In today's digital communication landscape, the security of End-to-End Encryption (E2EE) protocols is paramount, particularly in safeguarding user privacy and data integrity. Formal verification has emerged as a critical methodology to ensure these protocols' effectiveness and reliability. However, certain scenarios reveal failures in guaranteeing E2EE communication security properties, leading to potential privacy breaches. We conducted a comprehensive survey of mainstream messaging protocols, systematically categorizing their vulnerabilities and causes. By focusing on identity authentication during user registration, key agreement protocols in session establishment, and data encapsulation processes in message transmission, we developed a formal verification framework for E2EE Messaging Protocols. This framework integrates dual perspectives: cryptographic protocol verification and implementation auditing.

In-context learning (ICL) has become a powerful, data-efficient paradigm for text classification using large language models. However, its robustness against realistic adversarial threats remains largely unexplored. We introduce ICL-Evader, a novel black-box evasion attack framework that operates under a highly practical zero-query threat model, requiring no access to model parameters, gradients, or query-based feedback during attack generation. We design three novel attacks—Fake Claim, Template, and Needle-in-a-Haystack—that exploit inherent limitations of LLMs in processing in-context prompts. Evaluated across sentiment analysis, toxicity, and illicit promotion tasks, our attacks significantly degrade classifier performance (e.g., achieving up to 95.3% attack success rate), drastically outperforming traditional NLP attacks which prove ineffective under the same constraints. To counter these vulnerabilities, we systematically investigate defense strategies and identify a joint defense recipe that effectively mitigates all attacks with minimal utility loss (<5% accuracy degradation). Finally, we translate our defensive insights into an automated tool that proactively fortifies standard ICL prompts against adversarial evasion. This work provides a comprehensive security assessment of ICL, revealing critical vulnerabilities and offering practical solutions for building more robust systems. Our source code and evaluation datasets are publicly available at: https://github.com/ChaseSecurity/ICL-Evader ICL-Evader Repository.

Abstract Spatial skyline queries represent a specialized category of skyline queries, applicable in various domains such as facility location, crisis management, and travel or event planning. The emergence of secure spatial skyline queries carries substantial practical implications. In this paper, we address the challenge posed by the point-geometry dependency problem inherent in existing spatial skyline query algorithms. Our approach involves a transformative strategy that simplifies the query into a more tractable range query problem. Building on this transformation approach, we design an efficient and secure spatial skyline query method for encrypted data, which requires alternating between ciphertext and plaintext queries. To ensure both security and optimal performance, we execute plaintext queries within a trusted execution environment. Experiments demonstrate the efficiency and effectiveness of our approach.

While Retrieval-Augmented Generation (RAG) effectively reduces hallucinations by integrating external knowledge bases, it introduces vulnerabilities to membership inference attacks (MIAs), particularly in systems handling sensitive data. Existing MIAs targeting RAG's external databases often rely on model responses but ignore the interference of non-member-retrieved documents on RAG outputs, limiting their effectiveness. To address this, we propose DCMI, a differential calibration MIA that mitigates the negative impact of non-member-retrieved documents. Specifically, DCMI leverages the sensitivity gap between member and non-member retrieved documents under query perturbation. It generates perturbed queries for calibration to isolate the contribution of member-retrieved documents while minimizing the interference from non-member-retrieved documents. Experiments under progressively relaxed assumptions show that DCMI consistently outperforms baselines—for example, achieving 97.42% AUC and 94.35% Accuracy against the RAG system with Flan-T5, exceeding the MBA baseline by over 40%. Furthermore, on real-world RAG platforms such as Dify and MaxKB, DCMI maintains a 10%-20% advantage over the baseline. These results highlight significant privacy risks in RAG systems and emphasize the need for stronger protection mechanisms. We appeal to the community's consideration of deeper investigations, like ours, against the data leakage risks in rapidly evolving RAG systems.

Local differential privacy (LDP) involves users perturbing their inputs to provide plausible deniability of their data. However, this also makes LDP vulnerable to poisoning attacks. In this paper, we first introduce novel poisoning attacks for ranking estimation. These attacks are intricate, as fake attackers do not merely adjust the frequency of target items. Instead, they leverage a limited number of fake users to precisely modify frequencies, effectively altering item rankings to maximize gains. To tackle this challenge, we introduce the concepts of attack cost and optimal attack item (set), and propose corresponding strategies for kRR, OUE, and OLH protocols. For kRR, we iteratively select optimal attack items and allocate suitable fake users. For OUE, we iteratively determine optimal attack item sets and consider the incremental changes in item frequencies across different sets. Regarding OLH, we develop a harmonic cost function based on the pre-image of a hash to select that supporting a larger number of effective attack items. Lastly, we present an attack strategy based on confidence levels to quantify the probability of a successful attack and the number of attack iterations more precisely. We demonstrate the effectiveness of our attacks through theoretical and empirical evidence, highlighting the necessity for defenses against these attacks. The source code and data have been made available at https://github.com/LDP-user/LDP-Ranking.git.

The susceptibility of deep neural networks to adversarial attacks is a well-established concern. To address this problem, robustness certification is proposed, which, unfortunately, suffers from precision or scalability issues. In this paper, we present DROWN (Dual CROWN), a novel method for certifying the robustness of DNNs. The advantage of DROWN is that it tightens classic LiRPA-based methods yet maintains similar scalability, which comes from refining pre-activation bounds of ReLU relaxations using two pairs of linear bounds derived from different relaxations of ReLU units in previous layers. The extensive evaluations show that DROWN achieves up to 83.39% higher certified robust accuracy than the baseline on CNNs and up to 4.68 times larger certified radii than the baseline on Transformers. Meanwhile, the running time of DROWN is about twice that of the baseline.

Privacy-preserving federated learning (PPFL) securely aggregates local gradients of clients into a global model, but it is vulnerable to poisoning attacks by Byzantine clients who upload malicious gradients to prevent convergence and compromise accuracy. Traditional defense methods, relying on observable characteristics in plaintext, are incompatible with PPFL, where client gradients are encrypted and indistinguishable to the central server. To detect malicious clients who upload poisoning updates in ciphertext, we design a privacy-preserving federated learning framework, namely Multi-GAFL. The framework initially employs a verifiable secret sharing (VSS) scheme combined with a masking technique to protect the privacy of uploading gradients. Subsequently, to accurately identify Byzantine clients and to remove the gradients of the detected clients, we propose a multi-group aggregation strategy and norm & sign detection algorithm. These approaches ultimately mitigate the impact of poisoning attacks. In addition, by setting a threshold in VSS, our Multi-GAFL achieves robustness against client dropout during training. Experiments demonstrate the effectiveness of Multi-GAFL in resisting poisoning attacks. Even when the proportion of Byzantine clients reaches 40%, MultiGAFL still effectively defends against multiple poisoning attacks and outperforms existing defense methods.

The open-source release of large language models (LLMs) enables malicious users to create unauthorized derivative models at low cost, posing significant threats to intellectual property (IP) and market stability. Existing IP protection methods either require access to model parameters or are vulnerable to fine-tuning attacks. To fill this gap, we propose ErrorTrace, a robust and black-box traceability mechanism for protecting LLM IP. Specifically, ErrorTrace leverages the unique error patterns of model families by mapping and analyzing their distinct error spaces, enabling robust and efficient IP protection without relying on internal parameters or specific query responses. Experimental results show that ErrorTrace achieves a traceability accuracy of 0.8518 for 27 base models when the suspect model is not included in ErrorTrace's training set, outperforming the baseline by 0.2593. Additionally,ErrorTrace successfully tracks 34 fine-tuned, pruned and merged models across various scenarios, demonstrating its broad applicability and robustness. In addition, ErrorTrace shows a certain level of resilience when subjected to adversarial attacks. Our code is available at: https://github.com/csdatazcc/ErrorTrace.

