-
Understanding IoT Domain Names: Analysis and Classification Using Machine Learning
Authors:
Ibrahim Ayoub,
Martine S. Lenders,
Benoît Ampeau,
Sandoche Balakrichenan,
Kinda Khawam,
Thomas C. Schmidt,
Matthias Wählisch
Abstract:
In this paper, we investigate the domain names of servers on the Internet that are accessed by IoT devices performing machine-to-machine communications. Using machine learning, we classify between them and domain names of servers contacted by other types of devices. By surveying past studies that used testbeds with real-world devices and using lists of top visited websites, we construct lists of d…
▽ More
In this paper, we investigate the domain names of servers on the Internet that are accessed by IoT devices performing machine-to-machine communications. Using machine learning, we classify between them and domain names of servers contacted by other types of devices. By surveying past studies that used testbeds with real-world devices and using lists of top visited websites, we construct lists of domain names of both types of servers. We study the statistical properties of the domain name lists and train six machine learning models to perform the classification. The word embedding technique we use to get the real-value representation of the domain names is Word2vec. Among the models we train, Random Forest achieves the highest performance in classifying the domain names, yielding the highest accuracy, precision, recall, and F1 score. Our work offers novel insights to IoT, potentially informing protocol design and aiding in network security and performance monitoring.
△ Less
Submitted 23 April, 2024;
originally announced April 2024.
-
Securing name resolution in the IoT: DNS over CoAP
Authors:
Martine S. Lenders,
Christian Amsüss,
Cenk Gündogan,
Marcin Nawrocki,
Thomas C. Schmidt,
Matthias Wählisch
Abstract:
In this paper, we present the design, implementation, and analysis of DNS over CoAP (DoC), a new proposal for secure and privacy-friendly name resolution of constrained IoT devices. We implement different design choices of DoC in RIOT, an open-source operating system for the IoT, evaluate performance measures in a testbed, compare with DNS over UDP and DNS over DTLS, and validate our protocol desi…
▽ More
In this paper, we present the design, implementation, and analysis of DNS over CoAP (DoC), a new proposal for secure and privacy-friendly name resolution of constrained IoT devices. We implement different design choices of DoC in RIOT, an open-source operating system for the IoT, evaluate performance measures in a testbed, compare with DNS over UDP and DNS over DTLS, and validate our protocol design based on empirical DNS IoT data. Our findings indicate that plain DoC is on par with common DNS solutions for the constrained IoT but significantly outperforms when additional standard features of CoAP are used such as caching. With OSCORE, we can save more than 10 kBytes of code memory compared to DTLS, when a CoAP application is already present, and retain the end-to-end trust chain with intermediate proxies, while leveraging features such as group communication or encrypted en-route caching. We also discuss a compression scheme for very restricted links that reduces data by up to 70%.
△ Less
Submitted 27 July, 2023; v1 submitted 15 July, 2022;
originally announced July 2022.
-
A Lesson in Scaling 6LoWPAN -- Minimal Fragment Forwarding in Lossy Networks
Authors:
Martine S. Lenders,
Thomas C. Schmidt,
Matthias Wählisch
Abstract:
This paper evaluates two forwarding strategies for fragmented datagrams in the IoT: hop-wise reassembly and a minimal approach to directly forward fragments. Minimal fragment forwarding is challenged by the lack of forwarding information at subsequent fragments in 6LoWPAN and thus requires additional data at nodes. We compared the two approaches in extensive experiments evaluating reliability, end…
▽ More
This paper evaluates two forwarding strategies for fragmented datagrams in the IoT: hop-wise reassembly and a minimal approach to directly forward fragments. Minimal fragment forwarding is challenged by the lack of forwarding information at subsequent fragments in 6LoWPAN and thus requires additional data at nodes. We compared the two approaches in extensive experiments evaluating reliability, end-to-end latency, and memory consumption. In contrast to previous work and due to our alternate setup, we obtained different results and conclusions. Our findings indicate that direct fragment forwarding should be deployed only with care, since higher packet transmission rates on the link-layer can significantly reduce its reliability, which in turn can even further reduce end-to-end latency because of highly increased link-layer retransmissions.
△ Less
Submitted 28 August, 2019; v1 submitted 20 May, 2019;
originally announced May 2019.
