ICT Systems Security and Privacy Protection

Private set intersection (PSI) has a wide range of applications such as privacy-preserving data mining. With the advent of cloud computing it is now desirable to take advantage of the storage and computation capabilities of the cloud to outsource datasets and delegate PSI computation. In this paper we design O-PSI, a protocol for delegated private set intersection on outsourced datasets based on a novel point-value polynomial representation. Our protocol allows multiple clients to independently prepare and upload their private datasets to a server, and then ask the server to calculate their intersection. The protocol ensures that intersections can only be calculated with the permission of all clients and that datasets and results remain completely confidential from the server. Once datasets are outsourced, the protocol supports an unlimited number of intersections with no need to download them or prepare them again for computation. Our protocol is efficient and has computation and communication costs linear to the cardinality of the datasets. We also provide a formal security analysis of the protocol.

Implicit authentication consists of a server authenticating a user based on the user’s usage profile, instead of/in addition to relying on something the user explicitly knows (passwords, private keys, etc.). While implicit authentication makes identity theft by third parties more difficult, it requires the server to learn and store the user’s usage profile. Recently, the first privacy-preserving implicit authentication system was presented, in which the server does not learn the user’s profile. It uses an

ad hoc

two-party computation protocol to compare the user’s fresh sampled features against an encrypted stored user’s profile. The protocol requires storing the usage profile and comparing against it using two different cryptosystems, one of them order-preserving; furthermore, features must be numerical. We present here a simpler protocol based on set intersection that has the advantages of: i) requiring only one cryptosystem; ii) not leaking the relative order of fresh feature samples; iii) being able to deal with any type of features (numerical or non-numerical).

The Minimal-Hitting-Set attack (HS-attack) is a well-known, provably optimal exact attack against the anonymity provided by Chaumian Mixes (Threshold-Mixes). This attack allows an attacker to identify the fixed set of communication partners of a given user by observing all messages sent and received by a Chaum Mix. In contrast to this, the Statistical Disclosure attack (SDA) provides a guess of that user’s contacts, based on statistical analyses of the observed message exchanges.

We contribute the first closed formula that shows the influence of traffic distributions on the least number of observations of the Mix to complete the HS-attack. This measures when the Mix fails to hide a user’s partners, such that the user cannot plausibly deny the identified contacts. It reveals that the HS-attack requires asymptotically less observations to identify a user’s partners than the SDA, which guesses them with a given bias. This number of observations is

$$O(\frac{1}{p})$$

for the HS-attack and

$$O(\frac{1}{p^2})$$

for the SDA, where

$$p$$

the probability that the attacked user contacts his least frequent partner.

A growing concern for end users of Online Social Networks (OSNs) is the privacy and control of user data due to the client-server architecture of the current ecosystems. In this paper, we introduce a privacy preserving decentralized OSN platform, which mimics real life social interactions. In particular, we decentralize the OSN platform and give direct control of the information to the user. The distributed platform removes central authorities from the OSN and users share their content only with intended peers through mobile devices. This decentralized system ensures that interaction happens between friends and third parties cannot access the user content or relationships. To be able to efficiently share objects and provide timely access in the POSN platform, we take advantage of free storage clouds to distribute encrypted user content. The combination of phone-to-phone applications with cloud infrastructure would address the availability limitation of peer-to-peer systems, while enjoying the benefits of peer-to-peer systems, such as no central authority and scalability.

Noninterference is a property that captures confidentiality of actions executed by a given process. However, the property is hard to guarantee in realistic scenarios. We show that the security of a system can be seen as an interplay between functionality requirements and the strategies adopted by users, and based on it we propose a weaker notion of noninterference which we call

strategic noninterference

. We also give a characterization of strategic noninterference through unwinding relations for specific subclasses of goals and for the simplified setting where a strategy is given as a parameter.

This paper proposes an approach to verify information flow security of concurrent programs. It discusses a hyperproperty called observational determinism which aims to ensure secure information flow in concurrent programs, and proves how this hyperproperty can be verified by stutter equivalence checking. More precisely, it defines observational determinism in terms of stutter equivalence of all traces having the same low initial value and shows how stutter trace equivalence can be verified by computing a divergence stutter bisimulation quotient. The approach is illustrated by verifying a small example.