Text-to-image (T2I) generative models have revolutionized content creation by transforming textual descriptions into high-quality images. However, these models are vulnerable to jailbreaking attacks, where carefully crafted prompts bypass safety mechanisms to produce unsafe content. While researchers have developed various jailbreak attacks to expose this risk, these methods face significant limitations, including impractical access requirements, easily detectable unnatural prompts, restricted search spaces, and high query demands on the target system. In this paper, we propose JailFuzzer, a novel fuzzing framework driven by large language model (LLM) agents, designed to efficiently generate natural and semantically meaningful jailbreak prompts in a black-box setting. Specifically, JailFuzzer employs fuzz-testing principles with three components: a seed pool for initial and jailbreak prompts, a guided mutation engine for generating meaningful variations, and an oracle function to evaluate jailbreak success. Furthermore, we construct the guided mutation engine and oracle function by LLM-based agents, which further ensures efficiency and adaptability in black-box settings. Extensive experiments demonstrate that JailFuzzer has significant advantages in jailbreaking T2I models. It generates natural and semantically coherent prompts, reducing the likelihood of detection by traditional defenses. Additionally, it achieves a high success rate in jailbreak attacks with minimal query overhead, outperforming existing methods across all key metrics. This study underscores the need for stronger safety mechanisms in generative models and provides a foundation for future research on defending against sophisticated jailbreaking attacks. JailFuzzer is open-source and available at this repository: https://github.com/YingkaiD/JailFuzzer.

Abstract Federated learning (FL) often uses local differential privacy (LDP) to prevent leaking data privacy through gradients. However, due to the high dimension of gradients, LDP will encounter the problem of privacy budget explosion in the application, resulting in low accuracy of the training model. To overcome this shortcoming, we propose a differential privacy FL protocol incorporating a control matrix and double shuffles. The control matrix, generated by the analyzer, is responsible for governing the selection and upload of clients’ gradients. Double shufflers shuffle the control matrix and clients’ gradients, respectively, so that the control matrix is invisible to the server and the gradient is anonymous to the server. In addition, the existing differential private FL often uses the same clipping scale for gradients clipping to facilitate determining the noise scale. However, this will bring too many clipping errors for the large gradients and too many noise errors for the small ones. To solve these problems, we propose an adaptive clipping scheme. Experiments on the real-world datasets show that our proposed methods achieve higher testing accuracy.

Abstract Repairing software defects is crucial to improving the security and robustness of software. Traditional methods repair software defects by using the “stop-repair-restart” approach. Unfortunately, in some scenarios, such as cloud environments, restarting critical software is expensive. Dynamic methods enable the defect repair while the software is running, which can avoid software restart. However, dynamic repair requires capture the safe state of the running software (process). Otherwise, it will cause execution exceptions or even process crashes. To address the complex state issues in multi-threaded environments, existing methods modify the kernel or compiler, or even pre-adding custom code to the target software, which reduces their generality. To solve this problem, we propose a hot repair method HotFix, which can fix the defects without any software suspending. HotFix places probes in the process to receive state signals, which can avoid complex and time-consuming state identification. Then, it selects the safe zone and repairs the defects when the code in the safe zone is called, which can prevent the target process from hanging for a long time. Finally, HotFix completes multi-threaded automated migration online. Experiments and analysis show that HotFix can achieve hot repair in complex environments. We found that the affected function was executed no more than 1,000 times during the fix. Introducing 3us and 11us delays per request respectively when repairing Redis and Mem cached, and the requests influenced are limited.

Fuzzing is a fundamental technique for detecting vulnerabilities in network protocols. However, existing approaches suffer from low fuzzing throughput caused by the overhead associated with server under test (SUT) restarts and connection setup. In this article, we present HSPFuzzer, a High-Speed Protocol Fuzzer that leverages connection reuse to reduce SUT restarts and connection re-establishments. To enable efficient connection reuse, it incorporates a prefix message identification algorithm to determine the essential packets required within a connection and a coverage monitoring mechanism to detect abnormal execution states. Additionally, HSPFuzzer employs an innovative message provision method that ensures input messages are delivered to the SUT with minimal delay within the same connection. HSPFuzzer also eliminates the need for manually implementing message-splitting logic by connection reuse. We evaluate HSPFuzzer on 12 widely used servers and experimental results show that HSPFuzzer achieves fuzzing throughput$1062{\times }$faster than AFLNet, whereas other state-of-the-art fuzzers, including AFLNet, SnapFuzz, HNPFuzzer, and AFL++, achieve, at most, a$12{\times }$speedup over AFLNet. Furthermore, HSPFuzzer attains an average code coverage increase of 25.1% compared to AFLNet, while competing fuzzers achieve, at most, 2.13% more coverage. Notably, HSPFuzzer also discovers more vulnerabilities, which further proves its effectiveness.

The prediction module, powered by deep learning models, constitutes a fundamental component of high-level Autonomous Vehicles (AVs). Given the direct influence of the module’s prediction accuracy on AV driving behavior, ensuring its security is paramount. However, limited studies have explored the adversarial robustness of the prediction modules. Furthermore, existing methods still generate adversarial trajectories that deviate significantly from human driving behavior. These deviations can be easily identified as hazardous by AVs’ anomaly detection models and thus cannot effectively evaluate and reflect the robustness of the prediction modules. To bridge this gap, we propose a stealthy and more effective optimization-based attack method. Specifically, we reformulate the optimization problem using Lagrangian relaxation and design a Frenet-based objective function along with a distinct constraint space. We conduct extensive evaluations on 2 popular prediction models and 2 benchmark datasets. Our results show that our attack is highly effective, with over 87% attack success rates, outperforming all baseline attacks. Moreover, our attack method significantly improves the stealthiness of adversarial trajectories while guaranteeing adherence to physical constraints. Our attack is also found robust to noise from upstream modules, transferable across trajectory prediction models, and high realizability. Lastly, to verify its effectiveness in real-world applications, we conduct further simulation evaluations using a production-grade simulator. These simulations reveal that the adversarial trajectory we created could convincingly induce autonomous vehicles (AVs) to initiate hard braking.

Transformers based on attention mechanisms exhibit vulnerability to adversarial examples, posing a substantial threat to the security of their applications. Aiming to solve this problem, the concept of robustness certification is introduced to formally ascertain the presence of any adversarial example within a specified region surrounding a given sample. However, prior works have neglected the dependencies among inputs of softmax (the most complex function in attention mechanisms) during linear relaxations. This oversight has consequently led to imprecise certification results. In this work, we introduce GaLileo, a general linear relaxation framework designed to certify the robustness of Transformers. GaLileo effectively surmounts the trade-off between precision and efficiency in robustness certification through our innovative n-dimensional relaxation approach. Notably, our relaxation technique represents a pioneering effort as the first linear relaxation for n-dimensional functions such as softmax. Our novel approach successfully transcends the challenges posed by the curse of dimensionality inherent in linear relaxations, thereby enhancing linear bounds by incorporating input dependencies. Our evaluations encompassed a thorough analysis utilizing the SST and Yelp datasets along with diverse Transformers of different depths and widths. The experimental results demonstrate that, as compared to the baseline method CROWN-BaF, GaLileo achieves up to 3.24 times larger certified radii while requiring similar running times. Additionally, GaLileo successfully attains certification for Transformers' robustness against multi-word lp perturbations, marking a notable accomplishment in this field.

Emerging in recent years, open edge computing platforms (OECPs) claim large-scale edge nodes, the extensive usage and adoption, as well as the openness to any third parties to join as edge nodes. For instance, OneThingCloud, a major OECP operated in China, advertises 5 million edge nodes, 70TB bandwidth, and 1,500PB storage. However, little information is publicly available for such OECPs with regards to their technical mechanisms and involvement in edge computing activities. Furthermore, different from known edge computing paradigms, OECPs feature an open ecosystem wherein any third party can participate as edge nodes and earn revenue for the contribution of computing and bandwidth resources, which, however, can introduce byzantine or even malicious edge nodes and thus break the traditional threat model for edge computing. In this study, we conduct the first empirical study on two representative OECPs, which is made possible through the deployment of edge nodes across locations, the efficient and semi-automatic analysis of edge traffic as well as the carefully designed security experiments. As the results, a set of novel findings and insights have been distilled with regards to their technical mechanisms, the landscape of edge nodes, the usage and adoption, and the practical security/privacy risks. Particularly, millions of daily active edge nodes have been observed, which feature a wide distribution in the network space and the extensive adoption in content delivery towards end users of 16 popular Internet services. Also, multiple practical and concerning security risks have been identified along with acknowledgements received from relevant parties, e.g., the exposure of long-term and crossedge-node credentials, the co-location with malicious activities of diverse categories, the failures of TLS certificate verification, the extensive information leakage against end users, etc.

