Automatic Device Selection and Access Policy Generation Based on User Preference for IoT Activity Workflow

This paper presents an overview of device identification techniques and the Manufacturer Usage Description (MUD) standard used for the Internet of things to reduce the IoT attack surface. The ongoing diversity and the sheer increase in the number of connected IoT devices have crumpled security efforts. There is a need to reconsider and redesign the underlying concept of developing security systems to resolve IoT security challenges. In this backdrop, device profiling and identification have emerged as an exciting technique that helps to reduce IoT device attack surface. One of the known approaches for device identification is to fingerprint a device. There are many ways to fingerprint the device, mostly using device network flows or device local attributes. The device identification ensures the authenticity of the device attached to the network, like user authentication. Since IoT devices mostly work using machine-to-machine (M2M) communication, this requires identifying each device properly. But there is no unified approach for device identification for the ever-growing world of IoT devices and applications. One of the major steps forward in this direction is the development of the Manufacturer Usage Description (MUD) standard that defines the role of a device within the network. It limits the device to execute the primary task only, which will help to reduce the attack surface. Since the inception of MUD, many security frameworks use this standard for IoT security. However, there is a need to scrutinize the security frameworks based on the MUD, to find out the claimed effectiveness of the standard in IoT security. This paper initially identifies and classifies the potential vulnerabilities in IoT devices. Then, the study provides an overview of the research that focuses on device identification techniques and analyzes their role in IoT security. Finally, the research presents an overview of MUD technology, its implementation scenarios, the limitation of the latest MUD standard, and its applications in the industry. The prime aim of this work is to examine the MUD benefits in IoT security along with the weaknesses and challenges while implementing this standard along with future directions. INDEX TERMSManufacturer usage description (MUD), Internet of Things (IoT), device identification (DI), software defined network (SDN), machine learning (ML), deep learning (DL). NOMAN MAZHAR received the B.E. degree in software engineering and the M.S. degree in information technology from the

show abstract

“…In comparison to this, MUD is not suitable for the dynamic IoT environment. As the MUD profiles are predefined and not updated dynamically [171].…”

Section: E Dynamic Policy Updatementioning

confidence: 99%

Role of Device Identification and Manufacturer Usage Description in IoT Security: A Survey

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In view of this problem, Al-Shaboti et al [4] developed a new technique for fulfilling user activities using a user preference model to automatically generate IoT policy for these activities.…”

Section: Policy Generationmentioning

confidence: 99%

“… 4. shows network access rules have no significant effects on the switch neither in the Ethernet connection nor in the WiFi links.…”

mentioning

confidence: 99%

Fine-grained Access Control for Internet of Things Smart Spaces Driven by User Inputs

Al-Shaboti¹

2021

Preprint

Self Cite

View full text Add to dashboard Cite

The increasing use of Internet of Things (IoT) devices raises security and privacy concerns. In smart spaces, multiple IoT devices are simultaneously used to fulfil user activity functions. However, these devices exhibit several security vulnerabilities that can compromise smart space security and privacy. The ability of fine-grained control network access in IoT devices and application messages can significantly reduce the risk resulting from the exploitation of IoT vulnerabilities due to unauthorised access, thereby improving smart space security. A well-recognised approach in the literature for IoT access control is to use pre-defined access policies to allow the necessary connections for a device to function correctly. However, these policies allow access to all device functions (i.e. coarse-grained access) including those functions that are not used by any user activity. The overall goal of this thesis is to develop an access control framework and techniques to achieve fine-grained access policies by using user inputs. The user inputs will be utilised to select devices to fulfil user activities aiming to build an access policy from the minimum access required for each device function. In this thesis, the use of user inputs to meet user security and privacy requirements in single- and multi-user smart spaces is studied. The main contributions are as follows: first, an access control framework that enables users to tailor IoT device policies to meet their security and privacy requirements is proposed. Validation results of the framework show the effectiveness of integrating user access rules into the existing security countermeasures (i.e. pre-defined policies and intrusion detection systems – IDS) to enforce user security and privacy. Second, the problem of selecting preferable devices to fulfil user activity functions is formulated as an optimisation problem. The optimisation problem is then solved by local and global optimisation searching algorithms that are guided by a developed user preference quantified model. The results show that global optimisation search algorithms such as Genetic Algorithm (GA) find the solution more effectively and efficiently than local search algorithms such as simulated annealing and hill-climbing. Third, sharing access control for multi-user smart spaces is proposed. Traditional access control that considers a single user is not suitable for multi-user smart spaces, where users share their IoT devices. The sharing between multiple users poses challenges different than in single-user access control. For example, users may abuse using shared devices and use vulnerable ones. This thesis addresses these two challenges through two contributions. First, it proposes a novel sharing policy language that enables users to precisely define their sharing policy. Second, this thesis formulates the sharing policies as constraints in the context of an optimisation problem with the objective function that maximises the use of secure devices. Results show that the IoT sharing issue can naturally be translated into an integer linear programming (ILP) problem and effectively solved using off-the-shelf ILP solvers. Fourth, this thesis explores the feasibility and practicality of the fine-grained access policy enforcement through a smart home case study. A case study is built using a hub-based architecture that uses Web of Things (WoT) technology. WoT provides a device semantic description that includes device functions with the corresponding Uniform Resource Identifier (URI) which is used to build access control policies. The case study results show that policy enforcement can be effectively achieved by directing network traffic through a device proxy for each IoT device to enforce application access control without introducing statistically significant overhead on the user activity running time. In summary, this thesis studies the use of user inputs to derive fine-grained access control in smart spaces. For a single-user access control system, this thesis considers using manual rules and user preferences in small and dense smart spaces, respectively. For a multi-user access control system, this thesis proposes a secure sharing system supported by a sharing policy language to share and use IoT devices securely. For each scenario analysed, user input is utilised to derive fine-grained access policies. Enforcement of these policies has been explored by implementing a smart space case study using WoT technology. The overall results show that user preferences and sharing policies can be used to derive fine-grained access policies that are transparent to users and meet their security and privacy requirements.

show abstract

A Decision Support System for Matching Users and Fall Risk Detection Technology

Caragiuli,

Brunzini,

Massera

et al. 2024

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Automatic Device Selection and Access Policy Generation Based on User Preference for IoT Activity Workflow

Cited by 7 publications

References 31 publications

Role of Device Identification and Manufacturer Usage Description in IoT Security: A Survey

Role of Device Identification and Manufacturer Usage Description in IoT Security: A Survey

Fine-grained Access Control for Internet of Things Smart Spaces Driven by User Inputs

A Decision Support System for Matching Users and Fall Risk Detection Technology

Contact Info

Product

Resources

About