USENIX Security '19 Fall Accepted Papers

USENIX Security '19 had two submission deadlines. Prepublication versions of the accepted papers from the fall submission deadline are available below. The full program will be available soon. All papers will be available on Wednesday, August 14, 2019.

Computer Security and Privacy in the Interactions Between Victim Service Providers and Human Trafficking Survivors

Christine Chen, University of Washington; Nicola Dell, Cornell Tech; Franziska Roesner, University of Washington

Distinguished Paper Award Winner

Distinguished Paper Award
Available Media

A victim service provider, or VSP, is a crucial partner in a human trafficking survivor’s recovery. VSPs provide or connect survivors to services such as medical care, legal services, employment opportunities, etc. In this work, we study VSP-survivor interactions from a computer security and privacy perspective. Through 17 semi-structured interviews with staff members at VSPs and survivors of trafficking, we surface the role technology plays in VSP-survivor interactions as well as related computer security and privacy concerns and mitigations. Our results highlight various tensions that VSPs must balance, including building trust with their clients (often by giving them as much autonomy as possible) while attempting to guide their use of technology to mitigate risks around revictimization. We conclude with concrete recommendations for computer security and privacy technologists who wish to partner with VSPs to support and empower trafficking survivors.

UWB-ED: Distance Enlargement Attack Detection in Ultra-Wideband

Mridula Singh, Patrick Leu, AbdelRahman Abdou, and Srdjan Capkun, ETH Zurich

Available Media

Mobile autonomous systems, robots, and cyber-physical systems rely on accurate positioning information. To conduct distance-measurement, two devices exchange signals and, knowing these signals propagate at the speed of light, the time of arrival is used for distance estimations. Existing distance-measurement techniques are incapable of protecting against adversarial distance enlargement---a highly devastating tactic in which the adversary reissues a delayed version of the signals transmitted between devices, after distorting the authentic signal to prevent the receiver from identifying it. The adversary need not break crypto, nor compromise any upper-layer security protocols for mounting this attack. No known solution currently exists to protect against distance enlargement. We present \textit{Ultra-Wideband Enlargement Detection} (UWB-ED), a new modulation technique to detect distance enlargement attacks, and securely verify distances between two mutually trusted devices. We analyze UWB-ED under an adversary that injects signals to block/modify authentic signals. We show how UWB-ED is a good candidate for 802.15.4z Low Rate Pulse and the 5G standard.

The KNOB is Broken: Exploiting Low Entropy in the Encryption Key Negotiation Of Bluetooth BR/EDR

Daniele Antonioli, SUTD; Nils Ole Tippenhauer, CISPA; Kasper B. Rasmussen, University of Oxford

Available Media

We present an attack on the encryption key negotiation protocol of Bluetooth BR/EDR. The attack allows a third party, without knowledge of any secret material (such as link and encryption keys), to make two (or more) victims agree on an encryption key with only 1 byte (8 bits) of entropy. Such low entropy enables the attacker to easily brute force the negotiated encryption keys, decrypt the eavesdropped ciphertext, and inject valid encrypted messages (in real-time). The attack is stealthy because the encryption key negotiation is transparent to the Bluetooth users. The attack is standard-compliant because all Bluetooth BR/EDR versions require to support encryption keys with entropy between 1 and 16 bytes and do not secure the key negotiation protocol. As a result, the attacker completely breaks Bluetooth BR/EDR security without being detected. We call our attack Key Negotiation Of Bluetooth (KNOB) attack.

The attack targets the firmware of the Bluetooth chip because the firmware (Bluetooth controller) implements all the security features of Bluetooth BR/EDR. As a standard-compliant attack, it is expected to be effective on any firmware that follows the specification and on any device using a vulnerable firmware. We describe how to perform the KNOB attack, and we implement it. We evaluate our implementation on more than 14 Bluetooth chips from popular manufacturers such as Intel, Broadcom, Apple, and Qualcomm. Our results demonstrate that all tested devices are vulnerable to the KNOB attack. We discuss countermeasures to fix the Bluetooth specification and its implementation.

Back to the Whiteboard: a Principled Approach for the Assessment and Design of Memory Forensic Techniques

Fabio Pagani and Davide Balzarotti, EURECOM

Available Media

Today memory analysis plays a fundamental role in computer forensics and is a very active area of research. However, the field is still largely driven by custom rules and heuristics handpicked by human experts. These rules describe how to overcome the semantic gap to associate high level structures to individual bytes contained in a physical memory dump. Structures are then traversed by following pointers to other objects, and the process is repeated until the required information is located and extracted from the memory image.