Mini-programs are lightweight apps running in super apps (such as WeChat, Baidu, Alipay, and TikTok), an emerging paradigm in the era of mobile computing. With the growing popularity of mini-programs, there is an increasing concern for their security and privacy. In essence, mini-programs are WebView-based apps. This means that they may be vulnerable to the same security risks associated with web apps. In this work, we discovered a new mini-program vulnerability called MiniCPRF (Cross-Page Request Forgery in Mini-Programs). The exploit of this vulnerability is easy, and the attack consequences are severe, leading to unauthorized operations, such as free shopping, and the exposure of confidential information, such as credit card numbers. The root causes of MiniCPRF can be attributed to multiple design flaws in both mini-programs and their super apps, including the insecure routing mechanism, lack of message integrity check, and plain-text storage. To evaluate the impacts of MiniCPRF, we designed an automated analysis framework called MiniCAT. It can automatically crawl mini-programs, perform static analysis on them, and generate detection reports. In large-scale real-world evaluations with MiniCAT, we identified that 32.0% (13,349/41,726) of analyzable mini-programs are potentially vulnerable to MiniCPRF, including some famous ones with millions of users, such as Sohu and Wenjuanxing. Following the responsible disclosure principle, we have reported verified vulnerable mini-programs to the corresponding vendors and developers, and three real-world cases have been confirmed by CNVD. Additionally, we suggest mitigation strategies to resolve the security issue related to MiniCPRF.

Generative Adversarial Networks (GANs) have become widely used in model training, as they can improve performance and/or protect sensitive information by generating data. However, this also raises potential risks, as malicious GANs may compromise or sabotage models by poisoning their training data. Therefore, it is important to verify the origin of a model’s training data for accountability purposes. In this work, we take the first step in the forensic analysis of models trained on GAN-generated data. Specifically, we first detect whether a model is trained on GAN-generated or real data. We then attribute these models, trained on GAN-generated data, to their respective source GANs. We conduct extensive experiments on three datasets, using four popular GAN architectures and four common model architectures. Empirical results show the remarkable performance of our detection and attribution methods. Furthermore, we conduct a more in-depth study and reveal that models trained on various data sources exhibit different decision boundaries and behaviours.

Mobile apps have become popular for providing artificial intelligence (AI) services via on-device machine learning (ML) techniques. Unlike accomplishing these AI services on remote servers traditionally, these on-device techniques process sensitive information required by AI services locally, which can mitigate the severe concerns of the sensitive data collection on the remote side. However, these on-device techniques have to push the core of ML expertise (e.g., models) to smartphones locally, which are still subject to similar vulnerabilities on the remote clouds and servers, especially when facing the model stealing attack. To defend against these attacks, developers have taken various protective measures. Unfortunately, we have found that these protections are still insufficient, and on-device ML models in mobile apps could be extracted and reused without limitation. To better demonstrate its inadequate protection and the feasibility of this attack, this paper presents DeMistify, which statically locates ML models within an app, slices relevant execution components, and finally generates scripts automatically to instrument mobile apps to successfully steal and reuse target ML models freely. To evaluate DeMistify and demonstrate its applicability, we apply it on 1,511 top mobile apps using on-device ML expertise for several ML services based on their install numbers from Google Play and DeMistify can successfully execute 1250 of them (82.73%). In addition, an in-depth study is conducted to understand the on-device ML ecosystem in the mobile application.

Deep learning models, pivotal in AI applications, are susceptible to model hijacking attacks. In model hijacking attacks, adversaries can misuse models for unintended tasks, shifting blame and maintenance costs onto the models’ deployers. Existing attack methods re-purpose target models by poisoning their training sets during training. However, leading models like GPT-4 and BERT with vast parameters are often pruned before deployment on resource-limited devices, which presents challenges for in-training attacks, including existing model hi- jacking attacks. In this paper, we propose PRJack, the first pruning-resistant hijacking attack. Specifically, the adversary re-purposes a model to perform a hijacking task different from the original task, which can still be activated even after model pruning. Our experiments across multiple datasets and pruning techniques highlight PRJack’s remarkable superiority on pruned models over existing model hijacking attacks.

Cryptographic techniques form the foundation of the security and privacy of computing solutions. However, if cryptographic APIs are not invoked correctly, they can result in significant security problems. In this paper, we abstract the intricate crypto misuse detection problem as a data-flow reasoning task. Towards this end, we propose CrypTody, a novel logic-inference-based framework for detecting crypto misuses via reasoning about data flows on multi-architecture IoT firmware images. It carries out cross-architecture analysis, with detection strategies to reduce false positives and false negatives, such as cross-flow misuse inference. To evaluate the effectiveness of CrypTody, we conducted a large-scale experiment on 1,431 firmware images from 16 vendors. Our evaluation shows that 46% of the firmware images have high-risk misuses and 95% have at least one cryptographic misuse. In total, we find 6,624 potential crypto misuses, with 760 being cross-flow misuses that are not detected by existing solutions. We have responsibly disclosed portions of our findings to the relevant vendors. From the feedback, we note that CrypTody has a low false-positive rate for the confirmed misuses. Some typical cases have been assigned CVEs and fixed by the vendors.

While DeepFake applications are becoming popular in recent years, their abuses pose a serious privacy threat. Unfortunately, most related detection algorithms to mitigate the abuse issues are inherently vulnerable to adversarial attacks because they are built atop DNN-based classification models, and the literature has demonstrated that they could be bypassed by introducing pixel-level perturbations. Though corresponding mitigation has been proposed, we have identified a new attribute-variation-based adversarial attack (AVA) that perturbs the latent space via a combination of Gaussian prior and semantic discriminator to bypass such mitigation. It perturbs the semantics in the attribute space of DeepFake images, which are inconspicuous to human beings (e.g., mouth open) but can result in substantial differences in DeepFake detection. We evaluate our proposed AVA attack on nine state-of-the-art DeepFake detection algorithms and applications. The empirical results demonstrate that AVA attack defeats the state-of-the-art black box attacks against DeepFake detectors and achieves more than a 95% success rate on two commercial DeepFake detectors. Moreover, our human study indicates that AVA-generated DeepFake images are often imperceptible to humans, which presents huge security and privacy concerns.

This work addresses the problem of collecting multidimensional data while adhering to personalized local differential privacy. In this local context, each user possesses a data record containing multiple attributes. Due to variations in attribute sensitivity, privacy requirements of the attributes differ. Balancing personalized privacy needs of attributes with maximizing the accuracy of statistical results presents a significant challenge. The accuracy of estimation results is closely tied to the allocation of privacy budgets and user grouping. Consequently, we define an optimization problem: determining the allocation of privacy budgets to each attribute within the group and identifying the corresponding number of users in the group to ensure optimal estimation result accuracy while meeting privacy constraints. To simplify the complexity of this optimization problem, we introduce a progressive optimization method. This method initially groups attributes and subsequently fine-tunes and optimizes them. Practically, our approach meets the personalized privacy protection requirements of attributes. Experimental results indicate that our approach achieves significant improvements in data utility.

Logic vulnerabilities associated with TLS protocol implementations often do not exhibit explicit erroneous behaviors, making them difficult to detect by testers. However, these vulnerabilities can pose serious security threats. While testing for TLS protocols lacks uniform test oracles, differential fuzzing effectively addresses this issue. Unfortunately, most of these vulnerabilities are triggered in deep protocol states, and no existing work on differential fuzzing targeting these states exists. In this paper, we propose a deep differential fuzzing framework that focuses on detecting logic issues in deep TLS protocol states. Our approach is based on the message tuples we proposed, which are semi-automatically extracted from RFCs using NLP techniques. We address the problem of test interruptions during early handshakes caused by original data inconsistencies by redefining the consistency determination to achieve deep differential fuzzing. In addition, we use encoding classification statistics to achieve quick and efficient analysis of the massive test results. Based on our approach, we implemented TLS-DeepDiffer and used it to test nine kinds of popular TLS libraries. We found four historical CVEs, one newly discovered high-risk vulnerability, and 24 security or implementation issues, demonstrating the usefulness of our approach.

