The Security of Smart Buildings: a Systematic Literature Review

Pierre Ciholas, Aidan Lennie, Parvin Sadigova, Jose M. SucharXiv 2019 · arXiv · OpenReview

TL;DR

This paper presents a systematic literature review of smart building (SB) security, revealing a rapidly growing and increasingly complex research field. While significant effort is being put into defenses, there's a critical lack of empirical evaluations and a concerning neglect of non-technical factors like human and organizational aspects, which are crucial for real-world security.

What Problem Does This Paper Address?

Smart Buildings (SBs) are becoming ubiquitous, integrating numerous interconnected devices and software systems to manage building functions from HVAC to security. This increased connectivity, however, opens them up to a wide range of new cyber threats. While research exists on SB security, there's a lack of a consolidated overview that synthesizes the current state of the art, identifies key challenges, and points towards future research directions. This paper aims to fill that gap by providing a comprehensive systematic literature review of SB security.

Methods

The authors conducted a systematic literature review (SLR) to provide a thorough understanding of the state of the art in SB security research. The process involved:

  1. Search Procedure: They searched major computer science and security databases (Web of Science, IEEE Xplore, ACM Digital Library, Scopus, Usenix) using two sets of keywords: one for SB-related terms (e.g., "Smart Building," "Building Automation System") and another for security aspects (e.g., "Security," "Vulnerability," "Attack"). This yielded an initial pool of 1697 papers.
  2. Vetting Procedure: Papers were initially screened based on title and abstract to determine if SBs were the primary focus and if they addressed vulnerabilities or security aspects. A rigorous vetting process involving multiple researchers and inter-rater reliability measures (kappa statistic) was used to select relevant papers. This process, including a third reviewer for disagreements, resulted in a final set of 90 papers for in-depth review.
  3. Reviewing Papers: The 90 selected papers were then fully retrieved and analyzed using a review template to extract key information, including paper type, proposed solutions, evaluation methods, and protocols discussed.

Datasets

Not discussed in the paper. The paper focuses on a systematic review of existing literature rather than introducing new datasets or benchmarks.

Key Insights and Ideas

  • Exploding Complexity: The field of SB security is growing rapidly in complexity, driven by the increasing number of interconnected devices and the proliferation of diverse communication protocols. This expanding attack surface is a major concern.
  • Empirical Evaluation Gap: Despite the growing research, there's a significant lack of empirical evaluations. While testbed and real-system evaluations are increasing, a substantial portion of research still lacks rigorous validation, making it difficult to assess the true effectiveness of proposed solutions.
  • The Human Factor is Missing: A striking observation is the almost complete lack of consideration for non-technical aspects like social, organizational, and human factors. These are critical in SB environments where ownership and liability are often unclear, and human error or social engineering can be major vulnerabilities.
  • SB-Specific vs. General IT Issues: The research highlights that many SB security issues are specific to the domain, rather than just being extensions of classic IT security problems. This underscores the need for domain-specific research and solutions.
  • Protocol Vulnerabilities: Many SB protocols were not designed with security in mind, leading to inherent vulnerabilities that are difficult to patch due to the long lifecycles of building systems.

Key Results

  • Publication Trend: Research on SB security has seen a steep increase since 2007, with a peak in publications in 2014-2015.
  • Main Topics: The majority of papers focus on "Defences" (54%), followed by "Attacks" (19%), "Discussions" (12%), "Vulnerabilities" (7%), and "New Systems/Features" (8%).
  • Popular Protocols: BACnet, KNX, and Lon are the most frequently discussed protocols, though newer wireless protocols are gaining traction.
  • Methodologies: The most common methodologies involve describing the design/implementation of systems/software (30%) and security protocols (18%), followed by experimental studies (17%) and technology/literature overviews (18%).
  • Evaluation Landscape: A significant portion of papers (37%) have no evaluation. While evaluations on testbeds and real systems are increasing, they still represent a minority (15% on real systems).
  • Attack/Vulnerability Focus: Attacks and vulnerabilities are categorized by SB layer. Field-level attacks are often wireless, while automation/management layer attacks include denial-of-service, protocol-specific, and configuration/legacy issues.

Limitations and Future Work

  • Lack of Empirical Evaluation: The authors explicitly highlight the lack of empirical evaluations as a major limitation and a key area for future research. They suggest that the cost and complexity of setting up SB testbeds and gaining access to real systems are significant hurdles.
  • Neglect of Non-Technical Aspects: The paper strongly emphasizes the need for future research to address social, organizational, and human factors, including social engineering, phishing, and user awareness.
  • Composite Vulnerabilities: The authors point out the limited consideration of composite vulnerabilities – how multiple single vulnerabilities can chain together to create more severe threats. Understanding these interdependencies is crucial.
  • AI for Security: The paper suggests a significant opportunity for using AI techniques, such as machine learning and knowledge-based systems, to develop more intelligent and adaptive security solutions for SBs.
  • Internet-Facing SB Security: Securing SB systems that are directly exposed to the internet remains a significant challenge, requiring further research to assess their security posture and develop robust defenses against remote attacks.
  • Responsibility and Liability: The lack of clear responsibility and liability frameworks for SB security is identified as a critical issue that needs further investigation.

Summary updated

Abstract

Smart Buildings are networks of connected devices and software in charge of automatically managing and controlling several building functions such as HVAC, fire alarms, lighting, shading and more. These systems evolved from mostly electronic and mechanical elements to complex systems relying on IT and wireless technologies and networks. This exposes smart buildings to new risks and threats that need to be enumerated and addressed. Research efforts have been done in several areas related to security in smart buildings but a clear overview of the research field is missing. In this paper, we present the results of a systematic literature review that provides a thorough understanding of the state of the art in research on the security of smart buildings. We found that the field of smart buildings security is growing significantly in complexity due to the many protocols introduced recently and that the research community is already studying. We also found an important lack of empirical evaluations, though evaluations on testbeds and real systems seems to be growing. Finally, we found an almost complete lack of consideration of non-technical aspects, such as social, organisational, and human factors, which are crucial in this type of systems, where ownership and liability is not always clear.
BibTeX
@misc{ciholas2019security,
  author = {Pierre Ciholas and Aidan Lennie and Parvin Sadigova and Jose M. Such},
  title = {{The Security of Smart Buildings: a Systematic Literature Review}},
  year = {2019},
  eprint = {1901.05837},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1901.05837},
}
Figure