PoisonIvy: (In)secure Practices of Enterprise IoT Systems in Smart Buildings

Luis Puche Rondon, Leonardo Babun, Ahmet Aris, Kemal Akkaya, A. Selcuk Uluagac

TL;DR

This paper introduces "PoisonIvy," a novel attack framework targeting Enterprise IoT (EIoT) systems, commonly found in smart buildings. By exploiting vulnerabilities in EIoT drivers, attackers can gain control, disrupt operations, and even leverage system resources for malicious purposes like cryptocurrency mining. This research highlights a critical, under-explored security gap in professional EIoT deployments.

What Problem Does This Paper Address?

The proliferation of smart devices has led to the widespread adoption of Enterprise IoT (EIoT) systems in environments like smart buildings, luxury homes, and government offices. Unlike consumer-grade IoT, EIoT systems are complex, proprietary, and professionally managed, often requiring specialized training and software. This complexity and the closed-source nature of many EIoT components have resulted in a significant lack of security research in this domain. Consequently, EIoT systems are often overlooked as potential attack vectors, despite their critical role in managing building functions and sensitive data. This paper addresses this gap by investigating the security vulnerabilities inherent in EIoT system development practices, specifically focusing on the role of drivers.

Methods

The core of this research is the development and demonstration of "PoisonIvy," a framework designed to exploit EIoT drivers. The methodology involves:

  • Driver-Based Attacks: PoisonIvy leverages "drivers," which are software modules responsible for integrating third-party devices and services into EIoT systems. The paper argues that these drivers, often sourced from untrusted third parties and lacking robust verification, are a prime attack vector.
  • Command and Control (C2) Architecture: A C2 server is established to act as an intermediary between a remote attacker and compromised EIoT controllers. This server facilitates attack initiation, command delivery, and status reporting.
  • Malicious Driver Implementation: A malicious driver is developed using Lua, the programming language commonly used for Control4 EIoT drivers. This driver is designed to mimic legitimate driver behavior while embedding malicious payloads.
  • Testbed Deployment: A realistic EIoT testbed is set up using Control4 hardware and software, simulating a smart building environment. This allows for the practical evaluation of PoisonIvy attacks.
  • Attack Scenarios: Three specific attack types are implemented and evaluated:
  • Denial-of-Service (DoS): Causing memory exhaustion in the EIoT controller to render the system inoperable.
  • Remote Control: Forcing infected devices to perform continuous malicious requests to a target website, simulating a botnet.
  • Malicious Resource Farming: Utilizing the controller's processing power for resource-intensive tasks like cryptocurrency mining.

Datasets

Not discussed in the paper. The research focuses on demonstrating attacks on a controlled EIoT testbed rather than training or evaluating on large datasets.

Key Insights and Ideas

  • Drivers are a Critical Vulnerability: The paper convincingly demonstrates that EIoT drivers, often overlooked due to their integration role, are a highly effective attack vector. Their proprietary nature and the reliance on third-party sources create a significant security blind spot.
  • Lack of Verification is Pervasive: The research highlights that unlike traditional operating systems, EIoT platforms often lack rigorous driver verification mechanisms. This forces integrators to trust unverified software, opening the door for malicious actors.
  • EIoT Systems are Prime Targets: The sophisticated nature and central control of EIoT systems make them attractive targets. Compromising a single EIoT controller can grant an attacker access to a wide range of integrated devices and functionalities within a building.
  • Attackers Can Mimic Benign Behavior: PoisonIvy's success relies on the malicious driver appearing legitimate. This means attackers can embed harmful code within seemingly functional drivers, making detection difficult.
  • Resource Farming is a Real Threat: The paper shows that EIoT controllers can be repurposed for illicit activities like cryptocurrency mining, impacting system performance and potentially leading to unexpected costs or legal issues for the building owner.

Key Results

The PoisonIvy attacks were successfully implemented and demonstrated on a real EIoT testbed:

  • DoS Attack: The memory exhaustion attack rendered the Control4 controller inoperable within five seconds, disrupting all communication and user interfaces. The system required a power cycle to recover.
  • Remote Control Attack: Infected devices were able to perform repeated, continuous requests to a target website, demonstrating the potential for botnet-like behavior and distributed attacks.
  • Resource Farming Attack: The system successfully executed multiple SHA-256 hashing operations, simulating cryptocurrency mining, showcasing the potential for attackers to leverage EIoT system resources for their benefit.

Limitations and Future Work

  • Limitations:
    • The research focuses on a specific EIoT platform (Control4) and may not be directly transferable to all EIoT systems without adaptation.
  • The study does not consider attacks on traditional operating systems or protocol-specific vulnerabilities (e.g., Zigbee).
  • The paper acknowledges that while the implemented attacks were successful, the complexity and impact could vary based on the specific EIoT system and its configuration.
  • Future Work:
    • The authors suggest the need for robust security verification mechanisms for EIoT drivers, similar to those found in traditional operating systems.
  • They advocate for inherent security features within EIoT controllers to limit external communications, resource consumption, and unauthorized access.
  • The paper calls for greater awareness among EIoT integrators regarding the risks of using unverified drivers and the importance of ongoing driver validation, even for updated versions.
  • Further research into standardization efforts for EIoT driver implementation is recommended.

Summary updated

Abstract

The rise of IoT devices has led to the proliferation of smart buildings, offices, and homes worldwide. Although commodity IoT devices are employed by ordinary end-users, complex environments such as smart buildings, smart offices, conference rooms, or hospitality require customized and highly reliable solutions. Those systems called Enterprise Internet of Things (EIoT) connect such environments to the Internet and are professionally managed solutions usually offered by dedicated vendors. As EIoT systems require specialized training, software, and equipment to deploy, this has led to very little research investigating the security of EIoT systems and their components. In effect, EIoT systems in smart settings such as smart buildings present an unprecedented and unexplored threat vector for an attacker. In this work, we explore EIoT system vulnerabilities and insecure development practices. Specifically, focus on the usage of drivers as an attack mechanism, and introduce PoisonIvy, a number of novel attacks that demonstrate an attacker can easily compromise EIoT system controllers using malicious drivers. Specifically, we show how drivers used to integrate third-party devices to EIoT systems can be misused in a systematic fashion. To demonstrate the capabilities of attackers, we implement and evaluate PoisonIvy using a testbed of real EIoT devices. We show that an attacker can perform DoS attacks, gain remote control, and maliciously abuse system resources of EIoT systems. To the best of our knowledge, this is the first work to analyze the (in)securities of EIoT deployment practices and demonstrate the associated vulnerabilities in this ecosystem. With this work, we raise awareness on the (in)secure development practices used for EIoT systems, the consequences of which can largely impact the security, privacy, reliability, and performance of millions of EIoT systems worldwide.
BibTeX
@inproceedings{rondon2020poisonivy,
  author = {Luis Puche Rondon and Leonardo Babun and Ahmet Aris and Kemal Akkaya and A. Selcuk Uluagac},
  title = {{PoisonIvy: (In)secure Practices of Enterprise IoT Systems in Smart Buildings}},
  booktitle = {BuildSys},
  year = {2020},
  doi = {10.1145/3408308.3427606},
}
Figure