English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Millions of vulnerable consumer IoT devices in home networks are the enablerfor cyber crimes putting user privacy and Internet security at risk. Internetservice providers (ISPs) are best poised to play key roles in mitigating risksby automatically inferring active IoT devices per household and notifying usersof vulnerable ones. Developing a scalable inference method that can performrobustly across thousands of home networks is a non-trivial task. This paperfocuses on the challenges of developing and applying data-driven inferencemodels when labeled data of device behaviors is limited and the distribution ofdata changes (concept drift) across time and space domains. Our contributionsare three-fold: (1) We collect and analyze network traffic of 24 types ofconsumer IoT devices from 12 real homes over six weeks to highlight thechallenge of temporal and spatial concept drifts in network behavior of IoTdevices; (2) We analyze the performance of two inference strategies, namely"global inference" (a model trained on a combined set of all labeled data fromtraining homes) and "contextualized inference" (several models each trained onthe labeled data from a training home) in the presence of concept drifts; and(3) To manage concept drifts, we develop a method that dynamically applies the``closest'' model (from a set) to network traffic of unseen homes during thetesting phase, yielding better performance in 20% of scenarios.

Quantifying and Managing Impacts of Concept Drifts on IoT Traffic Inference in Residential ISP Networks