Biblio

With the end of CPU core scaling due to dark silicon limitations, customized accelerators on FPGAs have gained increased attention in modern datacenters due to their lower power, high performance and energy efficiency. Evidenced by Microsoft's FPGA deployment in its Bing search engine and Intel's 16.7 billion acquisition of Altera, integrating FPGAs into datacenters is considered one of the most promising approaches to sustain future datacenter growth. However, it is quite challenging for existing big data computing systems—like Apache Spark and Hadoop—to access the performance and energy benefits of FPGA accelerators. In this paper we design and implement Blaze to provide programming and runtime support for enabling easy and efficient deployments of FPGA accelerators in datacenters. In particular, Blaze abstracts FPGA accelerators as a service (FaaS) and provides a set of clean programming APIs for big data processing applications to easily utilize those accelerators. Our Blaze runtime implements an FaaS framework to efficiently share FPGA accelerators among multiple heterogeneous threads on a single node, and extends Hadoop YARN with accelerator-centric scheduling to efficiently share them among multiple computing tasks in the cluster. Experimental results using four representative big data applications demonstrate that Blaze greatly reduces the programming efforts to access FPGA accelerators in systems like Apache Spark and YARN, and improves the system throughput by 1.7× to 3× (and energy efficiency by 1.5× to 2.7×) compared to a conventional CPU-only cluster.

It is estimated that 50% of the global population lives in urban areas occupying just 0.4% of the Earth's surface. Understanding urban activity constitutes monitoring population density and its changes over time, in urban environments. Currently, there are limited mechanisms to non-intrusively monitor population density in real-time. The pervasive use of cellular phones in urban areas is one such mechanism that provides a unique opportunity to study population density by monitoring the mobility patterns in near real-time. Cellular carriers such as AT&T harvest such data through their cell towers; however, this data is proprietary and the carriers restrict access, due to privacy concerns. In this work, we propose a system that passively senses the population density and infers mobility patterns in an urban area by monitoring power spectral density in cellular frequency bands using periodic beacons from each cellphone without knowing who and where they are located. A wireless sensor network platform is being developed to perform spectral monitoring along with environmental measurements. Algorithms are developed to generate real-time fine-resolution population estimates.

Cache side-channel attacks have been extensively studied on x86 architectures, but much less so on ARM processors. The technical challenges to conduct side-channel attacks on ARM, presumably, stem from the poorly documented ARM cache implementations, such as cache coherence protocols and cache flush operations, and also the lack of understanding of how different cache implementations will affect side-channel attacks. This paper presents a systematic exploration of vectors for flush-reload attacks on ARM processors. flush-reload attacks are among the most well-known cache side-channel attacks on x86. It has been shown in previous work that they are capable of exfiltrating sensitive information with high fidelity. We demonstrate in this work a novel construction of flush-reload side channels on last-level caches of ARM processors, which, particularly, exploits return-oriented programming techniques to reload instructions. We also demonstrate several attacks on Android OS (e.g., detecting hardware events and tracing software execution paths) to highlight the implications of such attacks for Android devices.

Diversity has been suggested as an effective alternative to the current trend in rules-based approaches to cybersecurity. However, little work to date has focused on how various techniques generalize to new attacks. That is, there is no accepted methodology that researchers use to evaluate diversity techniques. Starting with the hypothesis that an attacker's effort increases as the common set of executable code snippets (return-oriented programming (ROP) gadgets) decreases across application variants, we explore how different diversification techniques affect the set of ROP gadgets that is available to an attacker. We show that a small population of diversified variants is sufficient to eliminate 90-99% of ROP gadgets across a collection of real-world applications. Finally, we observe that the number of remaining gadgets may still be sufficient for an attacker to mount an effective attack regardless of the presence of software diversity.

Because of rampant security breaches in IoT devices, searching vulnerabilities in massive IoT ecosystems is more crucial than ever. Recent studies have demonstrated that control-flow graph (CFG) based bug search techniques can be effective and accurate in IoT devices across different architectures. However, these CFG-based bug search approaches are far from being scalable to handle an enormous amount of IoT devices in the wild, due to their expensive graph matching overhead. Inspired by rich experience in image and video search, we propose a new bug search scheme which addresses the scalability challenge in existing cross-platform bug search techniques and further improves search accuracy. Unlike existing techniques that directly conduct searches based upon raw features (CFGs) from the binary code, we convert the CFGs into high-level numeric feature vectors. Compared with the CFG feature, high-level numeric feature vectors are more robust to code variation across different architectures, and can easily achieve realtime search by using state-of-the-art hashing techniques. We have implemented a bug search engine, Genius, and compared it with state-of-art bug search approaches. Experimental results show that Genius outperforms baseline approaches for various query loads in terms of speed and accuracy. We also evaluated Genius on a real-world dataset of 33,045 devices which was collected from public sources and our system. The experiment showed that Genius can finish a search within 1 second on average when performed over 8,126 firmware images of 420,558,702 functions. By only looking at the top 50 candidates in the search result, we found 38 potentially vulnerable firmware images across 5 vendors, and confirmed 23 of them by our manual analysis. We also found that it took only 0.1 seconds on average to finish searching for all 154 vulnerabilities in two latest commercial firmware images from D-LINK. 103 of them are potentially vulnerable in these images, and 16 of them were confirmed.