A fundamental problem with this approach is that we have no way to measure these heuristics to know precisely how well they work, under which circumstances, how prone they are to evasions or to errors, and how stable they are over different versions of the OS kernel. In addition, without a method to measure the quality and effectiveness of a given heuristic, it is impossible to compare one approach against the others. If a tool adopts a certain heuristic to list the sockets associated to a program, how do we know if that is the only possible way to extract this information? Maybe other, even better, solutions exist, just waiting to be "discovered'' by human analysts.

For this reason, we believe we need to go back to the drawing board and rethink memory forensics from its foundations. In this paper we propose a framework and a set of metrics we can use as a basis to assess existing methodologies, understand their characteristics and limitations, and propose new techniques in a principled way. The memory of a modern operating system is a very large and very complex network of interconnected objects. Because of this, we argue that automated algorithms, rather than human intuition, should play a fundamental role in evaluating and designing future memory forensics techniques.

Point Break: A Study of Bandwidth Denial-of-Service Attacks against Tor

Rob Jansen, U.S. Naval Research Laboratory; Tavish Vaidya and Micah Sherr, Georgetown University

Available Media

As the Tor network has grown in popularity and importance as a tool for privacy-preserving online communication, it has increasingly become a target for disruption, censorship, and attack. A large body of existing work examines Tor's susceptibility to attacks that attempt to block Tor users' access to information (e.g., via traffic filtering), identify Tor users' communication content (e.g., via traffic fingerprinting), and de-anonymize Tor users (e.g., via traffic correlation). This paper focuses on the relatively understudied threat of denial-of-service (DoS) attacks against Tor, and specifically, DoS attacks that intelligently utilize bandwidth as a means to significantly degrade Tor network performance and reliability.

We demonstrate the feasibility of several bandwidth DoS attacks through live-network experimentation and high-fidelity simulation while quantifying the cost of each attack and its effect on Tor performance. First, we explore an attack against Tor's most commonly used default bridges (for censorship circumvention) and estimate that flooding those that are operational would cost $17K/mo. and could reduce client throughput by 44% while more than doubling bridge maintenance costs. Second, we explore attacks against the TorFlow bandwidth measurement system and estimate that a constant attack against all TorFlow scanners would cost $2.8K/mo. and reduce the median client download rate by 80%. Third, we explore how an adversary could use Tor to congest itself and estimate that such a congestion attack against all Tor relays would cost $1.6K/mo. and increase the median client download time by 47%. Finally, we analyze the effects of Sybil DoS and deanonymization attacks that have costs comparable to those of our attacks.

Reading the Tea leaves: A Comparative Analysis of Threat Intelligence

Vector Guo Li, University of California, San Diego; Matthew Dunn, Northeastern University; Paul Pearce, Georgia Tech; Damon McCoy, New York University; Geoffrey M. Voelker and Stefan Savage, University of California, San Diego; Kirill Levchenko, University of Illinois Urbana-Champaign

Available Media

The term "threat intelligence" has swiftly become a staple buzzword in the computer security industry. The entirely reasonable premise is that, by compiling up-to-date information about known threats (i.e., IP addresses, domain names, file hashes, etc.), recipients of such information may be able to better defend their systems from future attacks. Thus, today a wide array of public and commercial sources distribute threat intelligence data feeds to support this purpose. However, our understanding of this data, its characterization and the extent to which it can meaningfully support its intended uses, is still quite limited. In this paper, we address these gaps by formally defining a set of metrics for characterizing threat intelligence data feeds and using these measures to systematically characterize a broad range of public and commercial sources. Further, we ground our quantitative assessments using external measurements to qualitatively investigate issues of coverage and accuracy. Unfortunately, our measurement results suggest that there are significant limitations and challenges in using existing threat intelligence data for its purported goals.

Rendered Private: Making GLSL Execution Uniform to Prevent WebGL-based Browser Fingerprinting

Shujiang Wu, Song Li, and Yinzhi Cao, Johns Hopkins University; Ningfei Wang, Lehigh University

Available Media

Browser fingerprinting, a substitute of cookies-based tracking, extracts a list of client-side features and combines them as a unique identifier for the target browser. Among all these features, one that has the highest entropy and the ability for an even sneakier purpose, i.e., cross-browser fingerprinting, is the rendering of WebGL tasks, which produce different results across different installations of the same browser on different computers, thus being considered as fingerprintable.

Such WebGL-based fingerprinting is hard to defend against, because the client browser executes a program written in OpenGL Shading Language (GLSL). To date, it remains unclear, in either the industry or the research community, about how and why the rendering of GLSL programs could lead to result discrepancies. Therefore, all the existing defenses, such as these adopted by Tor Browser, can only disable WebGL, i.e., a sacrifice of functionality over privacy, to prevent WebGL-based fingerprinting.

