This paper proposes FingerprinTV, a fully automated methodology for extracting fingerprints from the network traffic of smart TV apps and assessing their performance. FingerprinTV (1) installs, repeatedly launches, and collects network traffic from smart TV apps; (2) extracts three different types of network fingerprints for each app, i.e., domain-based fingerprints (DBF), packet-pair-based fingerprints (PBF), and TLS-based fingerprints (TBF); and (3) analyzes the extracted fingerprints in terms of their prevalence, distinctiveness, and sizes. From applying FingerprinTV to the top-1000 apps of the three most popular smart TV platforms, we find that smart TV app network fingerprinting is feasible and effective: even the least prevalent type of fingerprint manifests itself in at least 68% of apps of each platform, and up to 89% of fingerprints uniquely identify a specific app when two fingerprinting techniques are used together. By analyzing apps that exhibit identical fingerprints, we find that these apps often stem from the same developer or “no code” app generation toolkit. Furthermore, we show that many apps that are present on all three platforms exhibit platformspecific fingerprints.

Abstract In this paper, we present a large-scale measurement study of the smart TV advertising and tracking ecosystem. First, we illuminate the network behavior of smart TVs as used in the wild by analyzing network traffic collected from residential gateways. We find that smart TVs connect to well-known and platform-specific advertising and tracking services (ATSes). Second, we design and implement software tools that systematically explore and collect traffic from the top-1000 apps on two popular smart TV platforms, Roku and Amazon Fire TV. We discover that a subset of apps communicate with a large number of ATSes, and that some ATS organizations only appear on certain platforms, showing a possible segmentation of the smart TV ATS ecosystem across platforms. Third, we evaluate the (in)effectiveness of DNS-based blocklists in preventing smart TVs from accessing ATSes. We highlight that even smart TV-specific blocklists suffer from missed ads and incur functionality breakage. Finally, we examine our Roku and Fire TV datasets for exposure of personally identifiable information (PII) and find that hundreds of apps exfiltrate PII to third parties and platform domains. We also find evidence that some apps send the advertising ID alongside static PII values, effectively eliminating the user’s ability to opt out of ad personalization.

Mobile devices play an essential role in the Internet today, and there is an increasing interest in using them as a vantage point for network measurement from the edge. At the same time, these devices store personal, sensitive information, and there is a growing number of applications that leak it. We propose AntMonitor-- the first system of its kind that supports (i) collection of large-scale, semantic-rich network traffic in a way that respects users' privacy preferences and (ii) detection and prevention of leakage of private information in real time. The first property makes AntMonitor a powerful tool for network researchers who want to collect and analyze large-scale yet fine-grained mobile measurements. The second property can work as an incentive for using AntMonitor and contributing data for analysis. As a proof-of-concept, we have developed a prototype of AntMonitor, deployed it to monitor 9 users for 2 months, and collected and analyzed 20 GB of mobile data from 151 applications. Preliminary results show that fine-grained data collected from AntMonitor could enable application classification with higher accuracy than state-of-the-art approaches. In addition, we demonstrated that AntMonitor could help prevent several apps from leaking private information over unencrypted traffic, including phone numbers, emails, and device identifiers.