Cache Timing Attacks (CTAs) have been shown to leak Web browsing history. Until recently, they were deemed a limited threat to individuals’ privacy because of their narrow attack surface and vectors, and a lack of robustness and efficiency. Our attack implementation exploits the Web Worker APIs to parallelise cache probing (300 requests/second) and applies time-outs on cache requests to prevent cache pollution. We demonstrate robust cache attacks at the browser, operating system and Web proxy level. Private browsing sessions, HTTPS and corporate intranets are vulnerable. Through case studies of (1) anti-phishing protection in online banking, (2) Web search using the address bar in browsers, (3) publishing of personal images in social media, and (4) use of desktop search, we show that CTAs can seriously compromise privacy and security of individuals and organisations. Options for protection from CTAs are limited. The lack of effective defence, and the ability to mount attacks without cooperation of other websites, makes the improved CTAs serious contenders for cyber-espionage and a broad consumer and corporate surveillance.

For using credential-based access control effectively, recent work identified the need to enforce usage constraints also on credentials. The enforcement of such constraints has not yet been investigated for web applications, although it is relevant when credential-based access control is employed in a web application. This article proposes an approach suitable for enforcing usage constraints on credentials in web applications. More concretely, we present a novel algorithm and an implementation of this algorithm that construct constraint-compliant proofs for credential-based access control policies. We proved that our solution is correct and showed that it is also efficient through extensive experiments.

In the recent years an incredible amount of data has been leaked from major websites such as

Adobe

,

Snapchat

and

LinkedIn

. There are hundreds of millions of usernames, email addresses, passwords, telephone numbers and credit card details in the wild. The aftermath of these breaches is the rise of

alerting websites

such as

http://haveibeenpwned.com

, which let users verify if their accounts have been compromised. Unfortunately, these seemingly innocuous websites can be easily turned into phishing tools. In this work, we provide a comprehensive study of the most popular ones. Our study exposes the associated privacy risks and evaluates existing solutions towards designing

privacy-friendly alerting websites

. In particular, we study three solutions: private set intersection, private set intersection cardinality and private information retrieval adapted to membership testing. Finally, we investigate the practicality of these solutions with respect to real world database leakages.

In this paper, a data structure to enhance remote authentication is proposed generalizing the concept of ISO/IEC 24761. Current technologies do not provide sufficient information on products which are used in the authentication process at the Claimant to the Verifier. As a result, the Verifier cannot sufficiently distinguish the authentication result executed with a trusted product from that without a trusted product. The difference is made clear if an evidence data of the execution of authentication process at the Claimant is generated by the trusted product and used for verification by the Verifier. Data structure for such a data is proposed in this paper as client Authentication Context (cAC) instance. Relation to other works and extension of the proposal are also described for further improvement of remote authentication. For this proposal to realize, standardization activities are to be taken as the next steps.

The use of deception to enhance security has shown promising results as a defensive technique. In this paper we present an authentication scheme that better protects users’ passwords than in currently deployed password-based schemes, without taxing the users’ memory or damaging the user-friendliness of the login process. Our scheme maintains comparability with traditional password-based authentication, without any additional storage requirements, giving service providers the ability to selectively enroll users and fall-back to traditional methods if needed. The scheme utilizes the ubiquity of smartphones; however, unlike previous proposals it does not require registration or connectivity of the phones used. In addition, no long-term secrets are stored in any user’s phone, mitigating the consequences of losing it. Our design significantly increases the difficulty of launching a phishing attack by automating the decisions of whether a website should be trusted and introducing additional risk at the adversary side of being detected and deceived. In addition, the scheme is resilient against Man-in-the-Browser (MitB) attacks and compromised client machines. We also introduce a covert communication mechanism between the user’s client and the service provider. This can be used to covertly and securely communicate the user’s context that comes with the use of this mechanism. The scheme also incorporates the use of deception that makes it possible to dismantle a large-scale attack infrastructure before it succeeds. As an added feature, the scheme gives service providers the ability to have full-transaction authentication.