Cryptocurrencies, such as Bitcoin and 250 similar alt-coins, embody at their core a blockchain protocol –- a mechanism for a distributed network of computational nodes to periodically agree on a set of new transactions. Designing a secure blockchain protocol relies on an open challenge in security, that of designing a highly-scalable agreement protocol open to manipulation by byzantine or arbitrarily malicious nodes. Bitcoin's blockchain agreement protocol exhibits security, but does not scale: it processes 3–7 transactions per second at present, irrespective of the available computation capacity at hand. In this paper, we propose a new distributed agreement protocol for permission-less blockchains called ELASTICO. ELASTICO scales transaction rates almost linearly with available computation for mining: the more the computation power in the network, the higher the number of transaction blocks selected per unit time. ELASTICO is efficient in its network messages and tolerates byzantine adversaries of up to one-fourth of the total computational power. Technically, ELASTICO uniformly partitions or parallelizes the mining network (securely) into smaller committees, each of which processes a disjoint set of transactions (or "shards"). While sharding is common in non-byzantine settings, ELASTICO is the first candidate for a secure sharding protocol with presence of byzantine adversaries. Our scalability experiments on Amazon EC2 with up to \$1, 600\$ nodes confirm ELASTICO's theoretical scaling properties.

The blockchain emerges as an innovative tool which proves to be useful in a number of application scenarios. A number of large industrial players, such as IBM, Microsoft, Intel, and NEC, are currently investing in exploiting the blockchain in order to enrich their portfolio of products. A number of researchers and practitioners speculate that the blockchain technology can change the way we see a number of online applications today. Although it is still early to tell for sure, it is expected that the blockchain will stimulate considerable changes to a large number of products and will positively impact the digital experience of many individuals around the globe. In this tutorial, we overview, detail, and analyze the security provisions of Bitcoin and its underlying blockchain-effectively capturing recently reported attacks and threats in the system. Our contributions go beyond the mere analysis of reported vulnerabilities of Bitcoin; namely, we describe and evaluate a number of countermeasures to deter threats on the system-some of which have already been incorporated in the system. Recall that Bitcoin has been forked multiple times in order to fine-tune the consensus (i.e., the block generation time and the hash function), and the network parameters (e.g., the size of blocks). As such, the results reported in this tutorial are not only restricted to Bitcoin, but equally apply to a number of "altcoins" which are basically clones/forks of the Bitcoin source code. Given the increasing number of alternative blockchain proposals, this tutorial extracts the basic security lessons learnt from the Bitcoin system with the aim to foster better designs and analysis of next-generation secure blockchain currencies and technologies.

Password vaults are used to store login credentials, usually encrypted by a master password, relieving the user from memorizing a large number of complex passwords. To manage accounts on multiple devices, vaults are often stored at an online service, which substantially increases the risk of leaking the (encrypted) vault. To protect the master password against guessing attacks, previous work has introduced cracking-resistant password vaults based on Honey Encryption. If decryption is attempted with a wrong master password, they output plausible-looking decoy vaults, thus seemingly disabling offline guessing attacks. In this work, we propose attacks against cracking-resistant password vaults that are able to distinguish between real and decoy vaults with high accuracy and thus circumvent the offered protection. These attacks are based on differences in the generated distribution of passwords, which are measured using Kullback-Leibler divergence. Our attack is able to rank the correct vault into the 1.3% most likely vaults (on median), compared to 37.8% of the best-reported attack in previous work. (Note that smaller ranks are better, and 50% is achievable by random guessing.) We demonstrate that this attack is, to a certain extent, a fundamental problem with all static Natural Language Encoders (NLE), where the distribution of decoy vaults is fixed. We propose the notion of adaptive NLEs and demonstrate that they substantially limit the effectiveness of such attacks. We give one example of an adaptive NLE based on Markov models and show that the attack is only able to rank the decoy vaults with a median rank of 35.1%.