In this paper, we propose a novel system, called UNIGL, to rewrite GLSL programs and make uniform WebGL rendering procedure with the support of existing WebGL functionalities. Particularly, we, being the first in the community, point out that such rendering discrepancies in state-of-the-art WebGL-based fingerprinting are caused by floating-point operations. After realizing the cause, we design UNIGL so that it redefines all the floating-point operations, either explicitly written in GLSL programs or implicitly invoked by WebGL, to mitigate the fingerprinting factors.

We implemented a prototype of UNIGL as an open-source browser add-on ( We also created a demo website (, i.e., a modified version of an existing fingerprinting website, which directly integrates our add-on at the server-side to demonstrate the effectiveness of UNIGL. Our evaluation using crowdsourcing workers shows that UNIGL can prevent state-of-the-art WebGL-based fingerprinting with reasonable FPSes.

The CrossPath Attack: Disrupting the SDN Control Channel via Shared Links

Jiahao Cao, Qi Li, and Renjie Xie, Tsinghua University; Kun Sun, George Mason University; Guofei Gu, Texas A&M University; Mingwei Xu and Yuan Yang, Tsinghua University

Available Media

Software-Defined Networking (SDN) enables network innovations with a centralized controller controlling the whole network through the control channel. Because the control channel delivers all network control traffic, its security and reliability are of great importance. For the first time in the literature, we propose the CrossPath attack that disrupts the SDN control channel by exploiting the shared links in paths of control traffic and data traffic. In this attack, crafted data traffic can implicitly disrupt the forwarding of control traffic in the shared links. As the data traffic does not enter the control channel, the attack is stealthy and cannot be easily perceived by the controller. In order to identify the target paths containing the shared links to attack, we develop a novel technique called adversarial path reconnaissance. Both theoretic analysis and experimental results demonstrate its feasibility and efficiency of identifying the target paths. We systematically study the impacts of the attack on various network applications in a real SDN testbed. Experiments show the attack significantly degrades the performance of existing network applications and causes serious network anomalies, e.g., routing blackhole, flow table resetting and even network-wide DoS.

A Billion Open Interfaces for Eve and Mallory: MitM, DoS, and Tracking Attacks on iOS and macOS Through Apple Wireless Direct Link

Milan Stute, Technische Universität Darmstadt; Sashank Narain, Northeastern University; Alex Mariotto, Alexander Heinrich, and David Kreitschmann, Technische Universität Darmstadt; Guevara Noubir, Northeastern University; Matthias Hollick, Technische Universität Darmstadt

Available Media

Apple Wireless Direct Link (AWDL) is a key protocol in Apple's ecosystem used by over one billion iOS and macOS devices for device-to-device communications. AWDL is a proprietary extension of the IEEE 802.11 (Wi-Fi) standard and integrates with Bluetooth Low Energy (BLE) for providing services such as Apple AirDrop. We conduct the first security and privacy analysis of AWDL and its integration with BLE. We uncover several security and privacy vulnerabilities ranging from design flaws to implementation bugs leading to a man-in-the-middle (MitM) attack enabling stealthy modification of files transmitted via AirDrop, denial-of-service (DoS) attacks preventing communication, privacy leaks that enable user identification and long-term tracking undermining MAC address randomization, and DoS attacks enabling targeted or simultaneous crashing of all neighboring devices. The flaws span across AirDrop's BLE discovery mechanism, AWDL synchronization, UI design, and Wi-Fi driver implementation. Our analysis is based on a combination of reverse engineering of protocols and code supported by analyzing patents. We provide proof-of-concept implementations and demonstrate that the attacks can be mounted using a low-cost ($20) micro:bit device and an off-the-shelf Wi-Fi card. We propose practical and effective countermeasures. While Apple was able to issue a fix for a DoS attack vulnerability after our responsible disclosure, the other security and privacy vulnerabilities require the redesign of some of their services.

BITE: Bitcoin Lightweight Client Privacy using Trusted Execution

Sinisa Matetic, Karl Wüst, Moritz Schneider, and Kari Kostiainen, ETH Zurich; Ghassan Karame, NEC Labs; Srdjan Capkun, ETH Zurich

Available Media

Blockchains offer attractive advantages over traditional payments such as the ability to operate without a trusted authority and increased user privacy. However, the verification of blockchain payments requires the user to download and process the entire chain which can be infeasible for resource-constrained devices like mobile phones. To address this problem, most major blockchain systems support so called lightweight clients that outsource most of the computational and storage burden to full blockchain nodes. However, such verification leaks critical information about clients’ transactions, thus defeating user privacy that is often considered one of the main goals of decentralized cryptocurrencies.

In this paper, we propose a new approach to protect the privacy of light clients in Bitcoin. Our main idea is to leverage the trusted execution capabilities of commonly available SGX enclaves. We design and implement a system called BITE where enclaves on full nodes serve privacy-preserving requests from light clients. However, as we will show, naive processing of client requests from within SGX enclaves still leaks client’s addresses and transactions. BITE therefore integrates several private information retrieval and side-channel protection techniques at critical parts of the system. We show that BITE provides significantly improved privacy protection for light clients without compromising the performance of the assisting full nodes.

