5 October 2018 in IEEE Transactions on Dependable and Secure Computing
Recently, many studies have demonstrated deep neural network (DNN) classifiers can be fooled by the adversarial example, which is crafted via introducing some perturbations into an original sample. Accordingly, some powerful defense techniques were proposed. However, existing defense techniques often require modifying the target model or depend on the prior knowledge of attacks. In this paper, we propose a straightforward method for detecting adversarial image examples, which can be directly deployed into unmodified off-the-shelf DNN models. We consider the perturbation to images as a kind of noise and introduce two classic image processing techniques, scalar quantization and smoothing spatial filter, to reduce its effect. The image entropy is employed as a metric to implement an adaptive noise reduction for different kinds of images. Consequently, the adversarial example can be effectively detected by comparing the classification results of a given sample and its denoised version, without referring to any prior knowledge of attacks. More than 20,000 adversarial examples against some state-of-the-art DNN models are used to evaluate the proposed method, which are crafted with different attack techniques. The experiments show that our detection method can achieve a high overall F1 score of 96.39% and certainly raises the bar for defense-aware attacks.
8 January 2021 in IEEE Transactions on Information Forensics and Security
Smart contract vulnerabilities have attracted lots of concerns due to the resultant financial losses. Matching-based detection methods extrapolating known vulnerabilities to unknown have proven to be effective in other platforms. However, directly adopting the technique to smart contracts is obstructed by two issues, i.e., diversity of bytecode generation resulting from the rapid evolution of compilers and interference of noise code easily caused by the homogeneous business logics. To address the problems, we propose contract bytecode-oriented normalization and slicing techniques to augment bytecode matching. Specifically, we conduct data- and instruction-level normalizations to uniform the bytecode generated by different compilers, and enforce contract-specific slicing by tracking data- and control-flows with simulated bytecode executions to prune the noise code as far as possible. Based on the above techniques, we design an unsupervised graph embedding algorithm to encode the code graphs into quantitatively comparable vectors. The potentially vulnerable smart contracts can be identified by measuring the similarities between their vectors and known vulnerable ones. Our evaluations have shown the efficiency (0.47 seconds per contract on average), effectiveness (160 verified true positives) and high precision (91.95% for top-ranked). It is worth noting that, we also identify dozens of honeypot contracts, further demonstrating the capability of our method.
15 August 2017 in IEEE Transactions on Dependable and Secure Computing
Dynamic taint analysis (DTA), as a mainstream information flow tracking technique, has been widely used in mobile security. On the Android platform, the existing DTA approaches are typically implemented by instrumenting the Dalvik virtual machine (DVM) interpreter or the Android emulator with taint enforcement code. The most prominent problem of the interpreter-based approaches is that they cannot work in the new Android RunTime (ART) environment introduced since the 5.0 release. For the emulator-based approaches, the most prominent problem is that they cannot be deployed on real devices. In addition, almost all the existing Android DTA approaches only concern the explicit information flow caused by data dependence, while completely ignore the impact of implicit information flow caused by control dependence. These problems limit their adoption in the latest Android system and make them ineffective in detecting the state-of-the-art malware whose privacy-breaching behaviors are inactivated in the analyzed environment (e.g., the emulator) or conducted via implicit information flow. In this paper, we present TaintMan, an ART-compatible DTA framework that can be deployed on unmodified and non-rooted Android devices. In TaintMan, the taint enforcement code is statically instrumented into both the target application and the system class libraries to track data flow and common control flow. A specially designed execution environment reconstruction technique, named reference hijacking, is proposed to force the target application to reference the instrumented system class libraries. By enforcing on-demand instrumentation and on-demand tracking, the performance overhead is significantly reduced. We have developed TaintMan and deployed it on two popular stock smartphones (HTC One S equipped with Android-4.0 and Motorola MOTO G equipped with Android-5.0). The evaluation with malware samples and real-world applications shows that TaintMan can effectively detect privacy leakage behaviors with an acceptable performance overhead.
19 October 2022 in IEEE Transactions on Software Engineering
Third-party library (TPL) detection is important for Android app security analysis nowadays. Unfortunately, the existing techniques often suffer from poor scalability. In some situations, the detection time cost is even unacceptable. Although a few existing methods run relatively fast, they cannot provide enough effectiveness, especially for non-structure-preserving obfuscated apps, e.g., repackaged and flattened. In this paper, we treat TPLs detection as a set inclusion problem to effectively and efficiently analyze obfuscated apps, and develop a scalable two-stage detection approach, Libloom . Specifically, the package and class signatures are encoded into two levels of Bloom filters respectively. At the first stage, the package filters are used to identify a limited number of candidate TPLs via set overlapping measurement to avoid unnecessary class-level set analysis. Subsequently, with the class filters, a similarity score is computed between the query app and each candidate to detect the integrated TPLs, and a novel entropy-based metric is presented to specially handle the repackaged and flattened apps. We have evaluated Libloom on some large-scale benchmarks involving tens of thousands of TPL instances. The experiment results demonstrate that Libloom outperforms state-of-the-art tools in both effectiveness and efficiency. Especially, the proposed two-stage method can run about ten times faster than the straightforward class-level analysis on flattened apps, and without loss of accuracy.
16 March 2018 in IEEE Transactions on Software Engineering
Code mining has been proven to be a promising approach to inferring implicit programming rules for finding software bugs. However, existing methods may report large numbers of false positives and false negatives. In this paper, we propose a novel approach called EAntMiner to improve the effectiveness of code mining. EAntMiner elaborately reduces noises from statements irrelevant to interesting rules and different implementation forms of the same logic. During preprocessing, we employ program slicing to decompose the original source repository into independent sub-repositories. In each sub-repository, statements irrelevant to critical operations (automatically extracted from source code) are excluded and various semantics-equivalent implementations are normalized into a canonical form as far as possible. Moreover, to tackle the challenge that some bugs are difficult to be detected by mining frequent patterns as rules, we further developed a kNN-based method to identify them. We have implemented EAntMiner and evaluated it on four large-scale C systems. EAntMiner successfully detected 105 previously unknown bugs that have been confirmed by corresponding development communities. A set of comparative evaluations also demonstrate that EAntMiner can effectively improve the precision of code mining.