With the use of our scheme, passwords are no longer communicated in plaintext format to the server, adding another layer of protection when secure channels of communication are compromised. Moreover, it gives service providers the ability to deploy risk-based authentication. It introduces the ability to make dynamic multi-level access decisions requiring extra authentication steps when needed. Finally, the scheme’s covert channel mechanisms give servers the ability to utilize a user’s context information — detecting the use of untrusted networks or whether the login was based on a solicitation email.

The cross-site information sharing and authorized actions of third-party identity management can have significant privacy implications for the users. In this paper, we use a combination of manual analysis of identified third-party identity management relationships and targeted case studies to (i) capture how the protocol usage and third-party selection is changing, (ii) profile what information is requested to be shared (and actions to be performed) between websites, and (iii) identify privacy issues and practical problems that occur when using multiple accounts (associated with these services). By characterizing and quantifying the third-party relationships based on their cross-site information sharing, the study highlights differences in the privacy leakage risks associated with different classes of websites, and provides concrete evidence for how the privacy risks are increasing. For example, many news and file/video-sharing sites ask users to authorize the site to post information to the third-party website. We also observe a general increase in the breadth of information that is shared across websites, and find that due to usage of multiple third-party websites, in many cases, the user can lose (at least) partial control over which identities they can merge/relate and the information that is shared/posted on their behalf.

Online rating systems are widely accepted as a means for quality assessment on the web, and users increasingly rely on these systems when deciding to purchase an item online. This fact motivates people to manipulate rating systems by posting unfair rating scores for fame or profit. Therefore, both providing useful realistic rating scores as well as detecting unfair behaviours are of very high importance. Existing solutions are mostly majority based, also employing temporal analysis and clustering techniques. However, they are still vulnerable to unfair ratings. They also ignore distance between options, provenance of information and different dimensions of cast rating scores while computing aggregate rating scores and trustworthiness of raters. In this paper, we propose a robust iterative algorithm which leverages the information in the profile of raters, provenance of information and a prorating function for the distance between options to build more robust and informative rating scores for items as well as trustworthiness of raters. We have implemented and tested our rating method using both simulated data as well as three real world datasets. Our tests demonstrate that our model calculates realistic rating scores even in the presence of massive unfair ratings and outperforms well-known ranking algorithms.

Network complexity continues to evolve and more robust measures are required to ensure network integrity and mitigate unauthorized access. A physical-layer (PHY) augmentation to Medium Access Control (MAC) authentication is considered using PHY-based Distinct Native Attribute (DNA) features to form device fingerprints. Specifically, a comparison of waveform-based Radio Frequency DNA (RF-DNA) and Constellation-Based DNA (CB-DNA) fingerprinting methods is provided using unintentional Ethernet cable emissions for 10BASE-T signaling. For the first time a direct comparison is achievable between the two methods given the evaluation uses the same experimentally collected emissions to generate RF-DNA and CB-DNA fingerprints. RF-DNA fingerprinting exploits device dependent features derived from instantaneous preamble responses within communication bursts. For these same bursts, the CB-DNA approach uses device dependent features derived from mapped symbol clusters within an adapted two-dimensional (2D) binary constellation. The evaluation uses 16 wired Ethernet devices from 4 different manufacturers and both Cross-Model (manufacturer) Discrimination (CMD) and Like-Model (serial number) Discrimination (LMD) is addressed. Discrimination is assessed using a Multiple Discriminant Analysis, Maximum Likelihood (MDA/ML) classifier. Results show that both RF-DNA and CB-DNA approaches perform well for CMD with average correct classification of

$$\%C$$

=90% achieved at Signal-to-Noise Ratios of

$$SNR\ge 12.0$$

dB. Consistent with prior related work, LMD discrimination is more challenging with CB-DNA achieving

$$\%C$$

=90.0% at

$$SNR$$

=

$$22.0$$

dB and significantly outperforming RF-DNA which only achieved

$$\%C$$

=56.0% at this same

$$SNR$$

.