PAC it up: Towards Pointer Integrity using ARM Pointer Authentication

Hans Liljestrand, Aalto University, Huawei Technologies Oy; Thomas Nyman, Aalto University; Kui Wang, Huawei Technologies Oy, Tampere University of Technology; Carlos Chinea Perez, Huawei Technologies Oy; Jan-Erik Ekberg, Huawei Technologies Oy, Aalto University; N. Asokan, Aalto University

Available Media

Run-time attacks against programs written in memory-unsafe programming languages (e.g., C and C++) remain a prominent threat against computer systems. The prevalence of techniques like return-oriented programming (ROP) in attacking real-world systems has prompted major processor manufacturers to design hardware-based countermeasures against specific classes of run-time attacks. An example is the recently added support for pointer authentication (PA) in the ARMv8-A processor architecture, commonly used in devices like smartphones. PA is a low-cost technique to authenticate pointers so as to resist memory vulnerabilities. It has been shown to enable practical protection against memory vulnerabilities that corrupt return addresses or function pointers. However, so far, PA has received very little attention as a general purpose protection mechanism to harden software against various classes of memory attacks. In this paper, we use PA to build novel defenses against various classes of run-time attacks, including the first PA-based mechanism for data pointer integrity. We present PARTS, an instrumentation framework that integrates our PA-based defenses into the LLVM compiler and the GNU/Linux operating system and show, via systematic evaluation, that PARTS provides better protection than current solutions at a reasonable performance overhead.

Fuzzification: Anti-Fuzzing Techniques

Jinho Jung, Hong Hu, David Solodukhin, and Daniel Pagan, Georgia Institute of Technology; Kyu Hyung Lee, University of Georgia; Taesoo Kim, Georgia Institute of Technology

Available Media

Fuzzing is a software testing technique that quickly and automatically explores the input space of a program without knowing its internals. Therefore, developers commonly use fuzzing as part of test integration throughout the software development process. Unfortunately, it also means that such a blackbox and the automatic natures of fuzzing are appealing to adversaries who are looking for zero-day vulnerabilities.

To solve this problem, we propose a new mitigation approach, called Fuzzification , that helps developers protect the released, binary-only software from attackers who are capable of applying state-of-the-art fuzzing techniques. Given a performance budget, this approach aims to hinder the fuzzing process from adversaries as much as possible. We propose three Fuzzification techniques: 1) SpeedBump, which amplifies the slowdown in normal executions by hundreds of times to the fuzzed execution, 2) BranchTrap, interfering with feedback logic by hiding paths and polluting coverage maps, and 3) AntiHybrid, hindering taint-analysis and symbolic execution. Each technique is designed with best-effort, defensive measures that attempt to hinder adversaries from bypassing Fuzzification .

Our evaluation on popular fuzzers and real-world applications shows that Fuzzification effectively reduces the number of discovered paths by 70.3% and decreases the number of identified crashes by 93.0% from real-world binaries, and decreases the number of detected bugs by 67.5% from LAVA-M dataset while under user-specified overheads for common workloads. We discuss the robustness of Fuzzification techniques against adversarial analysis techniques. We open-source our Fuzzification system to foster future research.

Leaky Images: Targeted Privacy Attacks in the Web

Cristian-Alexandru Staicu and Michael Pradel, TU Darmstadt

Available Media

Sharing files with specific users is a popular service provided by various widely used websites, e.g., Facebook, Twitter, Google, and Dropbox. A common way to ensure that a shared file can only be accessed by a specific user is to authenticate the user upon a request for the file. This paper shows a novel way of abusing shared image files for targeted privacy attacks. In our attack, called leaky images, an image shared with a particular user reveals whether the user is visiting a specific website. The basic idea is simple yet effective: an attacker-controlled website requests a privately shared image, which will succeed only for the targeted user whose browser is logged into the website through which the image was shared. In addition to targeted privacy attacks aimed at single users, we discuss variants of the attack that allow an attacker to track a group of users and to link user identities across different sites. Leaky images require neither JavaScript nor CSS, exposing even privacy-aware users, who disable scripts in their browser, to the leak. Studying the most popular websites shows that the privacy leak affects at least eight of the 30 most popular websites that allow sharing of images between users, including the three most popular of all sites. We disclosed the problem to the affected sites, and most of them have been fixing the privacy leak in reaction to our reports. In particular, the two most popular affected sites, Facebook and Twitter, have already fixed the leaky images problem. To avoid leaky images, we discuss potential mitigation techniques that address the problem at the level of the browser and of the image sharing website.

