A promising avenue for improving the effectiveness of behavioral-based malware detectors is to leverage two-phase detection mechanisms. Existing problem in two-phase detection is that after the first phase produces borderline decision, suspicious behaviors are not well contained before the second phase completes. This article improvesChameleon, a framework to realize the uncertain environment.Chameleonoffers two environments: standard—for software identified as benign by the first phase, and uncertain—for software received borderline classification from the first phase. The uncertain environment adds obstacles to software execution through random perturbations applied probabilistically. We introduce a dynamic perturbation threshold that can target malware disproportionately more than benign software. We analyzed the effects of the uncertain environment by manually studying 113 software and 100 malware, and found that 92 percent malware and 10 percent benign software disrupted during execution. The results were then corroborated by an extended dataset (5,679 Linux malware samples) on a newer system. Finally, a careful inspection of the benign software crashes revealed some software bugs, highlightingChameleon's potential as a practical complementary anti-malware solution.

Dynamic Information Flow Tracking (DIFT) is a technique that facilitates run-time data-flow analysis on a running process, allowing a system to overcome the limitations of finding data dependencies statically at compilation time. DIFT serves as the backbone for applications including data-flow integrity (DFI). However, previous uses of DIFT towards DFI often have large overhead in terms of hardware, software or both, and often cannot provide fine-granularity tracking for software object, such as variables. To address these limitations, we present FineDIFT as a DFI framework which utilizes DIFT to generate a live data-flow graph of a running process and perform hardware-based assisted analysis at fine-granularity, thus being able to enforce the application’s Data-Flow Graph (DFG). We provide a sample implementation on a RISC-V core with a performance overhead of 5.03% for BEEBS benchmarks and hardware overhead of 6% LUTs and 8% Flip-Flops in the FPGA implementation, if excluding the Content-Addressable Memory (CAM) like structure used for metadata storage. With CAM-like structure being synthesized using FPGA logic, the total hardware overhead is$\approx 2 \times $LUTs and 33% Flip-Flops compared to the original RISC-V core. We also use the real-world application and customized vulnerable application to demonstrate the effectiveness of the proposed framework in protecting computing systems.

Existing malware detectors on safety-critical devices have difficulties in runtime detection due to the performance overhead. In this article, we introduce Propedeutica, a framework for efficient and effective real-time malware detection, leveraging the best of conventional machine learning (ML) and deep learning (DL) techniques. In Propedeutica, all software start executions are considered as benign and monitored by a conventional ML classifier for fast detection. If the software receives a borderline classification from the ML detector (e.g., the software is 50% likely to be benign and 50% likely to be malicious), the software will be transferred to a more accurate, yet performance demanding DL detector. To address spatial-temporal dynamics and software execution heterogeneity, we introduce a novel DL architecture (DeepMalware) for Propedeutica with multistream inputs. We evaluated Propedeutica with 9115 malware samples and 1338 benign software from various categories for the Windows OS. With a borderline interval of [30%, 70%], Propedeutica achieves an accuracy of 94.34% and a false-positive rate of 8.75%, with 41.45% of the samples moved for DeepMalwareanalysis. Even using only CPU, Propedeutica can detect malware within less than 0.1 s.

Malware analysis is an essential task to understand infection campaigns, the behavior of malicious codes, and possible ways to mitigate threats. Malware analysis also allows better assessment of attackers’ capabilities, techniques, and processes. Although a substantial amount of previous work provided a comprehensive analysis of the international malware ecosystem, research on regionalized, country-, and population-specific malware campaigns have been scarce. Moving towards addressing this gap, we conducted a longitudinal (2012-2020) and comprehensive (encompassing an entire population of online banking users) study of MS Windows desktop malware that actually infected Brazilian banks’ users. We found that the Brazilian financial desktop malware has been evolving quickly: it started to make use of a variety of file formats instead of typical PE binaries, relied on native system resources, and abused obfuscation techniques to bypass detection mechanisms. Our study on the threats targeting a significant population on the ecosystem of the largest and most populous country in Latin America can provide invaluable insights that may be applied to other countries’ user populations, especially those in the developing world that might face cultural peculiarities similar to Brazil’s. With this evaluation, we expect to motivate the security community/industry to seriously consider a deeper level of customization during the development of next-generation anti-malware solutions, as well as to raise awareness towards regionalized and targeted Internet threats.

Computer usage behaviour information can be used by anomaly detection algorithms to identify the current user of the computer system for security reasons. However, the data collected in this setup can be binary and very sparse, resulting in poor performance for some widely used anomaly detection methods. In this study, we propose a novel reconstruction criterion inspired by the F1score and the cross-entropy loss, that tackles the class imbalance problem introduced by binary and sparse data distribution with effectively merging reconstruction criterion calculated from vector elements of both positive and negative classes. Our experiments show that the proposed criterion can effectively improve the performance of reconstruction based anomaly detection methods, including both the PCA and the autoencoder.