Security is a critical concern for any information system. Security properties such as confidentiality, integrity and availability need to be enforced in order to make systems safe. In complex environments, where information systems are composed of a number of heterogeneous subsystems, each must participate in their achievement. Therefore, security integration mechanisms are needed in order to 1) achieve the global security goal and 2) facilitate the analysis of the security status of the whole system. For the specific case of access-control, access-control policies may be found in several components (databases, networks and applications) all, supposedly, working together in order to meet the high level security property. In this work we propose an integration mechanism for access-control policies to enable the analysis of the system security. We rely on model-driven technologies and the XACML standard to achieve this goal.

The File Broadcast Protocol (FBP) was developed as a part of the DETIboot system. DETIboot allows a host to broadcast an operating system image through an 802.11 wireless network to an arbitrary number of receivers. Receivers can load the image and immediately boot a Linux live session. The initial version of FBP had no security mechanisms. In this paper we present an authentication protocol developed for FBP that ensures a correct file distribution from the intended source to the receivers. The performance evaluations have shown that, with the best operational configuration tested, the file download time is increased by less than 5%.

We present WebVisor, an automated tool to derive patterns from malware Command and Control (C&C) server connections. From collective network communications stored on a large-scale malware dataset, WebVisor establishes the underlying patterns among samples of the same malware families (e.g., families in terms of development tools). WebVisor focuses on C&C channels based on the Hypertext Transfer Protocol (HTTP). First, it builds clusters based on the statistical features of the HTTP-based Uniform Resource Locators (URLs) stored in the malware dataset. Then, it conducts a fine-grained, noise-agnostic clustering process, based on the structure and semantic features of the URLs. We present experimental results using a software prototype of WebVisor and real-world malware datasets.

Honeypots are used in IT Security to detect and gather information about ongoing intrusions by presenting an interactive system as attractive target to an attacker. They log all actions of an attacker for further analysis. The longer an attacker interacts with a honeypot, the more valuable information about the attack can be collected. Thus, it should be one of the main goals of a honeypot to stay unnoticed as long as possible. Also, a honeypot should appear to be a valuable target system to motivate attackers to attacks the honeypot. This paper presents a novel honeypot concept (B.Hive) that fulfills both requirements: it protects existing web application in productive use, hence offering an attractive attack target, and it uses a novel technique to conceal the honeypot components such that it is hard to detect the honeypot even by manual inspection. B.Hive does not need configuration or changes of existing web applications, it is web framework agnostic, and it only has a slight impact on the performance of the web application it protects. The evaluation shows that B.Hive can be used to protect the majority of the 10,000 most popular web sites (based on the Alexia Global Top 10,000 list), and that the honeypot cannot be identified by humans.

At a time of rapid business globalisation, it is necessary to understand employee security behaviour within diverse cultural settings. While general deterrence theory has been extensively used in Behavioural Information Security research with the aim to explain the effect of deterrent factors on employees’ security actions, these studies provide inconsistent and even contradictory findings. Therefore, a further examination of deterrent factors in the security context is required. The aim of this study is to contribute to the emerging field of Behavioural Information Security research by investigating how a combination of security countermeasures and cultural factors impact upon employee security behaviour in organisations. A particular focus of this project is to explore the effect of national culture and organisational culture on employee actions as regards information security. Preliminary findings suggest that organisational culture, national culture, and security countermeasures do have an impact upon employee security behaviour.

Employees’ poor compliance with information security policies is a perennial problem for many organizations. Existing research shows that about half of all breaches caused by insiders are accidental, which means that one can question the usefulness of information security policies. In order to support the formulation of practical, from the employees’ perspective, information security policies, we propose eight tentative quality criteria. These criteria were developed using practice-based discourse analysis on three information security policy documents from a health care organisation.

IT security preparedness exercises allow for practical collaborative training, which in turn leads to improved response capabilities to information security incidents for an organization. However, such exercises are not commonly performed in the electric power industry. We have observed a tabletop exercise as performed by three organizations with the aim of understanding challenges of performing such exercises. We argue that challenges met during exercises could affect the response process during a real incident as well, and by improving the exercises the response capabilities would be strengthened accordingly. We found that the response team must be carefully selected to include the right competences and all parties that would be involved in a real incident response process, such as technical, managerial, and business responsible. Further, the main goal of the exercise needs to be well understood among the whole team and the facilitator needs to ensure a certain time pressure to increase the value of the exercise, and both the exercise and existing procedures need to be reviewed. Finally, there are many ways to conduct preparedness exercises. Therefore, organizations need to both optimize current exercise practices and experiment with new ones.