All Your Clicks Belong to Me: Investigating Click Interception on the Web

Mingxue Zhang and Wei Meng, Chinese University of Hong Kong; Sangho Lee, Microsoft Research; Byoungyoung Lee, Seoul National University and Purdue University; Xinyu Xing, Pennsylvania State University

Available Media

Click is the prominent way that users interact with web applications. For example, we click hyperlinks to navigate among different pages on the Web, click form submission buttons to send data to websites, and click player controls to tune video playback. Clicks are also critical in online advertising, which fuels the revenue of billions of websites. Because of the critical role of clicks in the Web ecosystem, attackers aim to intercept genuine user clicks to either send malicious commands to another application on behalf of the user or fabricate realistic ad click traffic. However, existing studies mainly consider one type of click interceptions in the cross-origin settings via iframes, i.e., clickjacking. This does not comprehensively represent various types of click interceptions that can be launched by malicious third-party JavaScript code.

In this paper, we therefore systematically investigate the click interception practices on the Web. We developed a browser-based analysis framework, Observer, to collect and analyze click related behaviors. Using Observer, we identified three different techniques to intercept user clicks on the Alexa top 250K websites, and detected 437 third-party scripts that intercepted user clicks on 613 websites, which in total receive around 43 million visits on a daily basis.

We revealed that some websites collude with third-party scripts to hijack user clicks for monetization. In particular, our analysis demonstrated that more than 36% of the 3,251 unique click interception URLs were related to online advertising, which is the primary monetization approach on the Web. Further, we discovered that users can be exposed to malicious contents such as scamware through click interceptions. Our research demonstrated that click interception has become an emerging threat to web users.

Evaluating the Contextual Integrity of Privacy Regulation: Parents' IoT Toy Privacy Norms Versus COPPA

Noah Apthorpe, Sarah Varghese, and Nick Feamster, Princeton University

Available Media

Increased concern about data privacy has prompted new and updated data protection regulations worldwide. However, there has been no rigorous way to test whether the practices mandated by these regulations actually align with the privacy norms of affected populations. Here, we demonstrate that surveys based on the theory of contextual integrity provide a quantifiable and scalable method for measuring the conformity of specific regulatory provisions to privacy norms. We apply this method to the U.S. Children’s Online Privacy Protection Act (COPPA), surveying 195 parents and providing the first data that COPPA’s mandates generally align with parents’ privacy expectations for Internet-connected "smart" children’s toys. Nevertheless, variations in the acceptability of data collection across specific smart toys, information types, parent ages, and other conditions emphasize the importance of detailed contextual factors to privacy norms, which may not be adequately captured by COPPA.

Stack Overflow Considered Helpful! Deep Learning Security Nudges Towards Stronger Cryptography

Felix Fischer, Technical University of Munich; Huang Xiao, Bosch Center for Artificial Intelligence; Ching-Yu Kao, Fraunhofer AISEC; Yannick Stachelscheid, Benjamin Johnson, and Danial Raza, Technical University of Munich; Paul Fawkesley and Nat Buckley, Projects by IF; Konstantin Böttinger, Fraunhofer AISEC; Paul Muntean and Jens Grossklags, Technical University of Munich

Available Media

Stack Overflow is the most popular discussion platform for software developers. Recent research found a large amount of insecure encryption code in production systems that has been inspired by examples given on Stack Overflow. By copying and pasting functional code, developers introduced exploitable software vulnerabilities into security-sensitive high-profile applications installed by millions of users every day. Proposed mitigations of this problem suffer from usability flaws and push developers to continue shopping for code examples on Stack Overflow once again. This points us to fighting the proliferation of insecure code directly at the root before it even reaches the clipboard. By viewing Stack Overflow as a market, implementation of cryptography becomes a decision-making problem: i. e. how to simplify the selection of helpful and secure examples. We focus on supporting software developers in making better decisions by applying nudges, a concept borrowed from behavioral science. This approach is motivated by one of our key findings: for 99.37% of insecure code examples on Stack Overflow, similar alternatives are available that serve the same use case and provide strong cryptography. Our system design is based on several nudges that are controlled by a deep neural network. It learns a representation for cryptographic API usage patterns and classification of their security, achieving average AUC-ROC of 0.992. With a user study we demonstrate that nudge-based security advice significantly helps tackling the most popular and error-prone cryptographic use cases in Android.