Password vaults are used to store login credentials, usually encrypted by a master password, relieving the user from memorizing a large number of complex passwords. To manage accounts on multiple devices, vaults are often stored at an online service, which substantially increases the risk of leaking the (encrypted) vault. To protect the master password against guessing attacks, previous work has introduced cracking-resistant password vaults based on Honey Encryption. If decryption is attempted with a wrong master password, they output plausible-looking decoy vaults, thus seemingly disabling offline guessing attacks. In this work, we propose attacks against cracking-resistant password vaults that are able to distinguish between real and decoy vaults with high accuracy and thus circumvent the offered protection. These attacks are based on differences in the generated distribution of passwords, which are measured using Kullback-Leibler divergence. Our attack is able to rank the correct vault into the 1.3% most likely vaults (on median), compared to 37.8% of the best-reported attack in previous work. (Note that smaller ranks are better, and 50% is achievable by random guessing.) We demonstrate that this attack is, to a certain extent, a fundamental problem with all static Natural Language Encoders (NLE), where the distribution of decoy vaults is fixed. We propose the notion of adaptive NLEs and demonstrate that they substantially limit the effectiveness of such attacks. We give one example of an adaptive NLE based on Markov models and show that the attack is only able to rank the decoy vaults with a median rank of 35.1%.

The IoT will host a large number of co-existing cyber-physical applications. Continuous change, application interference, environment dynamics and uncertainty lead to complex effects which must be controlled to give performance and application guarantees. Application and platform self-configuration and self-awareness are one paradigm to approach this challenge. They can leverage context knowledge to control platform and application functions and their interaction. They could play a dominant role in large scale cyber-physical systems and systems-of-systems, simply because no person can oversee the whole system functionality and dynamics. IoT adds a new dimension because Internet based services will increasingly be used in such system functions. Autonomous vehicles accessing cloud services for efficiency and comfort as well as to reach the required level of safety and security are an example. Such vehicle platforms will communicate with a service infrastructure that must be reliable and highly responsive. Automated continuous self-configuration of data storage might be a good basis for such services up to the point where the different self-x strategies might affect each other, in a positive or negative form. This paper contains three contributions from different domains representing the current status of self-aware systems as they will meet in the Internet-of-Things and closes with a short discussion of upcoming challenges.

Current intrusion detection systems (IDS) for industrial control systems (ICS) mostly involve the retrofitting of conventional network IDSs, such as SNORT. Such an approach is prone to missing highly targeted and specific attacks against ICS. Where ICS-specific approaches exist, they often rely on passive network monitoring techniques, offering a low cost solution, and avoiding any computational overhead arising from actively polling ICS devices. However, the use of passive approaches alone could fail in the detection of attacks that alter the behaviour of ICS devices (as was the case in Stuxnet). Where active solutions exist, they can be resource-intensive, posing the risk of overloading legacy devices which are commonplace in ICSs. In this paper we aim to overcome these challenges through the combination of a passive network monitoring approach, and selective active monitoring based on attack vectors specific to an ICS context. We present the implementation of our IDS, SENAMI, for use with Siemens S7 devices. We evaluate the effectiveness of SENAMI in a comprehensive testbed environment, demonstrating validity of the proposed approach through the detection of purely passive attacks at a rate of 99%, and active value tampering attacks at a rate of 81-93%. Crucially, we reach recall values greater than 0.96, indicating few attack scenarios generating false negatives.

Nowadays, both the amount of cyberattacks and their sophistication have considerably increased, and their prevention concerns many organizations. Cooperation by means of information sharing is a promising strategy to address this problem, but unfortunately it poses many challenges. Indeed, looking for a win-win environment is not straightforward and organizations are not properly motivated to share information. This work presents a model to analyse the benefits and drawbacks of information sharing among organizations that present a certain level of dependency. The proposed model applies functional dependency network analysis to emulate attacks propagation and game theory for information sharing management. We present a simulation framework implementing the model that allows for testing different sharing strategies under several network and attack settings. Experiments using simulated environments show how the proposed model provides insights on which conditions and scenarios are beneficial for information sharing.

In recent years, prodigious explosion of social network services may trigger new business models. However, it has negative aspects such as personal information spill or spamming, as well. Amongst conventional spam detection approaches, the studies which are based on vertex degrees or Local Clustering Coefficient have been caused false positive results so that normal vertices can be specified as spammers. In this paper, we propose a novel approach by employing the circuit structure in the social networks, which demonstrates the advantages of our work through the experiment.

