As third-party cookie blocking is becoming the norm in mainstream web browsers, advertisers and trackers have started to use first-party cookies for tracking. To understand this phenomenon, we conduct a differential measurement study with versus without third-party cookies. We find that first-party cookies are used to store and exfiltrate identifiers to known trackers even when third-party cookies are blocked. As opposed to third-party cookie blocking, first-party cookie blocking is not practical because it would result in major breakage of website functionality. We propose CookieGraph, a machine learning-based approach that can accurately and robustly detect and block first-party tracking cookies. CookieGraph detects first-party tracking cookies with 90.18% accuracy, outperforming the state-of-the-art CookieBlock by 17.31%. We show that CookieGraph is robust against cookie name manipulation, while CookieBlock's accuracy drops by 15.87%. While blocking all first-party cookies results in major breakage on 32% of the sites with SSO logins, and CookieBlock reduces it to 10%, we show that CookieGraph does not cause any major breakage on these sites. Our deployment of CookieGraph shows that first-party tracking cookies are used on 89.86% of the top-million websites. We find that 96.61% of these first-party tracking cookies are in fact ghostwritten by third-party scripts embedded in the first-party context. We also find evidence of first-party tracking cookies being set by fingerprinting scripts. The most prevalent first-party tracking cookies are set by major advertising entities such as Google, Facebook, and TikTok.
Machine-learning (ML) models are increasingly being deployed on edge devices to provide a variety of services. However, their deployment is accompanied by challenges in model privacy and auditability. Model providers want to ensure that (i) their proprietary models are not exposed to third parties; and (ii) be able to get attestations that their genuine models are operating on edge devices in accordance with the service agreement with the user. Existing measures to address these challenges have been hindered by issues such as high overheads and limited capability (processing/secure memory) on edge devices. In this work, we propose GuaranTEE, a framework to provide attestable private machine learning on the edge. GuaranTEE uses Confidential Computing Architecture (CCA), Arm's latest architectural extension that allows for the creation and deployment of dynamic Trusted Execution Environments (TEEs) within which models can be executed. We evaluate CCA's feasibility to deploy ML models by developing, evaluating, and openly releasing a prototype. We also suggest improvements to CCA to facilitate its use in protecting the entire ML deployment pipeline on edge devices.
In this work, we compare the performance of a passive eavesdropper in 802.11b/n/ac WLAN networks. In particular, we investigate the downlink of 802.11 networks in infrastructure mode (e. g. from an access point to a terminal) using Commercial-Of-The-Shelf (COTS) devices. Recent 802.11n/ac amendments introduced several physical and link layer features, such as MIMO, spatial diversity, and frame aggregation, to increase the throughput and the capacity of the channel. Several information theoretical studies state that some of those 802.11n/ac features (e. g. beamforming) should provide a degradation of performance for a passive eavesdropper. However, the real impact of those features has not yet been analyzed in a practical context and experimentally evaluated. We present a theoretical discussion and a statistical analysis (using path loss models) to estimate the effects of such features on a passive eavesdropper in 802.11n/ac, using 802.11b as a baseline. We use Signal-to-Noise-Ratio (SNR) and Packet-Error-Rate (PER) as our main metrics. We compute lower and upper bounds for the expected SNR difference between 802.11b and 802.11n/ac using high-level wireless channel characteristics. We show that the PER in 802.11n/ac increases up to 98% (compared to 802.11b) at a distance of 20 m between the sender and the eavesdropper. To obtain a PER of 0.5 in 802.11n/ac, the attacker’s maximal distance is reduced by up to 129.5 m compared to 802.11b. We perform an extensive set of experiments, using COTS devices in an indoor office environment, to verify our theoretical estimations. The experimental results validate our predicted effects and show that every amendment add extra resiliency against passive COTS eavesdropping.