Empirical Evaluation of Attacks Against IEEE 802.11 Enterprise Networks: The AWID3 Dataset

Chatzoglou, Efstratios; Kambourakis, Georgios; Kolias, Constantinos

doi:10.1109/access.2021.3061609

Cited by 57 publications

(42 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Tofino switch uses a snake configuration for throughput tests. More details in Appendix D. Workloads: Our evaluation explores four use cases: attack detection (using AWID3 [8], CICIDS 2017 [52], KDD99 [58], and UNSW-NB15 [42]), finance (NASDAQ TotalView-ITCH [47], Jane Street Market Prediction [21]), QoE (Requet [24]) and flowers classification (Iris [17]). The results for attack detection (using CICIDS and UNSW) and finance are presented below, and the rest are described in appendix E.3.…”

Section: Methodology and Testbed Setupmentioning

confidence: 99%

Automating In-Network Machine Learning

Zheng¹,

Zang²,

Hong³

et al. 2022

Preprint

View full text Add to dashboard Cite

Using programmable network devices to aid in-network machine learning has been the focus of significant research. However, most of the research was of a limited scope, providing a proof of concept or describing a closed-source algorithm. To date, no general solution has been provided for mapping machine learning algorithms to programmable network devices. In this paper, we present Planter, an opensource, modular framework for mapping trained machine learning models to programmable devices. Planter supports a wide range of machine learning models, multiple targets and can be easily extended. The evaluation of Planter compares different mapping approaches, and demonstrates the feasibility, performance, and resource efficiency for applications such as anomaly detection, financial transactions, and quality of experience. The results show that Planterbased in-network machine learning algorithms can run at line rate, have a negligible effect on latency, coexist with standard switching functionality, and have no or minor accuracy trade-offs.

show abstract

Section: Methodology and Testbed Setupmentioning

confidence: 99%

Automating In-Network Machine Learning

Zheng¹,

Zang²,

Hong³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Multiple Flow Derived Features (MFD) correspond to the aggregation of information belonging to multiple flow records, containing higher level statistics (e.g., time window delimited flows, last n flows and so on). Finally, dataset labeling could be done manually (M) by a skillful professional; automatically (A) using a rule repository and a script; or on a scheduled (S) way, [20] wireless IoT 2020 IOT-23 [21] A IoT 2020 TON IOT [22] A IoT/IIoT 2019 CICIDS-2017 [23] S application traffic 2017 ISCXTor2016 [24] S application traffic (raw/Tor) 2016 ISCXVPN2016 [25] S application traffic (raw/VPN) 2016 WSN-DS [26] S wireless IoT 2016 AWID2 [27] M wireless IoT 2015 SEA [28] A UNIX commands 2001 NSL-KDD [29] MS networking 1998 1 M: manually, A: automatically, S: scheduled. launching specific attacks in pre-established time windows.…”

Section: B Datasets For Evaluating Idsmentioning

confidence: 99%

“…Regarding the monitored devices, datasets [21], [22] collect data from IoT devices using the Zeek "Bro" network security monitoring tool. Similarly, datasets [20], [26], [27] capture specific data coming from wireless IoT parties. The generalized use of DARPA type datasets [30] (i.e., storing and inferring new instances from the information exchange of multiple agents) and its derivatives (e.g., [29]) are also present in Table I.…”

Section: B Datasets For Evaluating Idsmentioning

confidence: 99%

A review of Federated Learning in Intrusion Detection Systems for IoT

Belenguer¹,

Navaridas²,

Pascual³

2022

Preprint

View full text Add to dashboard Cite

“…The respective contemporary AP models were ASUS RT-AX88U, DIR-X1860, MR7350, AX1800, RAX40, and AX10v1, respectively. Excluding legacy or common Wi-Fi vulnerabilities [36][37][38], for such devices, an essential concern is the protection of the Wi-Fi passphrase, i.e., the key(s) to connect to the Wi-Fi network. Another major concern is the safeguarding of the user credentials used to connect to the AP's web-based management interface.…”

Section: Access Pointsmentioning

confidence: 99%

Let the Cat out of the Bag: Popular Android IoT Apps under Security Scrutiny

Chatzoglou

Kambourakis

Smiliotopoulos

2022

Sensors

Self Cite

View full text Add to dashboard Cite

The impact that IoT technologies have on our everyday life is indisputable. Wearables, smart appliances, lighting, security controls, and others make our life simpler and more comfortable. For the sake of easy monitoring and administration, such devices are typically accompanied by smartphone apps, which are becoming increasingly popular, and sometimes are even required to operate the device. Nevertheless, the use of such apps may indirectly magnify the attack surface of the IoT device itself and expose the end-user to security and privacy breaches. Therefore, a key question arises: do these apps curtail their functionality to the minimum needed, and additionally, are they secure against known vulnerabilities and flaws? In seek of concrete answers to the aforesaid question, this work scrutinizes more than forty chart-topping Android official apps belonging to six diverse mainstream categories of IoT devices. We attentively analyse each app statically, and almost half of them dynamically, after pairing them with real-life IoT devices. The results collected span several axes, namely sensitive permissions, misconfigurations, weaknesses, vulnerabilities, and other issues, including trackers, manifest data, shared software, and more. The short answer to the posed question is that the majority of such apps still remain susceptible to a range of security and privacy issues, which in turn, and at least to a significant degree, reflects the general proclivity in this ecosystem.

show abstract

Empirical Evaluation of Attacks Against IEEE 802.11 Enterprise Networks: The AWID3 Dataset

Cited by 57 publications

References 15 publications

Automating In-Network Machine Learning

Automating In-Network Machine Learning

A review of Federated Learning in Intrusion Detection Systems for IoT

Let the Cat out of the Bag: Popular Android IoT Apps under Security Scrutiny

Contact Info

Product

Resources

About