FastKitten: Practical Smart Contracts on Bitcoin

Poulami Das, Lisa Eckey, Tommaso Frassetto, David Gens, Kristina Hostáková, Patrick Jauernig, Sebastian Faust, and Ahmad-Reza Sadeghi, Technische Universität Darmstadt, Germany

Available Media

Smart contracts are envisioned to be one of the killer applications of decentralized cryptocurrencies. They enable self-enforcing payments between users depending on complex program logic. Unfortunately, Bitcoin – the largest and by far most widely used cryptocurrency – does not offer support for complex smart contracts. Moreover, simple contracts that can be executed on Bitcoin are often cumbersome to design and very costly to execute. In this work we present FastKitten, a practical framework for executing arbitrarily complex smart contracts at low costs over decentralized cryptocurrencies which are designed to only support simple transactions. To this end, FastKitten leverages the power of trusted computing environments (TEEs), in which contracts are run off-chain to enable efficient contract execution at low cost. We formally prove that FastKitten satisfies strong security properties when all but one party are malicious. Finally, we report on a prototype implementation which supports arbitrary contracts through a scripting engine, and evaluate performance through benchmarking a provably fair online poker game. Our implementation illustrates that FastKitten is practical for complex multi-round applications with a very small latency. Combining these features, FastKitten is the first truly practical framework for complex smart contract execution over Bitcoin.

Towards the Detection of Inconsistencies in Public Security Vulnerability Reports

Ying Dong, University of Chinese Academy of Sciences and The Pennsylvania State University; Wenbo Guo, Yueqi Chen, and Xinyu Xing, The Pennsylvania State University and JD Security Research Center; Yuqing Zhang, University of Chinese Academy of Sciences; Gang Wang, Virginia Tech

Available Media

Public vulnerability databases such as Common Vulnerabilities and Exposures (CVE) and National Vulnerability Database (NVD) have achieved a great success in promoting vulnerability disclosure and mitigation. While these databases have accumulated massive data, there is a growing concern for their information quality and consistency.

In this paper, we propose an automated system VIEM to detect inconsistent information between the fully standardized NVD database and the unstructured CVE descriptions and their referenced vulnerability reports. VIEM allows us, for the first time, to quantify the information consistency at a massive scale, and provides the needed tool for the community to keep the CVE/NVD databases up-to date. VIEM is developed to extract vulnerable software names and vulnerable versions from unstructured text. We introduce customized designs to deep-learning-based named entity recognition (NER) and relation extraction (RE) so that VIEM can recognize previous unseen software names and versions based on sentence structure and contexts. Ground-truth evaluation shows the system is highly accurate (0.941 precision and 0.993 recall). Using VIEM, we examine the information consistency using a large dataset of 78,296 CVE IDs and 70,569 vulnerability reports in the past 20 years. Our result suggests that inconsistent vulnerable software versions are highly prevalent. Only 59.82% of the vulnerability reports/CVE summaries strictly match the standardized NVD entries, and the inconsistency level increases over time. Case studies confirm the erroneous information of NVD that either overclaims or underclaims the vulnerable software versions.

Robust Website Fingerprinting Through the Cache Occupancy Channel

Anatoly Shusterman, Ben-Gurion University of the Negev; Lachlan Kang, University of Adelaide; Yarden Haskal and Yosef Meltser, Ben-Gurion University of the Negev; Prateek Mittal, Princeton University; Yossi Oren, Ben-Gurion University of the Negev; Yuval Yarom, University of Adelaide and Data61

Available Media

Website fingerprinting attacks, which use statistical analysis on network traffic to compromise user privacy, have been shown to be effective even if the traffic is sent over anonymity-preserving networks such as Tor. The classical attack model used to evaluate website fingerprinting attacks assumes an on-path adversary, who can observe all traffic traveling between the user's computer and the secure network.

In this work we investigate these attacks under a different attack model, in which the adversary is capable of sending a small amount of malicious JavaScript code to the target user's computer. The malicious code mounts a cache side-channel attack, which exploits the effects of contention on the CPU's cache, to identify other websites being browsed. The effectiveness of this attack scenario has never been systematically analyzed, especially in the open-world model which assumes that the user is visiting a mix of both sensitive and non-sensitive sites.

We show that cache website fingerprinting attacks in JavaScript are highly feasible. Specifically, we use machine learning techniques to classify traces of cache activity. Unlike prior works, which try to identify cache conflicts, our work measures the overall occupancy of the last-level cache. We show that our approach achieves high classification accuracy in both the open-world and the closed-world models. We further show that our attack is more resistant than network-based fingerprinting to the effects of response caching, and that our techniques are resilient both to network-based defenses and to side-channel countermeasures introduced to modern browsers as a response to the Spectre attack. To protect against cache-based website fingerprinting, new defense mechanisms must be introduced to privacy-sensitive browsers and websites. We investigate one such mechanism, and show that generating artificial cache activity reduces the effectiveness of the attack and completely eliminates it when used in the Tor Browser.