Private browsing is a common feature of web browsers on desktop platforms. This feature protects the privacy of users browsing the Internet and, therefore, is widely welcomed by users. In recent years, with the popularity of smartphones, the private browsing mode has been introduced into mobile browsers. However, its deployment on mobile platforms has not been well evaluated. To bridge the gap, in this work, we systemically studied the private browsing modes of Android browser apps. Specifically, we proposed six private rules for mobile browsers to follow by combining the mobile browsing features with the previous research on private browsing. Furthermore, we designed an automated analysis framework, BroDroid, to detect whether mobile browsers violate these rules. Also, with BroDroid, we evaluated 49 popular browser apps crawled from Google Play. Finally, BroDroid successfully identified 58 violations, some of which come from the promised capabilities of the browser. We reported our discovered issues to the corresponding developers, and four of them (Yandex Browser, Mint Browser, Web Explorer, and Net Fast Web Browser) have acknowledged our findings. Our observation may be the tip of the iceberg, and more efforts should be put into improving the privacy protections of mobile browsers.

The public key encryption with keyword search (PEKS) schemes are mostly applied to small data sets in mail forwarding systems. When retrieving large databases, the typical search mechanism makes them inefficient and impractical. When designing a PEKS scheme, except for remedying the vulnerability of keyword guessing attacks (KGAs), other leakage issues, such as multipattern privacy and forward/backward security are rarely considered, which may lead to information leakage. Moreover, most existing PEKS only consider applications in single-user scenarios, and cannot be directly transferred to multiuser scenarios, which undermines the value of data utilization. To cope with the above concerns, we propose a PEKS scheme based on an inverted index where the bitmap is used to build the index for the first time in PEKS to meet some seemingly conflicting yet desirable characteristics. First, it has high search efficiency under multiwriter and multiuser. Through linear transformation, users quickly retrieve data and control other users’ access to their data without relying on a third party for authentication. Second, we prove its security in an enhanced security model that achieves multipattern privacy and forward and backward security. It can also resist KGA attacks without a designated tester, which makes it more practical. Finally, it can be extended to achieve search result verification. Compare to the scheme (Zhang et al. ICWS 2016), it has absolute advantages in security and computational cost where the search efficiency is improved by two orders of magnitude.

IoT devices have brought invaluable convenience to our daily life. However, their pervasiveness also amplifies the impact of security vulnerabilities. Many widespread vulnerabilities of embedded systems reside in their vulnerable border services. Unfortunately, existing vulnerability detection methods can neither effectively nor efficiently analyze such border services: they either introduce heavy execution overheads or have many false positives and negatives. In this paper, we propose a novel static taint checking solution, SaTC, to effectively detect security vulnerabilities in border services provided by embedded devices. Our key insight is that string literals on border interfaces are commonly shared between front-end files and back-end binaries to encode user input. Thus, we extract common keywords from the front-end and use them to locate reference points in the back-end, which indicate the input entry. Then, we apply targeted data-flow analysis to detect dangerous uses of the untrusted user input accurately. We implemented a prototype of SaTC and evaluated it on 39 firmware samples from six popular vendors. SaTC discovered 36 unknown bugs, of which CVE/CNVD/PSV confirms 33. Compared to the state-of-the-art tool KARONTE, SaTC found significantly more bugs in the test set. It shows that SaTC is effective in discovering bugs in embedded systems.

Deepfake data contains realistically manipulated faces—its abuses pose a huge threat to the security and privacy-critical applications. Intensive research from academia and industry has produced many deepfake/detection models, leading to a constant race of attack and defense. However, due to the lack of a unified evaluation platform, many critical questions on this subject remain largely unexplored. How is the anti-detection ability of the existing deepfake models? How generalizable are existing detection models against different deepfake samples? How effective are the detection APIs provided by the cloud-based vendors? How evasive and transferable are adversarial deepfakes in the lab and real-world environment? How do various factors impact the performance of deepfake and detection models? To bridge the gap, we design and implement DEEPFAKER 1 a unified and comprehensive deepfake detection evaluation platform. Specifically, DEEPFAKER has integrated 10 state-of-the-art deepfake methods and 9 representative detection methods, while providing a user-friendly interface and modular design that allows for easy integration of new methods. Leveraging DEEPFAKER , we conduct a large-scale empirical study of facial deepfake/detection models and draw a set of key findings: (i) the detection methods have poor generalization on samples generated by different deepfake methods; (ii) there is no significant correlation between anti-detection ability and visual quality of deepfake samples; (iii) the current detection APIs have poor detection performance and adversarial deepfakes can achieve about 70% attack success rate on all cloud-based vendors, calling for an urgent need to deploy effective and robust detection APIs; (iv) the detection methods in the lab are more robust against transfer attacks than the detection APIs in the real-world environment; and (v) deepfake videos may not always be more difficult to detect after video compression. We envision that DEEPFAKER will benefit future research on facial deepfake and detection.

Fully homomorphic encryption (FHE) is a promising data privacy solution for machine learning, which allows the inference to be performed with encrypted data. However, it typically leads to 5-6 orders of magnitude higher computation and storage overhead. This paper proposes the first full-fledged FPGA acceleration framework for FHE-based convolution neural network (HE-CNN) inference. We then design parameterized HE operation modules with intra- and inter- HE-CNN layer resource management based on FPGA high-level synthesis (HLS) design flow. With sophisticated resource and performance modeling of the HE operation modules, the proposed FxHENN framework automatically performs design space exploration to determine the optimized resource provisioning and generates the accelerator circuit for a given HE-CNN model on a target FPGA device. Compared with the state-of-the-art CPU-based HE-CNN inference solution, FxHENN achieves up to 13.49X speedup of inference latency, and 1187.12X energy efficiency. Meanwhile, given this is the first attempt in the literature on FPGA acceleration of fullfledged non-interactive HE-CNN inference, our results obtained on low-power FPGA devices demonstrate HE-CNN inference for edge and embedded computing is practical.

Popular Secure Instant Messaging (SIM) applications like Telegram and WhatsApp have deployed state-of-the-art encryption schemes in recent years to protect the security of user communications. However, SIM applications are still not completely secure. Governments can surveil and censor users who participate in groups on sensitive topics based on the leaked information of their SIM clients. In this paper, we find two types of padding flaws in SIM applications where the padding length is not long enough, thereby exposing users' encrypted traffic characteristics. Furthermore, we first present an Active Text Traffic Analysis Attack (ATTAA) that enables the adversary to obtain sensitive information about target users' clients by merely monitoring their encrypted SIM traffic. Specifically, the adversary can quickly identify the participants of target SIM groups with high accuracy. Our study demonstrates a significant, real-world threat to SIM users due to increasing government regulation on social media. We demonstrate the practicality of our ATTAA through extensive experiments on real-world SIM communications. Although SIM applications have various restrictions on message sending, our results show that only ten text messages in 10 seconds are enough to successfully attack Telegram and WhatsApp with an accuracy of 99.94% and 98.66%, and a false positive rate of$4.3\times 10^{-3}$and$1.5\times 10^{-4}$.

With the emergence of Web3, decentralized network protocol technologies have been vigorously developed. As a pioneer for decentralized real-time communication systems, Matrix is an open standard based on a federation specification protocol. Anyone can set up a self-hosted homeserver to participate in the global Matrix network and communicate with others in chat rooms. In this paper, we conduct the first in-depth measurement and exploratory research on Matrix’s ecosystem and security. We designed and implemented several investigation techniques to empirically delve into Matrix federation from various aspects (homeservers, rooms, and users). In the end, we identified a number of interesting findings and potential vulnerabilities, including anti-decentralization phenomena, cybersecurity threats in homeservers, and the confidentiality of encrypted rooms being compromised.