Transformations form an important part of developing domain specific languages, where they are used to provide semantics for typing and evaluation. Yet, few solutions exist for verifying transformations written in expressive high-level transformation languages. We take a step towards that goal, by developing a general symbolic execution technique that handles programs written in these high-level transformation languages. We use logical constraints to describe structured symbolic values, including containment, acyclicity, simple unordered collections (sets) and to handle deep type-based querying of syntax hierarchies. We evaluate this symbolic execution technique on a collection of refactoring and model transformation programs, showing that the white-box test generation tool based on symbolic execution obtains better code coverage than a black box test generator for such programs in almost all tested cases.

Recent incidents have once again brought the topic of encryption to public discourse, while researchers continue to demonstrate attacks that highlight the difficulty of implementing encryption even without the presence of "backdoors". However, apart from the threat of implementation flaws in encryption libraries, another significant threat arises when web services fail to enforce ubiquitous encryption. A recent study explored this phenomenon in popular services, and demonstrated how users are exposed to cookie hijacking attacks with severe privacy implications. Many security mechanisms purport to eliminate this problem, ranging from server-controlled options such as HSTS to user-controlled options such as HTTPS Everywhere and other browser extensions. In this paper, we create a taxonomy of available mechanisms and evaluate how they perform in practice. We design an automated testing framework for these mechanisms, and evaluate them using a dataset of 30 days of HTTP requests collected from the public wireless network of our university's campus. We find that all mechanisms suffer from implementation flaws or deployment issues and argue that, as long as servers continue to not support ubiquitous encryption across their entire domain (including all subdomains), no mechanism can effectively protect users from cookie hijacking and information leakage.

Defect-prediction techniques can enhance the quality assurance activities for software systems. For instance, they can be used to predict bugs in source files or functions. In the context of a software product line, such techniques could ideally be used for predicting defects in features or combinations of features, which would allow developers to focus quality assurance on the error-prone ones. In this preliminary case study, we investigate how defect prediction models can be used to identify defective features using machine-learning techniques. We adapt process metrics and evaluate and compare three classifiers using an open-source product line. Our results show that the technique can be effective. Our best scenario achieves an accuracy of 73 % for accurately predicting features as defective or clean using a Naive Bayes classifier. Based on the results we discuss directions for future work.

Semi-autonomous driver assists are already widely deployed and fully autonomous cars are progressively leaving the realm of laboratories. This evolution coexists with a progressive connectivity and cooperation, creating important safety and security challenges, the latter ranging from casual hackers to highly-skilled attackers, requiring a holistic analysis, under the perspective of fully-fledged ecosystems of autonomous and cooperative vehicles. This position paper attempts at contributing to a better understanding of the global threat plane and the specific threat vectors designers should be attentive to. We survey paradigms and mechanisms that may be used to overcome or at least mitigate the potential risks that may arise through the several threat vectors analyzed.

This paper builds on prior psychological studies that identify signals of trustworthiness between two human negotiators. Unlike prior work, the current work tracks such signals automatically and fuses them into computational models that predict trustworthiness. To achieve this goal, we apply automatic trackers to recordings of human dyads negotiating in a multi-issue bargaining task. We identify behavioral indicators in different modalities (facial expressions, gestures, gaze, and conversational features) that are predictive of trustworthiness. We predict both objective trustworthiness (i.e., are they honest) and perceived trustworthiness (i.e., do they seem honest to their interaction partner). Our experiments show that people are poor judges of objective trustworthiness (i.e., objective and perceived trustworthiness are predicted by different indicators), and that multimodal approaches better predict objective trustworthiness, whereas people overly rely on facial expressions when judging the honesty of their partner. Moreover, domain knowledge (from the literature and prior analysis of behaviors) facilitates the model development process.

Interpersonal trust is one of the key components of efficient teamwork. Research suggests two main approaches for trust formation: personal information exchange (e.g., social icebreakers), and creating a context of risk and interdependence (e.g., trust falls). However, because these strategies are difficult to implement in an online setting, trust is more difficult to achieve and preserve in distributed teams. In this paper, we argue that games are an optimal environment for trust formation because they can simulate both risk and interdependence. Results of our online experiment show that a social game can be more effective than a social task at fostering interpersonal trust. Furthermore, trust formation through the game is reliable, but trust depends on several contingencies in the social task. Our work suggests that gameplay interactions do not merely promote impoverished versions of the rich ties formed through conversation; but rather engender genuine social bonds. \textbackslash