Individuals’ compliance with information security policies is important for the overall security of organizations. It has been suggested that obedience cultures exist in organizations and that social processes and structures play a role for the compliance intentions and compliance behavior of individuals. This paper investigates if individuals’ compliance intention is more homogenous within social groups in the workplace than they are within the workplace overall workplace and the effect these groups have are in line with the theory of planned behavior. The results show that a considerable portion of variance in information security policy compliance intentions is explained by the respondents’ organizational department (15%), professional knowledge area (17%), and the same lunch room (18%). While sizeable and significant effects can be found on intentions the effects on attitudes, norm and perceived behavior control are less clear. The only statistically significant (p<0.05) effect is from department on attitudes and perceived norm, each with 6% explained variance. This suggests that the theory of planned behavior fails to account for factors tied to these types of social groups.

We provide the first formal foundation of

SAND

attack trees which are a popular extension of the well-known attack trees. The

SAND

attack tree formalism increases the expressivity of attack trees by introducing the sequential conjunctive operator

$${\mathtt {SAND}}$$

. This operator enables the modeling of ordered events.

We give a semantics to

SAND

attack trees by interpreting them as sets of series-parallel graphs and propose a complete axiomatization of this semantics. We define normal forms for

SAND

attack trees and a term rewriting system which allows identification of semantically equivalent trees. Finally, we formalize how to quantitatively analyze

SAND

attack trees using attributes.

Text based CAPTCHAs are the de facto method of choice to ensure that humans (rather than automated bots) are interacting with websites. Unfortunately, users often find it inconvenient to read characters and type them in. Image CAPTCHAs provide an alternative that is often preferred to text-based implementations. However, Image CAPTCHAs have their own set of security and usability problems. A key issue is their susceptibility to Reverse Image Search (RIS) and Computer Vision (CV) attacks. In this paper, we present a generalized methodology to transform existing images by applying various noise generation algorithms into variants that are resilient to such attacks. To evaluate the usability/security tradeoff, we conduct a user study to determine if the method can provide “usable” images that meet our security requirements – thus improving the overall security provided by Image CAPTCHAs.

Using data flow tracking technology, one can observe how data flows from inputs (sources) to outputs (sinks) of a software system. It has been proposed [

1

] to do runtime data flow tracking at various layers simultaneously (operating system, application, data base, window manager, etc.), and connect the monitors’ observations to exploit semantic information about the layers to make analyses more precise. This has implications on performance—multiple monitors running in parallel—and on methodology—there needs to be one dedicated monitor per layer. We address both aspects of the problem. We replace a

runtime

monitor at a layer L by its

statically

computed input-output dependencies. At runtime, these relations are used by monitors at other layers to model flows of data through L, thus allowing cross-layer system-wide tracking. We achieve this in three steps: (1) static analysis of the application at layer L, (2) instrumentation of the application’s source and sink instructions and (3) runtime execution of the instrumented application in combination with monitors at other layers. The result allows for system-wide tracking of data dissemination, across and through multiple applications. We implement our solution at the Java Bytecode level, and connect it to a runtime OS-level monitor. In terms of precision and performance, we outperform binary-level approaches

and

can exploit high-level semantics.

Dynamically-linked libraries are widely adopted in application programs to achieve extensibility. However, faults in untrusted libraries could allow an attacker to compromise both integrity and confidentiality of the host system (the main program and trusted libraries), as no protection boundaries are enforced between them. Previous systems address this issue through the technique named data sandboxing that relies on instrumentation to sandbox memory reads and writes in untrusted libraries. However, the instrumentation method causes relatively high overhead due to frequent memory reads in code.

In this paper, we propose an efficient and practical data sandboxing approach (called ISboxing) on contemporary x86 platforms, which sandboxes a memory read/write by directly substituting it with a self-sandboxed and function-equivalent one. Our substitution-based method does not insert any additional instructions into library code and therefore incurs almost no measurable runtime overhead. Our experimental results show that ISboxing incurs only 0.32%/1.54% (average/max) overhead for SPECint2000 and 0.05%/0.24% (average/max) overhead for SFI benchmarks, which indicates a notable performance improvement on prior work.