Dynamic Information Flow Tracking (DIFT), also called Dynamic Taint Analysis (DTA), is a technique for tracking the information as it flows through a program’s execution. Specifically, some inputs or data get tainted and then these taint marks (tags) propagate usually at the instruction-level. While DIFT has been a fundamental concept in computer and network security for the past decade, it still faces open challenges that impede its widespread application in practice; one of them being the indirect flow propagation dilemma: should the tags involved in an indirect flow, e.g., in a control or address dependency, be propagated? Propagating all these tags, as is done for direct flows, leads to overtainting (all taintable objects become tainted), while not propagating them leads to undertainting (information flow becomes incomplete). In this paper, we analytically model that decisioning problem for indirect flows, by considering various tradeoffs including undertainting versus overtainting, importance of heterogeneous code semantics and context. Towards tackling this problem, we design MITOS, a distributed-optimization algorithm, that: decides about the propagation of indirect flows by properly weighting all these tradeoffs, is of low-complexity, is scalable, is able to flexibly adapt to different application scenarios and security needs of large distributed systems. Additionally, MITOS is applicable to most DIFT systems that consider an arbitrary number of tag types, and introduces the key properties of fairness and tag-balancing to the DIFT field. To demonstrate MITOS’s applicability in practice, we implement and evaluate MITOS on top of an open-source DIFT, and we shed light on the open problem. We also perform a case-study scenario with a real in-memory only attack and show that MITOS improves simultaneously (i) system’s spatiotemporal overhead (up to 40%), and (ii) system’s fingerprint on suspected bytes (up to 167%) compared to traditional DIFT, even though these metrics usually conflict.

This paper introduces a game-based framework to compare the usability of authentication methods. The framework uses a dual-task interference technique to determine the usability of authentication methods. In the experiment, subjects participate in a multi-tasking game that simulates a conversation being interrupted by authentication requirements. By simulating a conversation, the goal is to reproduce a real use of authentication, and collect ecologically sound data. Participants also perform each authentication method in a standalone manner, which allows for comparison of the usability under two different cognitive loads. The authentication techniques evaluated represent each of the three main authentication factors, specifically password, fingerprint, and coauthentication. The three aspects of usability used to compare authentication techniques in this framework are efficiency, effectiveness, and satisfaction. An experiment with 43 participants enrolled was conducted to collect data pertaining to these aspects. The results show that fingerprint and coauthentication (both laptop and phone) are the more usable techniques evaluated.

Phishing is fundamental to cyber attacks. This research determined the effect of Internet user age and email content such as weapons of influence (persuasive techniques that attackers can use to lure individuals to fall for an attack) and life domains (a specific topic or aspect of an individual's life that attackers can focus an emails on) on spear-phishing (targeted phishing) susceptibility. One-hundred young and 58 older users received, without their knowledge, daily simulated phishing emails over 21 days. A browser plugin recorded their clicking on links in the emails as an indicator of their susceptibility. Forty-three percent of users fell for the simulated phishing emails, with older women showing the highest susceptibility. While susceptibility in young users declined across the study, susceptibility in older users remained stable. The relative effectiveness of the attacks differed by weapons of influence and life domains with age-group variability. In addition, older compared to young users reported lower susceptibility awareness. These findings support effects of Internet user demographics and email content on susceptibility to phishing and emphasize the need for personalization of the next generation of security solutions.

In-memory injection attacks are extremely challenging to reverse engineer because they operate stealthily without leaving artifacts in the system or in any easily observable events from outside of a virtual machine. Because these attacks perform their actions in memory only, current malware analysis solutions cannot expose their behavior. This paper introduces FAROS^1 a reverse engineering tool for Windows malware analysis based on dynamic information flow tracking (DIFT), which can flag stealthy in-memory-only malware injection attacks by leveraging the synergy of: (i) whole-system taint analysis; (ii) per security policy-based handling of the challenge of indirect flows via the application of tags of different types, and (iii) the use of tags with fine-grained provenance information. We evaluated FAROS with six advanced in-memory-injecting malware and it flagged the attacks for all samples. We also analyzed FAROS' false positive rate with 90 non-injecting malware samples and 14 benign software from various categories. FAROS presented a very low false positive rate of 2%, which shows its potential towards practical solutions against advanced in-memory-only anti-reverse-engineering attacks.

Spear phishing emails are key in many cyber attacks. Successful emails employ psychological weapons of influence and relevant life domains. This paper investigates spear phishing susceptibility as a function of Internet user age (old vs young), weapon of influence, and life domain. A 21-day study was conducted with 158 participants (younger and older Internet users). Data collection took place at the participants' homes to increase ecological validity. Our results show that older women were the most vulnerable group to phishing attacks. While younger adults were most susceptible to scarcity, older adults were most susceptible to reciprocation. Further, there was a discrepancy, particularly among older users, between self-reported susceptibility awareness and their behavior during the intervention. Our results show the need for demographic personalization for warnings, training and educational tools in targeting the specifics of the older adult population.