In the age of pervasive computing, traditionally centralized cloud storage (CCS) services may not fit in well due to their centralized architecture, limited worldwide availability, high expense, and security and privacy concerns. To address these issues, decentralized cloud storage (DCS) services emerged recently. However, previous works focus on analyzing the technical design of DCS services, e.g., their incentive mechanisms. Little is known regarding how well these DCS services work in real-world operations. In this paper, we fill this gap by providing the first empirical measurement of Storj, one of the most extensive in-operation decentralized cloud storage services, focusing on its ecosystem, performance, and security implications. Our study is made possible through multiple measurement techniques to automatically capture storage nodes, profile Storj's quality of service, understand co-located network threats, and evaluate potential attacks in a simulated environment. Leveraging these techniques, a set of insightful findings have been distilled. Particularly, we have observed over 32K storage nodes as well as 155K unique node IP addresses, which are widely distributed in 122 countries, 2,418 ASNs, and 205 /8 IPv4 prefixes. Regarding performance, storage customers located in Europe or the United States tend to enjoy a better storage performance than those in Asia-Pacific, likely due to the imbalanced distribution of storage nodes in different regions. Lastly, what is concerning is that 4.48% of IPs of storage nodes were found to have been associated with various malicious activities, especially botnets and cryptomining. Another vulnerability is that a malicious storage node could exploit multiple channels to boost its storage reputation while demoting that of benign nodes.

Recently, Deep learning (DL) compilers have been widely developed to optimize the deployment of DL models. These DL compilers transform DL models into high-level intermediate representation (IR) and then into low-level IR, ultimately generating optimized codes for different hardware targets. However, DL compilers are not immune to generating incorrect code, leading to potentially severe consequences. Testing techniques for low-level IR are limited, and efficient approaches for detecting some categories of non-crashing bugs are lacking. In this paper, we address the limitations of existing low-level IR DL compiler testing techniques and introduce DeepDiffer, a priority-guided differential testing framework designed to detect bugs resulting from low-level optimizations in the DL compiler, specifically TVM. We propose a novel DL compiler coverage metric and establish an optimization goal to maximize the detection of valuable differences between DL compilers. Our experiments demonstrate that DeepDiffer outperforms existing low-level IR fuzzers, detecting a wider range of bug types. In fact, DeepDiffer has successfully identified 13 bugs in TVM, which can be categorized into 9 distinct root causes, and 9 bugs are first found. We have submitted these bugs to the TVM community, where they have been confirmed.

The blockchain-powered decentralized applications and systems have been widely deployed in recent years. The decentralization feature promises users anonymity, security, and non-censorship, which is especially welcomed in the areas of decentralized finance and digital assets. From the perspective of most common users, a decentralized ecosystem means every service follows the principle of decentralization. However, we find that the services in a decentralized ecosystem still may contain centralized components or scenarios, like third-party SDKs and privileged operations, which violate the promise of decentralization and may cause a series of centralized security risks. In this work, we systematically study the centralized security risks existing in decentralized ecosystems. Specifically, we identify seven centralized security risks in the deployment of two typical decentralized services – crypto wallets and DApps, such as anonymity loss and overpowered owner. Also, to measure these risks in the wild, we designed an automated detection tool called Naga and carried out large-scale experiments. Based on the measurement of 28 Ethereum crypto wallets (Android version) and 110,506 on-chain smart contracts, the result shows that the centralized security risks are widespread. Up to 96.4% of wallets and 83.5% of contracts exist at least one security risk, including 260 well-known tokens with a total market cap of over $98 billion.

Given a set of local sequential datasets held by multiple parties, we study the problem of publishing a synthetic dataset that preserves approximate sequentiality information of the integrated dataset while satisfying differential privacy for each local dataset. The existing solutions for publishing differentially private sequential data in the centralized setting mostly adopt tree-based approaches. Such approaches rely on different tree structures that encode sequential data's statistical information. The construction of a tree structure is normally done by recursively splitting nodes whose noisyscores(e.g., entropy or count) are larger than a given threshold. However, extending similar ideas to the multi-party setting is challenging. First, the comparison between noisy scores and a given threshold needs to be done in a distributed manner without letting the parties know the noisy scores, while satisfying differential privacy for each local dataset. Second, in the multi-party setting the large number of node splitting decisions incurs prohibitive computation costs. In addressing the above challenges, we presentDPST, a distributed prediction suffix tree construction solution. In DPST, we first introduce a novel node splitting decision method that calculates the comparison result under encryption with substantially improved efficiency. Then we present a novel batch-based tree construction approach to reduce computation costs. In order to achieve high parallel performance without incurring any extra communication cost, we introduce theconjunctionandslidemethods to ensure that each batch contains a stable number of carefully arrangeddecision tasks. To further reduce communication and computation costs, we propose a prefix-based pre-pruning method to reduce the number of nodes that need to be judged whether to split by an interactive protocol. Extensive experiments on real datasets demonstrate that our DPST solution offers desirable data utility with low computation and communication costs.

Video captioning aims to automatically describe a video clip with informative sentences. At present, deep learning-based models have become the mainstream for this task and achieved competitive results on public datasets. Usually, these methods leverage different types of features to generate sentences, e.g., semantic information, 2D or 3D features. However, some methods only treat semantic information as a complement of visual representations and cannot fully exploit it; some of them ignore the relationship between different types of features. In addition, most of them select multiple frames of a video with an equally spaced sampling scheme, resulting in much redundant information. To address these issues, we present a novel video-captioning framework, Semantic Enhanced video captioning with Multi-feature Fusion, SEMF for short. It optimizes the use of different types of features from three aspects. First, a semantic encoder is designed to enhance meaningful semantic features through a semantic dictionary to boost performance. Second, a discrete selection module pays attention to important features and obtains different contexts at different steps to reduce feature redundancy. Finally, a multi-feature fusion module uses a novel relation-aware attention mechanism to separate the common and complementary components of different features to provide more effective visual features for the next step. Moreover, the entire framework can be trained in an end-to-end manner. Extensive experiments are conducted on Microsoft Research Video Description Corpus (MSVD) and MSR-Video to Text (MSR-VTT) datasets. The results demonstrate that SEMF is able to achieve state-of-the-art results.

Android is the most popular smartphone platform with over 85% market share. Its success is built on openness, and phone vendors can utilize the Android source code to make customized products with unique software/hardware features. On the other hand, the fragmentation and customization of Android also bring many security risks that have attracted the attention of researchers. Many efforts were put in to investigate the security of customized Android firmware. However, most of the previous works focus on designing efficient analysis tools or analyzing particular aspects of the firmware. There still lacks a panoramic view of Android firmware ecosystem security and the corresponding understandings based on large-scale firmware datasets. In this work, we made a large-scale comprehensive measurement of the Android firmware ecosystem security. Our study is based on 8,325 firmware images from 153 vendors and 813 Android-related CVEs, which is the largest Android firmware dataset ever used for security measurements. In particular, our study followed a series of research questions, covering vulnerabilities, patches, security updates, and pre-installed apps. To automate the analysis process, we designed a framework,AndScanner+, to complete firmware crawling, firmware parsing, patch analysis, and app analysis. Through massive data analysis and case explorations, several interesting findings are obtained. For example, the patch delay and missing issues are widespread in Android firmware images, say 31.4% and 5.6% of all images, respectively. The latest images of several phones still contain vulnerable pre-installed apps, and even the corresponding vulnerabilities have been publicly disclosed. In addition to data measurements, we also explore the causes behind these security threats through case studies and demonstrate that the discovered security threats can be converted into exploitable vulnerabilities. There are 46 new vulnerabilities found byAndScanner+, 36 of which have been assigned CVE/CNVD IDs. This study provides much new knowledge of the Android firmware ecosystem with a deep understanding of software engineering security practices.

Real-Time Operating System (RTOS) has become the main category of embedded systems. It is widely used to support tasks requiring real-time response such as printers and switches. The security of RTOS has been long overlooked as it was running in special environments isolated from attackers. However, with the rapid development of IoT devices, tremendous RTOS devices are connected to the public network. Due to the lack of security mechanisms, these devices are extremely vulnerable to a wide spectrum of attacks. Even worse, the monolithic design of RTOS combines various tasks and services into a single binary, which hinders the current program testing and analysis techniques working on RTOS. In this paper, we propose SFuzz, a novel slice-based fuzzer, to detect security vulnerabilities in RTOS. Our insight is that RTOS usually divides a complicated binary into many separated but single-minded tasks. Each task accomplishes a particular event in a deterministic way and its control flow is usually straightforward and independent. Therefore, we identify such code from the monolithic RTOS binary and synthesize a slice for effective testing. Specifically, SFuzz first identifies functions that handle user input, constructs call graphs that start from callers of these functions, and leverages forward slicing to build the execution tree based on the call graphs and pruning the paths independent of external inputs. Then, it detects and handles roadblocks within the coarse-grain scope that hinder effective fuzzing, such as instructions unrelated to the user input. And then, it conducts coverage-guided fuzzing on these code snippets. Finally, SFuzz leverages forward and backward slicing to track and verify each path constraint and determine whether a bug discovered in the fuzzer is a real vulnerability. SFuzz successfully discovered 77 zero-day bugs on 35 RTOS samples, and 67 of them have been assigned CVE or CNVD IDs. Our empirical evaluation shows that SFuzz outperforms the state-of-the-art tools (e.g., UnicornAFL) on testing RTOS.