We present a method to generate automatically exploits for information flow leaks in object-oriented programs. Our approach combines self-composition and symbolic execution to compose an

insecurity formula

for a given information flow policy and a specification of the security level of the program locations. The insecurity formula gives then rise to a model which is used to generate input data for the exploit.

A prototype tool called KEG implementing the described approach for Java programs has been developed, which generates exploits as executable JUnit tests.

Identifying differences between two executable binaries (binary diffing) has compelling security applications, such as software vulnerability exploration, “1-day” exploit generation and software plagiarism detection. Recently, binary diffing based on symbolic execution and constraint solver has been proposed to look for the code pairs with the same semantics, even though they are ostensibly different in syntactics. Such logical-based method captures intrinsic differences of binary code, making it a natural choice to analyze highly-obfuscated malicious program. However, semantics-based binary diffing suffers from significant performance slowdown, hindering it from analyzing large-scale malware samples. In this paper, we attempt to mitigate the high overhead of semantics-based binary diffing with application to malware lineage inference. We first study the key obstacles that contribute to the performance bottleneck. Then we propose

basic blocks fast matching

to speed up semantics-based binary diffing. We introduce an union-find set structure that records semantically equivalent basic blocks. Managing the union-find structure during successive comparisons allows direct reuse of previously computed results. Moreover, we purpose to concretize symbolic formulas and cache equivalence queries to further cut down the invocation times of constraint solver. We have implemented our technique on top of iBinHunt and evaluated it on 12 malware families with respect to the performance improvement when performing intra-family comparisons. Our experimental results show that our methods can accelerate symbolic execution from

$$2.8$$

x to

$$5.3$$

x (with an average

$$4.0$$

x), and reduce constraint solver invocation by a factor of

$$3.0$$

x to

$$6.0$$

x (with an average

$$4.3$$

x).

Recently, code-reuse attack (CRA) is becoming the most prevalent attack vector which reuses fragments of existing code to make up malicious code. Recent studies show that CRAs especially jump-oriented programming (JOP) attacks are hard and costly to detect and protect from, especially on CISC processors. One reason for this is that the instructions of CISC architecture are of variable-length, and lots of unintended but legal instructions can be exploited by starting from in the middle of a legal instruction. This feature of CISC architectures makes the finding of so called

gadgets

for CRAs is much easier than that of RISC architectures. Most of previous studies for mitigating CRA on CISC processors rely on software-only means to tackle the unintended instruction problem, which makes their approaches either very costly or can only be applied under restricted conditions. In this paper, we propose two hardware supported techniques. The first, which is the main contribution of this paper, is to eliminate the execution of an unintended instruction. This technique only requires a few modifications to the processor and operating system. Furthermore, the proposed mechanism has little performance impact on the examined SPEC CPU 2006 benchmarks (-0.093% ~2.993%). Second, we propose using hardware control-flow locking as a complementary technique to our protection mechanism. By using the two techniques together, an attacker will have little chance to carry out CRAs on a CISC processor.

In the last few years, many efforts have been devoted to the development of solutions aiming at ensuring the confidentiality and integrity of data and computations in the cloud. In particular, a recent solution for verifying the integrity of equi-join queries is based on the insertion of checks (markers and twins) whose presence provides probabilistic guarantees on the integrity of the computation. In this paper, we propose an approach for verifying the integrity of

approximate join queries

, which is based on the introduction of a discretized version of the join attribute and on the translation of the approximate join into an equi-join defined over the discrete attribute added to the original relations. The proposed approach guarantees the correctness and completeness of the join result, while causing a limited overhead for the user.

Attribute-based credentials allow a user to prove properties about herself anonymously. Revoking such credentials, which requires singling them out, is hard because it is at odds with anonymity. All revocation schemes proposed to date either sacrifice anonymity altogether, require the parties to be online, or put high load on the user or the verifier. As a result, these schemes are either too complicated for low-powered devices like smart cards or they do not scale. We propose a new revocation scheme that has a very low computational cost for users and verifiers, and does not require users to process updates. We trade only a limited, but well-defined, amount of anonymity to make the first practical revocation scheme that is efficient at large scales and fast enough for smart cards.