XONN: XNOR-based Oblivious Deep Neural Network Inference

M. Sadegh Riazi and Mohammad Samragh, UC San Diego; Hao Chen, Kim Laine, and Kristin Lauter, Microsoft Research; Farinaz Koushanfar, UC San Diego

Available Media

Advancements in deep learning enable cloud servers to provide inference-as-a-service for clients. In this scenario, clients send their raw data to the server to run the deep learning model and send back the results. One standing challenge in this setting is to ensure the privacy of the clients' sensitive data. Oblivious inference is the task of running the neural network on the client's input without disclosing the input or the result to the server. This paper introduces XONN (pronounced /ZAn/), a novel end-to-end framework based on Yao's Garbled Circuits (GC) protocol, that provides a paradigm shift in the conceptual and practical realization of oblivious inference. In XONN, the costly matrix-multiplication operations of the deep learning model are replaced with XNOR operations that are essentially free in GC. We further provide a novel algorithm that customizes the neural network such that the runtime of the GC protocol is minimized without sacrificing the inference accuracy.

We design a user-friendly high-level API for XONN, allowing expression of the deep learning model architecture in an unprecedented level of abstraction. We further provide a compiler to translate the model description from high-level Python (i.e., Keras) to that of XONN. Extensive proof-of-concept evaluation on various neural network architectures demonstrates that XONN outperforms prior art such as Gazelle (USENIX Security'18) by up to 7×, MiniONN (ACM CCS'17) by 93×, and SecureML (IEEE S&P'17) by 37×. State-of-the-art frameworks require one round of interaction between the client and the server for each layer of the neural network, whereas, XONN requires a constant round of interactions for any number of layers in the model. XONN is first to perform oblivious inference on Fitnet architectures with up to 21 layers, suggesting a new level of scalability compared with state-of-the-art. Moreover, we evaluate XONN on four datasets to perform privacy-preserving medical diagnosis. The datasets include breast cancer, diabetes, liver disease, and Malaria.

“Johnny, you are fired!” – Spoofing OpenPGP and S/MIME Signatures in Emails

Jens Müller and Marcus Brinkmann, Ruhr University Bochum; Damian Poddebniak, Münster University of Applied Sciences; Hanno Böck, unaffiliated; Sebastian Schinzel, Münster University of Applied Sciences; Juraj Somorovsky and Jörg Schwenk, Ruhr University Bochum

Available Media

OpenPGP and S/MIME are the two major standards to encrypt and digitally sign emails. Digital signatures are supposed to guarantee authenticity and integrity of messages. In this work we show practical forgery attacks against various implementations of OpenPGP and S/MIME email signature verification in five attack classes: (1) We analyze edge cases in S/MIME's container format. (2) We exploit in-band signaling in the GnuPG API, the most widely used OpenPGP implementation. (3) We apply MIME wrapping attacks that abuse the email clients' handling of partially signed messages. (4) We analyze weaknesses in the binding of signed messages to the sender identity. (5) We systematically test email clients for UI redressing attacks. Our attacks allow the spoofing of digital signatures for arbitrary messages in 14 out of 20 tested OpenPGP-capable email clients and 15 out of 22 email clients supporting S/MIME signatures. While the attacks do not target the underlying cryptographic primitives of digital signatures, they raise concerns about the actual security of OpenPGP and S/MIME email applications. Finally, we propose mitigation strategies to counter these attacks.

TESSERACT: Eliminating Experimental Bias in Malware Classification across Space and Time

Feargus Pendlebury, Fabio Pierazzi, and Roberto Jordaney, King's College London & Royal Holloway, University of London; Johannes Kinder, Bundeswehr University Munich; Lorenzo Cavallaro, King's College London

Available Media

Is Android malware classification a solved problem? Published F1 scores of up to 0.99 appear to leave very little room for improvement. In this paper, we argue that results are commonly inflated due to two pervasive sources of experimental bias: "spatial bias" caused by distributions of training and testing data that are not representative of a real-world deployment; and "temporal bias" caused by incorrect time splits of training and testing sets, leading to impossible configurations. We propose a set of space and time constraints for experiment design that eliminate both sources of bias. We introduce a new metric that summarizes the expected robustness of a classifier in a real-world setting, and we present an algorithm to tune its performance. Finally, we demonstrate how this allows us to evaluate mitigation strategies for time decay such as active learning. We have implemented our solutions in TESSERACT, an open source evaluation framework that allows for the comparison of malware classifiers in a realistic setting. We used TESSERACT to evaluate three Android malware classifiers from the literature on a dataset of 129K applications spanning over three years. Our evaluation confirms that earlier results were biased, but also reveals counterintuitive performance and shows that appropriate tuning can lead to significant improvements.

