mainly through the frequency of keywords and phrases.The method follows a prior bibliometric process and organizes publication analysis alongside search and source-error reduction procedures.
- Screening and inclusion: Results are cross-referenced, screened for smart-grid cybersecurity relevance through abstracts, and filtered to exclude non-English or irrelevant papers.The included literature is then categorized by timeline, journals, authors, and research output.
- Data handling: Zotero and VOSviewer are used to sort and visualize the bibliometric data.The tools support organization and visualization after the literature has been retrieved, screened, and categorized.
3. Findings
The bibliometric analysis identifies a rapidly expanding, widely distributed research field, with 1,722 articles contributed by many authors and journals. Publications cluster in recent years, while keyword analysis maps the field’s principal concepts.
- 1,722 journal articles remained after duplicate filtering from 2,314 database search results.
- 61.2% of findings came from the Web of Science database, compared with 30.1% from ScienceDirect and 8.67% from IEEE Xplore.
- Most articles were published during the last five years, and publication growth indicates that smart-grid security is a recently developing field.
- Articles were distributed across many journals, although some journals published more than others, indicating broad interest across scientific and societal domains.
- The 1,722 articles included contributions from 4,952 authors, while the leading author accounted for only 1.16% of publications.
- The keyword heat-map identifies prominent terms, colored clusters, and connections representing the main concepts associated with smart-grid security.
4. Reported Attacks on the Smart Grid
Reported incidents show that smart-grid security failures can affect power generation, monitoring, metering, and broader critical-infrastructure operations. The section links cyber incidents and situational-awareness failures to operational disruption and cascading damage.
- Malware disabled processing and safety-monitoring systems at an Ohio nuclear plant for several hours.
- Excessive traffic believed to result from a DoS attack caused circulation pumps at Alabama’s Brown Ferry nuclear plant to fail.
- Hackers changed smart-meter consumption readings using software readily available on the internet, enabling power theft.
- The authors argue that smart-grid security should be examined at every level because security systems can also minimize damage from faults or incidents.
- A lack of situational awareness contributed to a cascading failure affecting 508 generators and 265 power plants across eight states and southern Ontario.
5. Security Systems in the Smart Grid
This section examines smart-grid security threats and the defense mechanisms proposed in the literature. It organizes the discussion around attack types and corresponding countermeasures.
- The section reviews security threats facing the smart grid and the state of current countermeasures.
- Figure 4 presents the number of journal articles studying each attack type.
- Subsection 5.1 discusses specific cyber threats in power systems, while subsection 5.2 discusses defense mechanisms proposed in the literature.
5.1. Types of Cyber Threats
The literature covers diverse passive and active cyber threats, including spoofing, replay, malware, false-data injection, and channel jamming. These attacks can mislead operators, corrupt measurements, disrupt communications, or contribute to power-system collapse.
- Cyber attacks are widely studied because identified vulnerabilities can potentially drive power systems into total collapse.
- Passive attacks include eavesdropping, spying, and traffic analysis, whereas active attacks include DoS and malware attacks.
- GPS spoofing can deceive phasor measurement units, mislead network operators, and affect subsequent corrective-control actions.
- Replay attacks reuse intercepted smart-meter usage patterns, and IoT integration increases exposure to these attacks.
- Malware threatens smart-grid communications, while similar hardware and firmware across smart meters can increase susceptibility to propagation.
- False Data Injection attacks manipulate meter measurements, potentially producing incorrect energy prices or inaccurate predictions.
- Jamming specific signal channels disrupts data transmission, producing unreliable communications and reduced power-system performance.
5.2. Defence Mechanisms
The literature organizes smart-grid defense mechanisms through a 7D cybersecurity model, covering vulnerability discovery, detection, prevention, disruption, deception, and degradation or destruction. Studies identify extensive vulnerabilities and attack strategies, while detection and mitigation of some threats remain open gaps.
- Framework: The 7D model structures proposed smart-grid security measures into seven cybersecurity phases.The surveyed literature discusses each phase and its proposed methods.
- Discovery and vulnerability analysis: Automated vulnerability-discovery methods extract security-related features and analyze smart-grid vulnerabilities across diverse environments and threats.Examples include binary-based discovery, automatic static analysis, and vulnerability modeling with incomplete topology information.
- Discovery and vulnerability analysis: Limited system knowledge can still expose exploitable smart-grid vulnerabilities, while risk-graph analysis identifies important nodes and enables new node-attack strategies.The risk graph represents node importance and relationships within the system.
- Discovery and vulnerability analysis: Vulnerability studies cover component-specific, coordinated, structural, functional, and emergency scenarios, including attacks on substations, transmission lines, SCADA networks, and GPS measurements.Joint substation-transmission-line assessment accounts for attacks occurring in either component or both.
- Research gaps: The surveyed literature provides substantial vulnerability and attack-strategy analysis, but detection and mitigation of some cyber threats remain research gaps.The survey also identifies a need for more versatile honeynet systems in deception strategies.
- Detection: Feature selection and dimensionality reduction are proposed to improve computational efficiency while maintaining detection accuracy, but few studies examine deep learning with automated or unsupervised feature selection.Purely model-based detection is described as insufficient to guarantee security at larger scale, motivating intelligent and machine-learning approaches.
- Detection: Supervised learning generally produces more accurate attack classification, and Gaussian-based SVM is reported as more robust and accurate on larger test systems.Other evaluated techniques include margin-setting algorithms, neural networks, AdaBoost, random forests, and common-path mining.
- Prevention: Encryption-based prevention commonly uses symmetric or asymmetric keys, although asymmetric encryption requires greater computational capacity and is unsuitable for time-sensitive applications.The prevention mechanisms target attacks including brute-force, replay, man-in-the-middle, and denial-of-service attacks.
6. Conclusion
The bibliometric analysis finds exponential growth in IoT-aided smart-grid security research and broad coverage of threats, security measures, and evaluation methods. It identifies priorities for intelligent security methods, faster algorithms with accurate detection and fewer false alarms, and greater attention to threat mitigation.
- The paper analyzes publications on the security of IoT-aided smart grids, reporting bibliometric findings on the field’s significance and growth.
- Exponential growth in smart-grid security research occurred over the last decade, while analyzed journal papers addressed diverse issues and solutions.
- The findings summarize cyber threats, security measures, and evaluation methods used for smart-grid security systems.
- The variety and complexity of smart-grid threats and systems call for comprehensive and intelligent security methods.
- Future work is expected to increase security-algorithm speed while maintaining high detection accuracy and low false-alarm rates.
- A research gap remains in mitigating threats that have already infected smart grids, because most papers focus on detection and prevention.
- Projected future trends include cyber-threat mitigation and robust deep-learning algorithms for efficient threat detection.