Applications are generally written assuming a predictable and well-behaved OS. In practice, they experience unpredictable misbehavior at the OS level and across OSes: different OSes can handle network events differently, APIs can behave differently across OSes, and OSes may be compromised or buggy. This unpredictability is challenging because its sources typically manifest during deployment and are hard to reproduce. This paper introduces Bear, a framework for statistical analysis of application sensitivity to OS unpredictability that can help developers build more resilient software, discover challenging bugs and identify the scenarios that most need validation. Bear analyzes a program with a set of perturbation strategies on a set of commonly used system calls in order to discover the most sensitive system calls for each application, the most impactful strategies, and how they predict abnormal program outcome. We evaluated Bear with 113 CPU and IO-bound programs, and our results show that null memory dereferencing and erroneous buffer operations are the most impactful strategies for predicting abnormal program execution and that their impacts increase ten-fold with workload increase (e.g. number of network requests from 10 to 1000). Generic system calls are more sensitive than specialized system calls-for example, write and sendto can both be used to send data through a socket, but the sensitivity of write is twice that of sendto. System calls with an array parameter (e.g. read) are more sensitive to perturbations than those having a struct parameter with a buffer (e.g readv). Moreover, the fewer parameters a system call has, the more sensitive it is.

We propose a new security paradigm that makes cross-layer personalization a premier component in the design of security solutions for computer infrastructure and situational awareness. This paradigm is based on the observation that computer systems have a personalized usage profile that depends on the user and his activities. Further, it spans the various layers of abstraction that make up a computer system, as if the user embedded his own DNA into the computer system. To realize such a paradigm, we discuss the design of a comprehensive and cross-layer profiling approach, which can be adopted to boost the effectiveness of various security solutions, e.g., malware detection, insider attacker prevention and continuous authentication. The current state-of-the-art in computer infrastructure defense solutions focuses on one layer of operation with deployments coming in a "one size fits all" format, without taking into account the unique way people use their computers. The key novelty of our proposal is the cross-layer personalization, where we derive the distinguishable behaviors from the intelligence of three layers of abstraction. First, we combine intelligence from: a) the user layer, (e.g., mouse click patterns); b) the operating system layer; c) the network layer. Second, we develop cross-layer personalized profiles for system usage. We will limit our scope to companies and organizations, where computers are used in a more routine and one-on-one style, before we expand our research to personally owned computers. Our preliminary results show that just the time accesses in user web logs are already sufficient to distinguish users from each other,with users of the same demographics showing similarities in their profiles. Our goal is to challenge today's paradigm for anomaly detection that seems to follow a monoculture and treat each layer in isolation. We also discuss deployment, performance overhead, and privacy issues raised by our paradigm.

Psychology and neuroscience literature shows the existance of upper bounds on the human capacity for executing cognitive tasks and for information processing. These bounds are where, demonstrably, people start experiencing cognitive strain and consequently committing errors in the tasks execution. We argue that the usable security discipline should scientifically understand such bounds in order to have realistic expectations about what people can or cannot attain when coping with security tasks. This may shed light on whether Johnny will be ever be able to encrypt. We propose a conceptual framework for evaluation of human capacities in security that also assigns systems to complexity categories according to their security and usability. From what we have initiated in this paper, we ultimately aim at providing designers of security mechanisms and policies with the ability to say: "This feature of the security mechanism X or this security policy element Y is inappropriate, because this evidence shows that it is beyond the capacity of its target community".

Despite the security community's emphasis on the importance of building secure software, the number of new vulnerabilities found in our systems is increasing. In addition, vulnerabilities that have been studied for years are still commonly reported in vulnerability databases. This paper investigates a new hypothesis that software vulnerabilities are blind spots in developer's heuristic-based decision-making processes. Heuristics are simple computational models to solve problems without considering all the information available. They are an adaptive response to our short working memory because they require less cognitive effort. Our hypothesis is that as software vulnerabilities represent corner cases that exercise unusual information flows, they tend to be left out from the repertoire of heuristics used by developers during their programming tasks.

Library OSes are a promising approach for applications to efficiently obtain the benefits of virtual machines, including security isolation, host platform compatibility, and migration. Library OSes refactor a traditional OS kernel into an application library, avoiding overheads incurred by duplicate functionality. When compared to running a single application on an OS kernel in a VM, recent library OSes reduce the memory footprint by an order-of-magnitude.