We carry out the first in-depth characterization of residential proxies (RESIPs) in China, for which little is studied in previous works. Our study is made possible through a semantic-based classifier to automatically capture RESIP services. In addition to the classifier, new techniques have also been identified to capture RESIPs without interacting with and relaying traffic through RESIP services, which can significantly lower the cost and thus allow continuous monitoring of RESIPs. Our RESIP service classifier has achieved good performance with a recall of 99.7% and a precision of 97.6% in 10-fold cross validation. Applying the classifier has identified 399 RESIP services, a much larger set compared to 38 RESIP services collected in all previous works. Our effort of RESIP capturing leads to a collection of 9,077,278 RESIP IPs (51.36% are located in China), 96.70% of which are not covered in publicly available RESIP datasets. An extensive measurement on RESIPs and their services has uncovered a set of interesting findings as well as several security implications. Especially, 80.05% RESIP IPs located in China have sourced at least one malicious traffic flows during 2021, resulting in 52-million malicious traffic flows in total. And RESIPs have also been observed in corporation networks of 559 sensitive organizations including government agencies, education institutions and enterprises. Also, 3,232,698 China RESIP IPs have opened at least one TCP/UDP port for accepting relaying requests, which incurs non-negligible security risks to the local network of RESIPs. Besides, 91% China RESIP IPs are of a lifetime fewer than 10 days while most China RESIP services show up a crest-trough pattern in terms of the daily active RESIPs across time.

Data privacy becomes a crucial concern in the AI and big data era. Fully homomorphic encryption (FHE) is a promising data privacy protection technique where the entire computation is performed on encrypted data. However, the dramatic increase of the computation workload restrains the usage of FHE for the real-world applications. In this paper, we propose an FPFA accelerator design framework for CKKS-based HE. While the KeySwitch operations are the primary performance bottleneck of FHE computation, we propose a low latency design of KeySwitch module with reduced intra-operation data dependency. Compared with the state-of-the-art FPGA based key-switch implementation that is based on Verilog, the proposed high-level synthesis (HLS) based design reduces the operation latency by 40%. Furthermore, we propose an automated design space exploration framework which generates optimal encryption parameters and accelerators for a given application kernel and the target FPGA device. Experimental results for a set of real HE application kernels on different FPGA devices show that our HLS-based flexible design framework produces substantially better accelerator design compared with a fixed-parameter HE accelerator in terms of security, approximation error, and overall performance.

With the development of data outsourcing technology, the data stored by cloud storage servers are exploding. Secure deduplication for encrypted data helps cloud servers reduce storage overhead in the scenario that cloud users outsource their data in ciphertext. To satisfy client-side semantic security, most existing deduplication schemes for encrypted data need trusted third parties. However, trusted third parties are difficult to deploy and may cause potential risks. Therefore, we propose a secure cloud ciphertext deduplication scheme based on Intel SGX. The proposed scheme uses the Enclave security container provided by Intel SGX as the trusted execution environment on the cloud server to replace the trusted third party to perform sensitive operations. At the same time, our scheme simplifies the secure management of the file encryption keys so that the encryption key of the files with the same data can be securely distributed to other owners of the same file without the need for the original uploader online. We prove the security of the proposed scheme and the experiment shows the efficiency of the scheme.

Android is the most popular smartphone platform with over 85% market share. Its success is built on openness, and phone vendors can utilize the Android source code to make products with unique software/hardware features. On the other hand, the fragmentation and customization of Android also bring many security risks that have attracted the attention of researchers. Many efforts were put in to investigate the security of customized Android firmware. However, most of the previous work focuses on designing efficient analysis tools or analyzing particular aspects of the firmware. There still lacks a panoramic view of Android firmware ecosystem security and the corresponding understandings based on large-scale firmware datasets. In this work, we made a large-scale comprehensive measurement of the Android firmware ecosystem security. Our study is based on 6,261 firmware images from 153 vendors and 602 Android-related CVEs, which is the largest Android firmware dataset ever used for security measurements. In particular, our study followed a series of research questions, covering vulnerabilities, patches, security updates, and pre-installed apps. To automate the analysis process, we designed a framework, AndScanner, to complete ROM crawling, ROM parsing, patch analysis, and app analysis. Through massive data analysis and case explorations, several interesting findings are obtained. For example, the patch delay and missing issues are widespread in Android images, say 24.2% and 6.1% of all images, respectively. The latest images of several phones still contain vulnerable pre-installed apps, and even the corresponding vulnerabilities have been publicly disclosed. In addition to data measurements, we also explore the causes behind these security threats through case studies and demonstrate that the discovered security threats can be converted into exploitable vulnerabilities via 38 newfound vulnerabilities by our framework, 32 of which have been assigned CVE/CNVD numbers. This study provides much new knowledge of the Android firmware ecosystem with deep understanding of software engineering security practices.

During the Android app development, the SDK is essential, which provides rich APIs to facilitate the implementations of functionalities. However, in the Android framework, there still exist plenty of non-SDK APIs that are not well documented. These non-SDK APIs can be invoked through unconventional ways, such as Java reflection. On the other hand, these APIs are not stable and may be changed or even removed in future Android versions, providing no guarantee for compatibility. From Android 9 (API level 28), Google began to strictly restrict the use of non-SDK APIs, and the corresponding checking mechanism has been integrated into the Android OS.

Bluetooth Low Energy (BLE) is widely deployed and has become the de-facto communication standard in the IoT ecosystem. Naturally, the security of BLE received much attention from both researchers and attackers. In another aspect, the BLE specifications provide the security guidelines for BLE deployments. Due to various reasons, the developers do not follow the guidelines in the implementation process, which introduces the misconfiguration issue. However, identifying these BLE mis-configurations in IoT device firmware is quite challenging. In this work, we do not handle the BLE-enabled devices directly. Instead, we focus on the security misconfiguration issues in their companion mobile apps, which can reflect the deployment conditions of the corresponding devices. Further, we designed an analysis tool - BSC-Checker to detect the misconfigurations based on pre-defined checking strategies. With BSC-Checker, we conducted large-scale experiments on 4,589 apps from multiple app markets. The result shows that the BLE configurations of most BLE apps disobey at least one security rule, and the current BLE deployment status is not optimistic.

Intent is an essential inter-component communication mechanism of Android OS, which can be used to request an action from another app component. The security of its design and implementation attracts lots of attention. However, the security of PendingIntent, a kind of delayed-triggered Intent, was neglected by most previous research, and the related analysis techniques are still imperfect. In this paper, we design a novel automated tool, PITracker, to detect the PendingIntent vulnerabilities in Android apps. It achieves the Intent flow tracking technique proposed by us, figuring out how an Intent is created and where it goes. In the real-world evaluations, PITracker discovered 2,939 potential threats in 10,000 third-party apps and 214 in 1,412 pre-installed apps. Among them, 11 exploitable vulnerabilities have been confirmed and acknowledged by the corresponding vendors.