Visible Light Communication (VLC) allows to reuse a lighting infrastructure for communication while its main purpose of illumination can be carried out at the same time. Light sources based on Light Emitting Diodes (LEDs) are attractive as they are inexpensive, ubiquitous, and allow rapid modulation. This paper describes how to integrate smartphones into such a communication system that supports networking for a wide range of devices, such as toys with single LEDs as transmitter and receivers as well as interconnected LED light bulbs. The main challenge is how to employ the smartphone without any (hardware) modification as a receiver, using the integrated camera as a (slow) light sampling device. This paper presents a simple software-based solution, exploiting the rolling shutter effect and slow motion video capturing capabilities of latest smartphones to enable continuous reception and real-time integration into an existing VLC system. Evaluation results demonstrate a working prototype and report communication distances up to 3m and a maximum data throughput of more than 1200b/s, improving upon previous work.

The HTTPS ecosystem, including the SSL/TLS protocol, the X.509 public-key infrastructure, and their cryptographic libraries, is the standardized foundation of Internet Security. Despite 20 years of progress and extensions, however, its practical security remains controversial, as witnessed by recent efforts to improve its design and implementations, as well as recent disclosures of attacks against its deployments. The Everest project is a collaboration between Microsoft Research, INRIA, and the community at large that aims at modelling, programming, and verifying the main HTTPS components with strong machine-checked security guarantees, down to core system and cryptographic assumptions. Although HTTPS involves a relatively small amount of code, it requires efficient low-level programming and intricate proofs of functional correctness and security. To this end, we are also improving our verifications tools (F*, Dafny, Lean, Z3) and developing new ones. In my talk, I will present our project, review our experience with miTLS, a verified reference implementation of TLS coded in F*, and describe current work towards verified, secure, efficient HTTPS.

Process-based isolation, suggested by several research prototypes, is a cornerstone of modern browser security architectures. Google Chrome is the first commercial browser that adopts this architecture. Unlike several research prototypes, Chrome's process-based design does not isolate different web origins, but primarily promises to protect "the local system" from "the web". However, as billions of users now use web-based cloud services (e.g., Dropbox and Google Drive), which are integrated into the local system, the premise that browsers can effectively isolate the web from the local system has become questionable. In this paper, we argue that, if the process-based isolation disregards the same-origin policy as one of its goals, then its promise of maintaining the "web/local system (local)" separation is doubtful. Specifically, we show that existing memory vulnerabilities in Chrome's renderer can be used as a stepping-stone to drop executables/scripts in the local file system, install unwanted applications and misuse system sensors. These attacks are purely data-oriented and do not alter any control flow or import foreign code. Thus, such attacks bypass binary-level protection mechanisms, including ASLR and in-memory partitioning. Finally, we discuss various full defenses and present a possible way to mitigate the attacks presented.

The security of order-revealing encryption (ORE) has been unclear since its invention. Dataset characteristics for which ORE is especially insecure have been identified, such as small message spaces and low-entropy distributions. On the other hand, properties like one-wayness on uniformly-distributed datasets have been proved for ORE constructions. This work shows that more plaintext information can be extracted from ORE ciphertexts than was previously thought. We identify two issues: First, we show that when multiple columns of correlated data are encrypted with ORE, attacks can use the encrypted columns together to reveal more information than prior attacks could extract from the columns individually. Second, we apply known attacks, and develop new attacks, to show that the leakage of concrete ORE schemes on non-uniform data leads to more accurate plaintext recovery than is suggested by the security theorems which only dealt with uniform inputs.

With its high penetration rate and relatively good clock accuracy, smartphones are replacing watches in several market segments. Modern smartphones have more than one clock source to complement each other: NITZ (Network Identity and Time Zone), NTP (Network Time Protocol), and GNSS (Global Navigation Satellite System) including GPS. NITZ information is delivered by the cellular core network, indicating the network name and clock information. NTP provides a facility to synchronize the clock with a time server. Among these clock sources, only NITZ and NTP are updated without user interaction, as location services require manual activation. In this paper, we analyze security aspects of these clock sources and their impact on security features of modern smartphones. In particular, we investigate NITZ and NTP procedures over cellular networks (2G, 3G and 4G) and Wi-Fi communication respectively. Furthermore, we analyze several European, Asian, and American cellular networks from NITZ perspective. We identify three classes of vulnerabilities: specification issues in a cellular protocol, configurational issues in cellular network deployments, and implementation issues in different mobile OS's. We demonstrate how an attacker with low cost setup can spoof NITZ and NTP messages to cause Denial of Service attacks. Finally, we propose methods for securely synchronizing the clock on smartphones.