Perfect forward secrecy is considered as the most important standard to evaluate a strong authentication scheme. There are many results researched to achieve this property without using hard problems. Recently, the result of Chang et al has some advances such as, the correctness of schemes mutual authentication and session key agreement demonstrated in BAN-logic or the overheads reduction of system implementation. However, in this paper, we prove that their scheme is still vulnerable to impersonation attacks and session key leakage. To overcome those limitations and be practical, we use different notion to propose time efficient scheme conducted in experiment. Our proposed method can be applied for remote user authentication in various scenarios, including systems with user authentication using mobile or wearable devices.

Relying on a trusted third party (TTP) in the design of a security protocol introduces obvious risks. Although the risks can be mitigated by distributing the trust across several parties, it still requires at least one party to be trustworthy. In the domain of exams this is critical because parties typically have conflicting interests, and it may be hard to find an entity who can play the role of a TTP, as recent exam scandals confirm. This paper proposes a new protocol for paper-based and computer-based exams that guarantees several security properties without the need of a TTP. The protocol combines oblivious transfer and visual cryptography to allow candidate and examiner to jointly generate a pseudonym that anonymises the candidate’s test. The pseudonym is revealed only to the candidate when the exam starts. We analyse the protocol formally in ProVerif and prove that it satisfies all the stated security requirements.

Android apps are made of components which can leak information between one another using the ICC mechanism. With the growing momentum of Android, a number of research contributions have led to tools for the intra-app analysis of Android apps. Unfortunately, these state-of-the-art approaches, and the associated tools, have long left out the security flaws that arise across the boundaries of single apps, in the interaction between several apps. In this paper, we present a tool called

ApkCombiner

which aims at reducing an inter-app communication problem to an intra-app inter-component communication problem. In practice,

ApkCombiner

combines different apps into a single apk on which existing tools can

indirectly

perform inter-app analysis. We have evaluated

ApkCombiner

on a dataset of 3,000 real-world Android apps, to demonstrate its capability to support static context-aware inter-app analysis scenarios.

This paper describes a study for assessing how many free Android games with in-app purchases were susceptible to data manipulation via the backup utility. To perform this study, a data set with more than 800 games available in the

Google Play

store was defined. The backup utility, provided by the Android Operating System (OS), was used to backup the app files into a Personal Computer (PC) in order to find and manipulate sensitive data. In the cases where sensitive data was found, the applications were restored and the games tested to assess if the manipulation was successful and if it could be used to the benefit of the user. The results included show that a significant percentage of the analyzed games save the user and app information in plaintext and do not include mechanisms to detect or prevent data from being modified.

With millions of apps provided from official and third-party markets, Android has become one of the most active mobile platforms in recent years. These apps facilitate people’s lives in a broad spectrum of ways but at the same time touch numerous users’ information, raising huge privacy concerns. To prevent leaks of sensitive information, especially from legitimate apps to malicious ones, developers are encouraged to store users’ sensitive data into private folders which are isolated and securely protected. But for non-sensitive data, there is no specific guideline on how to manage them, and in many cases, they are simply stored on public storage which lacks fine-grained access control and is almost open to all apps.

Such storage model appears to be capable of preventing privacy leaks, as long as the sensitive data are correctly identified and kept in private folders by app developers. Unfortunately, this is not true in reality. In this paper, we carry out a thorough study over a number of Android apps to examine how the sensitive data are handled, and the results turn out to be pretty alarming: most of the apps we surveyed fail to handle the data correctly, including extremely popular apps. Among these problematic apps, some directly store the sensitive data into public storage, while others leave non-sensitive data on public storage which could give out users’ private information when being combined with data from other sources. An adversary can exploit these leaks to infer users’ location, friends and other information without requiring any critical permission. We refer to both types of data as “non-shared” data, and argue that Android’s storage model should be refined to protect the non-shared data if they are saved to public storage. In the end, we propose several approaches to mitigate such privacy leaks.