One intriguing property of deep neural networks (DNNs) is their vulnerability to adversarial examples – those maliciously crafted inputs that deceive target DNNs. While a plethora of defenses have been proposed to mitigate the threats of adversarial examples, they are often penetrated or circumvented by even stronger attacks. To end the constant arms race between attackers and defenders, significant efforts have been devoted to providing certifiable robustness bounds for DNNs, which ensures that for a given input its vicinity does not admit any adversarial instances. Yet, most prior works focus on the case of symmetric vicinities (e.g., a hyperrectangle centered at a given input), while ignoring the inherent heterogeneity of perturbation direction (e.g., the input is more vulnerable along a particular perturbation direction). To bridge the gap, in this article, we propose the concept ofasymmetric robustnessto account for the inherent heterogeneity of perturbation directions, and presentAmoeba1, an efficient certification framework for asymmetric robustness. Through extensive empirical evaluation on state-of-the-art DNNs and benchmark datasets, we show that compared with its symmetric counterpart, the asymmetric robustness bound of a given input describes its local geometric properties in a more precise manner, which enables use cases including (i) modeling stronger adversarial threats, (ii) interpreting DNN predictions, and makes it a more practical definition of certifiable robustness for security-sensitive domains.

Fine-grained hashing is a new topic in the field of hashing-based retrieval and has not been well explored up to now. In this paper, we raise three key issues that fine-grained hashing should address simultaneously, i.e., fine-grained feature extraction, feature refinement as well as a well-designed loss function. In order to address these issues, we propose a novel Fine-graIned haSHing method with a double-filtering mechanism and a proxy-based loss function, FISH for short. Specifically, the double-filtering mechanism consists of two modules, i.e., Space Filtering module and Feature Filtering module, which address the fine-grained feature extraction and feature refinement issues, respectively. Thereinto, the Space Filtering module is designed to highlight the critical regions in images and help the model to capture more subtle and discriminative details; the Feature Filtering module is the key of FISH and aims to further refine extracted features by supervised re- weighting and enhancing. Moreover, the proxy-based loss is adopted to train the model by preserving similarity relationships between data instances and proxy-vectors of each class rather than other data instances, further making FISH much efficient and effective. Experimental results demonstrate that FISH achieves much better retrieval performance compared with state-of-the-art fine-grained hashing methods, and converges very fast. The source code is publicly available: https://github.com/chenzhenduo/FISH.

With the wide use of Automatic Speech Recognition (ASR) in applications such as human machine interaction, simultaneous interpretation, audio transcription, and so on, its security protection becomes increasingly important. Although recent studies have brought to light the weaknesses of popular ASR systems that enable out-of-band signal attack, adversarial attack, and so on, and further proposed various remedies (signal smoothing, adversarial training, etc.), a systematic understanding of ASR security (both attacks and defenses) is still missing, especially on how realistic such threats are and how general existing protection could be. In this article, we present our systematization of knowledge for ASR security and provide a comprehensive taxonomy for existing work based on a modularized workflow. More importantly, we align the research in this domain with that on security in Image Recognition System (IRS), which has been extensively studied, using the domain knowledge in the latter to help understand where we stand in the former. Generally, both IRS and ASR are perceptual systems. Their similarities allow us to systematically study existing literature in ASR security based on the spectrum of attacks and defense solutions proposed for IRS, and pinpoint the directions of more advanced attacks and the directions potentially leading to more effective protection in ASR. In contrast, their differences, especially the complexity of ASR compared with IRS, help us learn unique challenges and opportunities in ASR security. Particularly, our experimental study shows that transfer attacks across ASR models are feasible, even in the absence of knowledge about models (even their types) and training data.

Permission is the fundamental security mechanism for protecting user data and privacy on Android. Given its importance, security researchers have studied the design and usage of permissions from various aspects. However, most of the previous research focused on the security issues ofsystem permissions. Overlooked by many researchers, an app can usecustom permissionsto share its resources and capabilities with other apps. However, the security implications of using custom permissions have not been fully understood. In this paper, we systematically evaluate the design and implementation of Android custom permissions. Notably, we built an automatic fuzzing tool, calledCuPerFuzzer+, to detect custom permission related vulnerabilities existing in the Android OS.CuPerFuzzer+treats the operations of the permission mechanism as a black-box and executes massive targeted test cases to trigger privilege escalation. In the experiments,CuPerFuzzer+discovered 5,932 effective cases with 47 critical paths successfully. Through investigating these vulnerable cases and analyzing the source code of Android OS, we further identified a series of severe design shortcomings lying in the Android permission framework, includingdangling custom permission,inconsistent permission-group mapping,custom permission elevating,inconsistent permission definition,dormant permission group, andinconsistent permission type. Exploiting these shortcomings, a malicious app can access unauthorized platform resources. On top of these observations, we propose three general design guidelines to secure custom permissions. Our findings have been acknowledged by the Android security team and assignedCVE-2020-0418,CVE-2021-0306,CVE-2021-0307, andCVE-2021-0317.

Given a set of local sequential datasets held by multiple parties, we study the problem of publishing a synthetic dataset that preserves approximate sequentiality information of the integrated dataset while satisfying differential privacy for each local dataset. The existing solutions for publishing differentially private sequential data in the centralized setting mostly adopt tree-based approaches. Such approaches rely on different tree structures that encode sequential data's statistical information. The construction of a tree structure is normally done by recursively splitting nodes whose noisy scores (e.g., entropy or count) are larger than a given threshold. However, extending similar ideas to the multi-party setting is challenging. First, the comparison between noisy scores and a given threshold needs to be done in a distributed manner without letting the parties know the noisy scores, while satisfying differential privacy for each local dataset. Second, in the multi-party setting the large number of node splitting decisions incurs prohibitive computation costs. In addressing the above challenges, we present DPST, a distributed prediction suffix tree construction solution. In DPST, we first introduce a novel node splitting decision method that calculates the comparison result under encryption with substantially improved efficiency. Then we present a novel batch-based tree construction approach to reduce the computation costs. In order to achieve high parallel performance without incurring any extra communication cost, we introduce the conjunction and slide methods to ensure that each batch contains a stable number of carefully arranged decision tasks. Extensive experiments on real datasets demonstrate that our DPST solution offers desirable data utility with low computation and communication costs.

Machine learning (ML) techniques have been widely deployed in the field of Android malware detection. On the other hand, ML-based malware detection also faces the threat of adversarial attacks. Recently, some research has demonstrated the possibility of such attacks under the settings of white-box or grey-box. However, a more practical threat model - black-box adversarial attack has not been well validated and evaluated. In this paper, we bridge this research gap and propose a black-box adversarial attack approach, ShadowDroid, against ML-based Android malware detection. On a high level, ShadowDroid tries to construct a substitute model of the target malware detection system. Utilizing this substitute model, we can identify and modify the key features of a malicious app to generate an adversarial sample. During the experiment, we evaluated the effectiveness of ShadowDroid against nine ML-based Android malware detection frameworks. It achieved successful malware evading on five platforms. Based on these results, we also discuss how to design a robust malware detection system to prevent adversarial attacks.

Permission is the fundamental security mechanism for protecting user data and privacy on Android. Given its importance, security researchers have studied the design and usage of permissions from various aspects. However, most of the previous research focused on the security issues of system permissions. Overlooked by many researchers, an app can use custom permissions to share its resources and capabilities with other apps. However, the security implications of using custom permissions have not been fully understood.In this paper, we systematically evaluate the design and implementation of Android custom permissions. Notably, we built an automatic fuzzing tool, called CuPerFuzzer, to detect custom permissions related vulnerabilities existing in the Android OS. CuPerFuzzer treats the operations of the permission mechanism as a black-box and executes massive targeted test cases to trigger privilege escalation. In the experiments, CuPerFuzzer discovered 2,384 effective cases with 30 critical paths successfully. Through investigating these vulnerable cases and analyzing the source code of Android OS, we further identified a series of severe design shortcomings lying in the Android permission framework, including dangling custom permission, inconsistent permission-group mapping, custom permission elevating, and inconsistent permission definition. Exploiting any of these shortcomings, a malicious app can obtain dangeroussystem permissions without user consent and further access unauthorized platform resources. On top of these observations, we propose some general design guidelines to secure custom permissions. Our findings have been acknowledged by the Android security team and rated as High severity.

The rapid development of Android apps primarily benefits from third-party libraries that provide well-encapsulated functionalities. On the other hand, more and more malicious libraries are discovered in the wild, which brings new security challenges. Despite some previous studies focusing on the malicious libraries, however, most of them only study specific types of libraries or individual cases. The security community still lacks a comprehensive understanding of potentially malicious libraries (PMLs) in the wild.

