Proactively Extracting IoT Device Capabilities: An Application to Smart Homes

Dolan, Andy; Ray, Indrakshi; Majumdar, Suryadipta

doi:10.1007/978-3-030-49669-2_3

Cited by 5 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To build knowledge of security standards, we develop a Python script that extracts the security expectations from NIST IR 8228 and the referenced NIST SP 800-53r5 security controls from control catalog (provided by NIST), and store them in a CSV file with attributes such as variables, expectations, controls, sub-controls and their discussions. To build device-specific knowledge, network data such as port numbers and protocols are extracted by leveraging MUDGEE [49], which creates MUD [48] profiles of IoT devices by monitoring network traces and technical specifications of different IoT devices are extracted by leveraging the approach from [8]. For the summarization of security controls, we use BERT-Extractive-Summarizer [51].…”

Section: Methodsmentioning

confidence: 99%

“…The structural knowledge of an IoT device includes different capabilities of its sensors and actuators, which are derived from the manufacture design specifications of different IoT devices. For this purpose, we leverage the approach proposed by Dolan et al [8] as follows. First, we gather all the technical information that provided on a device's website.…”

Section: Building Device Specific Knowledgementioning

confidence: 99%

“…Second, we extract those information from the device's API documentation that describes API calls to change system states. Third, we gather essential information from IoT device configuration files, which are publicly accessible and include all essential device characteristics [8]. The functional knowledge of an IoT device includes its network behaviors that can be captured through manufacturer usage descriptor (MUD) (i.e., a framework by IETF for formally describing the network behaviour [48])…”

Section: Building Device Specific Knowledgementioning

confidence: 99%

“…As there are no trained NER models available to extract security values, we had to annotate the security values manually to fine-tune the NER model. We leverage the approach of Dolan et al in [8] to automatically extract device specifications, configs, and API documentation, while all additional device-specific data is manually verified to assure its completeness.…”

Section: Manualmentioning

confidence: 99%

See 3 more Smart Citations

On Reducing Underutilization of Security Standards by Deriving Actionable Rules: An Application to IoT

Shuvo¹,

Hoq²,

Majumdar³

et al. 2023

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Even though there exist a number of security guidelines and recommendations from various worldwide standardization authorities (e.g., NIST, ISO, ENISA), it is evident from many of the recent attacks that these standards are not strictly followed in the implementation of real-world products. Furthermore, most security applications (e.g., monitoring and auditing) do not consider those standards as the basis of their security check. Therefore, regardless of continuous efforts in publishing security standards, they are still under-utilized in practice. Such under-utilization might be caused by the fact that existing security standards are intended more for high-level recommendations than for being readily adopted to automated security applications on the system-level data.Bridging this gap between high-level recommendations and low-level system implementations becomes extremely difficult, as a fully automated solution might suffer from high inaccuracy, whereas a fully manual approach might require tedious efforts. Therefore, in this thesis, we aim for a more practical solution by proposing a partially automated approach, where it automates the tedious tasks (e.g., summarizing long standard documents, and extracting device specifications) and relies on manual efforts from security experts to avoid mistakes in finalizing security rules. We apply our solution to IoT by implementing it with IoT-specific standards (NISTIR 8228) and smart home networks. We further demonstrate the actionability of our derived rules in three major applications: security auditing, Intrusion Detection systems (IDS), and secure application development.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Building Device Specific Knowledgementioning

confidence: 99%

Section: Building Device Specific Knowledgementioning

confidence: 99%

Section: Manualmentioning

confidence: 99%

See 2 more Smart Citations

On Reducing Underutilization of Security Standards by Deriving Actionable Rules: An Application to IoT

Shuvo¹,

Hoq²,

Majumdar³

et al. 2023

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…For example, smart light may come with a microphone that may leave the IoT system vulnerable to security and privacy intrusions. A common household resident may not have knowledge and skills to interpret and safeguard the IoT system [4]. Hence, there is a huge need for an automated mechanism to protect smart homes from security and privacy intrusions.…”

mentioning

confidence: 99%

Explainable and Optimally Configured Artificial Neural Networks for Attack Detection in Smart Homes

Sohail¹,

Fan²,

Gu³

et al. 2022

Preprint

View full text Add to dashboard Cite

In recent years cybersecurity has become a major concern in adaptation of smart applications. Specially, in smart homes where a large number of IoT devices are used having a secure and trusted mechanisms can provide peace of mind for users. Accurate detection of cyber attacks is crucial, however precise identification of the type of attacks plays a huge role if devising the countermeasure for protecting the system. Artificial Neural Networks (ANN) have provided promising results for detecting any security attacks for smart applications. However, due to complex nature of the model used for this technique it is not easy for normal users to trust ANN based security solutions. Also, selection of right hyperparameters for ANN architecture plays a crucial role in the accurate detection of security attacks, especially when it come to identifying the subcategories of attacks. In this paper, we propose a model that considers both the issues of explainability of ANN model and the hyperparameter selection for this approach to be easily trusted and adapted by users of smart home applications. Also, our approach considers a subset of the dataset for optimal selection of hyperparamters to reduce the overhead of the process of ANN architecture design. Distinctively this paper focuses on configuration, performance and evaluation of ANN architecture for identification of five categorical attacks and nine subcategorical attacks. Using a very recent IoT dataset our approach showed high performance for intrusion detection with 99.9%, 99.7%, and 97.7% accuracy for Binary, Category, and Subcategory level classification of attacks.

show abstract