The dual-execution-environment approach (dual-EE) is a trusted model that was defined to allow mobile smart devices to guarantee tamper-resistant execution for highly sensitive applications. Although various solutions implementing dual-EE have been proposed in the literature, this model has not been formalized yet. In this paper, we revisit the dual-EE approach and propose a theoretical framework to systematize the design of dual-EE solutions regarding well-established primitives defined in the Multiple Independent Levels of Security (MILS) architecture. We provide a general classification of the different dual-EE proposals based on their isolation properties. We introduce a comparative framework allowing dual-EE solutions to be evaluated across a common set of criteria. The relevance of our framework is examined by applying it on three technologies, each one represents one category in our classification. Results are consistent and explain some hidden and unexpected properties of each technology. For instance, we find that bare-metal hypervisors are ill-adapted to provide high assurance security even though they might improve the overall security level of the system.

Mobile health (mHealth) apps are an ideal tool for monitoring and tracking long-term health conditions. In this paper, we examine whether mHealth apps succeed in ensuring the privacy, security, and safety of the health data entrusted to them. We investigate 154 apps from Android app stores using both automatic code and metadata analysis and a manual analysis of functionality and data leakage. Our study focuses on hypertension and diabetes, two common health conditions that require careful tracking of personal health data.

We find that many apps do not provide privacy policies or safe communications, are implemented in an insecure fashion, fail basic input validation tests and often have overall low code quality which suggests additional security and safety risks. We conclude with recommendations for App Stores, App developers, and end users.

In this paper, we address the problem of building an anonymized medical database from multiple sources. Our proposed solution defines how to achieve data integration in a heterogeneous network of many clinical institutions, while preserving data utility and patients’ privacy. The contribution of the paper is twofold: Firstly, we propose a secure and scalable cloud eHealth architecture to store and exchange patients’ data for the treatment. Secondly, we present an algorithm for efficient aggregation of the health data for the research purposes from multiple sources independently.

In this work we present an Intrusion Detection (ID) and an Intrusion Prevention (IP) model for Wireless Sensor Networks (WSNs). The attacker’s goal is to compromise the deployment by causing nodes to report faulty sensory information. The defender, who is the WSN’s operator, aims to detect the presence of faulty sensor measurements (ID) and to subsequently recover compromised nodes (IP). In order to address the conflicting interests involved, we adopt a Game Theoretic approach that takes into consideration the strategies of both players and we attempt to identify the presence of Nash Equilibria in the two games. The results are then verified in two simulation contexts: Firstly, we evaluate the model in a middleware-based WSN which uses clustering over a bespoke network stack. Subsequently, we test the model in a simulated IPv6-based sensor deployment. According to the findings, the results of both simulation models confirm the results of the theoretic one.

Building Automation Systems (BAS) are crucial for monitoring and controlling buildings, ranging from small homes to critical infrastructure, such as airports or military facilities. A major concern in this context is the security of BAS communication protocols and devices. The

building automation and control networking

protocol (BACnet) is integrated into products of more than 800 vendors worldwide. However, BACnet devices are vulnerable to attacks. We present a novel solution for the two most important BACnet layers, i.e. those independent of the data link layer technology, namely the network and the application layer. We provide the first implementation and evaluation of traffic normalization for BAS traffic. Our proof of concept code is based on the open source software Snort.

Product counterfeit is a tremendous challenge for vendors in many areas. Particularly important is a prevention of product counterfeit where products like telecommunication devices interact with other systems and thus a malfunctioning of a single device can jeopardize the complete system. This can also deteriorate the reputation of the vendor. Furthermore, violation of intellectual properties can cause financial losses. Detection of product counterfeit can be based on tracking back each device to the production process of the vendor to ensure the product origin. Devices without a verified source can then be considered counterfeit with a high potential to be malicious or of low quality. Vendors already apply vendor-specific security technologies protecting the distribution. These often employ special hardware-based security mechanisms specifically designed for a particular range of products.

This publication shows the usage of the already available Trusted Platform Module to allow for distribution channel protection and to leverage overall security by allowing the secure identification of a specific device. It also explains a few additional Trusted Platform Module functionalities that can be used.