The security community spares no effort in emphasizing security awareness and the importance of building secure software. However, the number of new vulnerabilities found in today's systems is still increasing. Furthermore, old and well-studied vulnerability types such as buffer overflows and SQL injections, are still repeatedly reported in vulnerability databases. Historically, the common response has been to blame the developers for their lack of security education. This paper discusses a new hypothesis to explain this problem by introducing a new security paradigm where software vulnerabilities are viewed as developers' blind spots in their decision making. We argue that such a flawed mental process is heuristic-based, where humans solve problems without considering all the information available, just like taking shortcuts. This paper's thesis is that security thinking tends to be left out by developers during their programming, as vulnerabilities usually exist in corner cases with unusual information flows. Leveraging this paradigm, this paper introduces a novel methodology for capturing and understanding security-related blind spots in Application Programming Interfaces (APIs). Finally, it discusses how this methodology can be applied to the design and implementation of the next generation of automated diagnosis tools.

We are always patching our systems against specific nstances of whatever the latest new, hot, trendy vulnerability type is. First it was time-of-check-to-time-of-use, then buffer overflows, then SQL injection, then cross-site scripting. Vulnerability studies are supposed to accomplish two main goals: to classify vulnerabilities into general classes so that unknown vulnerabilities of that class can be discovered in a proactive way, and to enable us to understand the fundamental nature of vulnerabilities so that when we build new systems we know how to make them secure.

The traditional virtual machine usage model advocates placing security mechanisms in a trusted VM layer and letting the untrusted guest OS run unaware of the presence of virtualization. In this work we challenge this traditional model and propose a collaboration approach between a virtualization-aware operating system and a VM layer to prevent tampering against kernel code and data. Our integrity model is a relaxed version of Biba's and the main idea is to have all attempted writes into kernel code and data segments checked for validity at VM level. The OS-VM collaboration bridges the semantic gap between tracing low integrity objects at OS-level (files, processes, modules, allocated areas) and architecture-level (memory and registers). We have implemented this approach in a proof-of-concept prototype and have successfully tested it against 6 rootkits (including a non-control data attack) and 4 real-world benign LKM/drivers. All rootkits were prevented from corrupting kernel space and no false positive was triggered for benign modules. Performance measurements show that the average overhead to the VM for the OS-VM communication is low (7%, CPU benchmarks). The greatest overhead is caused by the memory monitoring module inside the VM: 1.38X alone and 1.46X when combined with the OS-VM communication. For OS microbenchmarks the slowdown for the OS-VM communication was 1.16X on average.

System availability is difficult for systems to maintain in the face of Internet worms. Large systems have vulnerabilities, and if a system attempts to continue operation after an attack, it may not behave properly. Traditional mechanisms for detecting attacks disrupt service and current recovery approaches are application-based and cannot guarantee recovery in the face of exploits that corrupt the kernel, involve multiple processes or target multithreaded network services. This paper presents Bezoar, an automated full-system virtual machine-based approach to recover from zero-day control-flow hijacking attacks. Bezoar tracks down the source of network bytes in the system and after an attack, replays the checkpointed run while ignoring inputs from the malicious source. We evaluated our proof-of-concept prototype on six notorious exploits for Linux and Windows. In all cases, it recovered the full system state and resumed execution. Bezoar incurs low overhead to the virtual machine: less than 1% for the recovery and log components and approximately 1.4X for the memory monitor component that tracks down network bytes, for five SPEC INT 2000 benchmarks.

Worms, viruses, and other malware can be ticking bombs counting down to a specific time, when they might, for example, delete files or download new instructions from a public web server. We propose a novel virtual-machine-based analysis technique to automatically discover the timetable of a piece of malware, or when events will be triggered, so that other types of analysis can discern what those events are. This information can be invaluable for responding to rapid malware, and automating its discovery can provide more accurate information with less delay than careful human analysis.Developing an automated system that produces the timetable of a piece of malware is a challenging research problem. In this paper, we describe our implementation of a key component of such a system: the discovery of timers without making assumptions about the integrity of the infected system's kernel. Our technique runs a virtual machine at slightly different rates of perceived time (time as seen by the virtual machine), and identifies time counters by correlating memory write frequency to timer interrupt frequency.We also analyze real malware to assess the feasibility of using full-system, machine-level symbolic execution on these timers to discover predicates. Because of the intricacies of the Gregorian calendar (leap years, different number of days in each month, etc.) these predicates will not be direct expressions on the timer but instead an annotated trace; so we formalize the calculation of a timetable as a weakest precondition calculation. Our analysis of six real worms sheds light on two challenges for future work: 1) time-dependent malware behavior often does not follow a linear timetable; and 2) that an attacker with knowledge of the analysis technique can evade analysis. Our current results are promising in that with simple symbolic execution we are able to discover predicates on the day of the month for four real worms. Then through more traditional manual analysis we conclude that a more control-flow-sensitive symbolic execution implementation would discover all predicates for the malware we analyzed.