Scalable Scanning and Automatic Classification of TLS Padding Oracle Vulnerabilities

Robert Merget and Juraj Somorovsky, Ruhr University Bochum; Nimrod Aviram, Tel Aviv University; Craig Young, Tripwire VERT; Janis Fliegenschmidt and Jörg Schwenk, Ruhr University Bochum; Yuval Shavitt, Tel Aviv University

Available Media

The TLS protocol provides encryption, data integrity, and authentication on the modern Internet. Despite the protocol's importance, currently-deployed TLS versions use obsolete cryptographic algorithms which have been broken using various attacks. One prominent class of such attacks is CBC padding oracle attacks. These attacks allow an adversary to decrypt TLS traffic by observing different server behaviors which depend on the validity of CBC padding.

We present the first large-scale scan for CBC padding oracle vulnerabilities in TLS implementations on the modern Internet. Our scan revealed vulnerabilities in 1.83% of the Alexa Top Million websites, detecting nearly 100 different vulnerabilities. Our scanner observes subtle differences in server behavior, such as responding with different TLS alerts, or with different TCP header flags.

We used a novel scanning methodology consisting of three steps. First, we created a large set of probes that detect vulnerabilities at a considerable scanning cost. We then reduced the number of probes using a preliminary scan, such that a smaller set of probes has the same detection rate but is small enough to be used in large-scale scans. Finally, we used the reduced set to scan at scale, and clustered our findings with a novel approach using graph drawing algorithms.

Contrary to common wisdom, exploiting CBC padding oracles does not necessarily require performing precise timing measurements. We detected vulnerabilities that can be exploited simply by observing the content of different server responses. These vulnerabilities pose a significantly larger threat in practice than previously assumed.

When the Signal is in the Noise: Exploiting Diffix's Sticky Noise

Andrea Gadotti and Florimond Houssiau, Imperial College London; Luc Rocher, Imperial College London and Université catholique de Louvain; Benjamin Livshits and Yves-Alexandre de Montjoye, Imperial College London

Available Media

Anonymized data is highly valuable to both businesses and researchers. A large body of research has however shown the strong limits of the de-identification release-and-forget model, where data is anonymized and shared. This has led to the development of privacy-preserving query-based systems. Based on the idea of "sticky noise", Diffix has been recently proposed as a novel query-based mechanism satisfying alone the EU Article 29 Working Party's definition of anonymization. According to its authors, Diffix adds less noise to answers than solutions based on differential privacy while allowing for an unlimited number of queries.

This paper presents a new class of noise-exploitation attacks, exploiting the noise added by the system to infer private information about individuals in the dataset. Our first differential attack uses samples extracted from Diffix in a likelihood ratio test to discriminate between two probability distributions. We show that using this attack against a synthetic best-case dataset allows us to infer private information with 89.4% accuracy using only 5 attributes. Our second cloning attack uses dummy conditions that conditionally strongly affect the output of the query depending on the value of the private attribute. Using this attack on four real-world datasets, we show that we can infer private attributes of at least 93% of the users in the dataset with accuracy between 93.3% and 97.1%, issuing a median of 304 queries per user. We show how to optimize this attack, targeting 55.4% of the users and achieving 91.7% accuracy, using a maximum of only 32 queries per user.

Our attacks demonstrate that adding data-dependent noise, as done by Diffix, is not sufficient to prevent inference of private attributes. We furthermore argue that Diffix alone fails to satisfy Art. 29 WP's definition of anonymization. We conclude by discussing how non-provable privacy-preserving systems can be combined with fundamental security principles such as defense-in-depth and auditability to build practically useful anonymization systems without relying on differential privacy.

AntiFuzz: Impeding Fuzzing Audits of Binary Executables

Emre Güler, Cornelius Aschermann, Ali Abbasi, and Thorsten Holz, Ruhr-Universität Bochum

Available Media

A general defense strategy in computer security is to increase the cost of successful attacks in both computational resources as well as human time. In the area of binary security, this is commonly done by using obfuscation methods to hinder reverse engineering and the search for software vulnerabilities. However, recent trends in automated bug finding changed the modus operandi. Nowadays it is very common for bugs to be found by various fuzzing tools. Due to ever-increasing amounts of automation and research on better fuzzing strategies, large-scale, dragnet-style fuzzing of many hundreds of targets becomes viable. As we show, current obfuscation techniques are aimed at increasing the cost of human understanding and do little to slow down fuzzing. In this paper, we introduce several techniques to protect a binary executable against an analysis with automated bug finding approaches that are based on fuzzing, symbolic/concolic execution, and taint-assisted fuzzing (commonly known as hybrid fuzzing). More specifically, we perform a systematic analysis of the fundamental assumptions of bug finding tools and develop general countermeasures for each assumption. Note that these techniques are not designed to target specific implementations of fuzzing tools, but address general assumptions that bug finding tools necessarily depend on. Our evaluation demonstrates that these techniques effectively impede fuzzing audits, while introducing a negligible performance overhead. Just as obfuscation techniques increase the amount of human labor needed to find a vulnerability, our techniques render automated fuzzing-based approaches futile.