Deep learning techniques have made tremendous progress in a variety of challenging tasks, such as image recognition and machine translation, during the past decade. Training deep neural networks is computationally expensive and requires both human and intellectual resources. Therefore, it is necessary to protect the intellectual property of the model and externally verify the ownership of the model. However, previous studies either fail to defend against the evasion attack or have not explicitly dealt with fraudulent claims of ownership by adversaries. Furthermore, they can not establish a clear association between the model and the creator's identity.

The features of modularity and inheritance in C++ facilitate the developers' usage, but also give rise to the problem of type confusion. As an ancestor class may have a different data layout from its descendant class, a dangerous downcasting operation from the ancestor to its descendant can lead to a critical attack, such as control flow hijacking, out-of-bounds access to neighbor memory area, etc. As reported in CVE, such vulnerabilities have been found in various common-used software, including Google Chrome, Firefox and Adobe Flash Player, and have a trend of increase in recent years. The urgency of addressing type confusion problems quickens the pace of researchers coming to corresponding solutions. However, the existing works either handle the problem partially, or suffer from the high performance and memory overhead, especially to the large-scale projects.

Recent years have witnessed the enormous growth of Android phones in the consumer market. On the other hand, as the most popular mobile platform, Android also attracts lots of attackers' attention. As a result, more and more Android malicious apps appear in the wild, which poses a serious threat to user's security and privacy. To such massive volume of Android malware, automated UI testing techniques have become the mainstream solutions because of the detection efficiency and accuracy. However, all existing UI testing techniques treat the Activity as the basic unit of UI interactions and cannot carry out a fine-grained analysis for Fragments. Due to the lack of Fragment-level analysis, the path coverage is usually quite limited. To fill this gap, in this paper, we propose FragDroid, a novel automated UI testing framework supporting both Activity and Fragment analysis. To achieve the Fragment-level testing, we design the Activity & Fragment Transition Model (AFTM) to simulate the internal interactions of an app, and ATFM could be utilized to generate test cases automatically through UI interactions. With the assist of AFTM, FragDroid achieves accessing most Activities and Fragments contained in the app along with the capability of detecting arbitrary API calls. We implemented a prototype of FragDroid and evaluated it on 15 popular apps. The results show FragDroid successfully covered 66% Fragments and the corresponding API calls of testing apps. Also, the traditional approaches have to miss at least 9.6% of API calls invoked in Fragments.

The Secure Sockets Layer (SSL) and Transport Layer Security (TLS) protocols are the foundation of network security. The certificate verification in SSL/TLS implementations is vital and may become the "weak link" in the whole network ecosystem. In previous works, some research focused on the automated testing of certificate verification, and the main approaches rely on generating massive certificates through randomly combining parts of seed certificates for fuzzing. Although the generated certificates could meet the semantic constraints, the cost is quite heavy, and the performance is limited due to the randomness. To fill this gap, in this paper, we propose DRLGENCERT, the first framework of applying deep reinforcement learning to the automated testing of certificate verification in SSL/TLS implementations. DRLGENCERT accepts ordinary certificates as input and outputs newly generated certificates which could trigger discrepancies with high efficiency. Benefited by the deep reinforcement learning, when generating certificates, our framework could choose the best next action according to the result of a previous modification, instead of simple random combinations. At the same time, we developed a set of new techniques to support the overall design, like new feature extraction method for X.509 certificates, fine-grained differential testing, and so forth. Also, we implemented a prototype of DRLGENCERT and carried out a series of real-world experiments. The results show DRLGENCERT is quite efficient, and we obtained 84,661 discrepancy-triggering certificates from 181,900 certificate seeds, say around 46.5% effectiveness. Also, we evaluated six popular SSL/TLS implementations, including GnuTLS, MatrixSSL, MbedTLS, NSS, OpenSSL, and wolfSSL. DRLGENCERT successfully discovered 23 serious certificate verification flaws, and most of them were previously unknown.

The development of positioning technologies has digitalized people's mobility traces for the first time in history. GPS sensors resided in people's mobile devices allow smart apps to access location data. This large amount of mobility data can help to build appealing applications. Meanwhile, location privacy has become a major concern. In this paper, we design a general system to assess whether an app is vulnerable to location inference attacks. We utilize a series of automatic testing mechanisms including UI match and API analysis to extract the location information an app provides. According to different characteristics of these apps, we classify them into two categories corresponding to two kinds of attacks, namely attack with distance limitation (AWDL) and attack without distance limitation (AWODL). After evaluating 800 apps, of which 109 passed automated testing, we found that 24.7% of the passing apps are vulnerable to AWDL and 11.0% to AWODL. Moreover, some apps even allow us to modify the parameters in http requests which largely increases the scope of the attacks. Our system demonstrates the severity of location privacy leakage to mobile devices and can serve as an auditing tool for future smart apps.

Hashing technique has attracted considerable attention for large-scale multimedia retrieval due to its low storage cost and fast query speed. Moreover, many hashing models have been proposed for cross-modal retrieval task. However, there are still some problems that need to be further considered. For example, a majority of them directly use linear projection matrix to project heterogeneous data into a common space, which may lead to large error as there are some heterogeneous data with semantic similarity hard to be close in latent space when linear projection is used. Besides, most existing cross-modal hashing methods use a simple pairwise similarity matrix for preserving the label information when learning. This kind of pairwise similarity cannot fully utilize the discriminative property of label information. Furthermore, most existing supervised ones try to solve a relaxed continuous optimization problem by dropping the discrete constraints, which may lead to large quantization error. To overcome these limitations, in this paper, we propose a novel cross-modal hashing method, called Dictionary Learning based Supervised Discrete Hashing (DLSDH). Specifically, it learns dictionaries and generates sparse representation for every instance, which is more suitable to be projected to a latent space. To make full use of label information, it uses cosine similarity to construct a new pairwise similarity matrix which can contain more information. Moreover, it directly learns the discrete hash codes instead of relaxing the discrete constraints. Extensive experiments are conducted on three benchmark datasets and the results demonstrate that it outperforms several state-of-the-art methods for cross-modal retrieval task.

Recently, hashing based approximate nearest neighbor search has attracted much attention in large scale data search task. Moreover, some cross-modal hashing methods have also been proposed to perform efficient search of different modalities. However, there are still some problems to be further considered. For example, some of them cannot make use of label information, which contains helpful information to generate hash codes; some of them firstly relax binary constraints during optimization, then threshold continuous outputs to binary, which could generate large quantization error. To consider these problems, in this paper, we propose a supervised cross-modal hashing without relaxation (SCMH-WR). It can not only make use of label information, but also generate the final binary codes directly, i.e., without relaxing binary constraints. Specifically, it maps different modalities into a common low-dimension subspace with preserving the similarity of labels; at the same time, it learns a rotation matrix to minimize the quantization error and gets the final binary codes. In addition, an iterative algorithm is proposed to tackle the optimization problem. SCMH-WR is tested on three benchmark data sets. Experimental results demonstrate that SCMH-WR outperforms state-of-the-art hashing methods for cross-modal search task.

In the Android system design, any app can start another app's public components to facilitate code reuse by sending an asynchronous message called Intent. In addition, Android also allows an app to have private components that should only be visible to the app itself. However, malicious apps can bypass this system protection and directly invoke private components in vulnerable apps through a class of newly discovered vulnerability, which is called next-intent vulnerability. In this paper, we design an intent flow analysis strategy which accurately tracks the intent in smali code to statically detect next-intent vulnerabilities efficiently and effectively on a large scale. We further propose an automated approach to dynamically verify the discovered vulnerabilities by generating exploit apps. Then we implement a tool named NIVAnalyzer and evaluate it on 20,000 apps downloaded from Google Play. As the result, we successfully confirms 190 vulnerable apps, some of which even have millions of downloads. We also confirmed that an open-source project and a third-party SDK, which are still used by other apps, have next intent vulnerabilities.

Recently, multimodal hashing techniques have received considerable attention due to their low storage cost and fast query speed for multimodal data retrieval. Many methods have been proposed; however, there are still some problems that need to be further considered. For example, some of these methods just use a similarity matrix for learning hash functions which will discard some useful information contained in original data; some of them relax binary constraints or separate the process of learning hash functions and binary codes into two independent stages to bypass the obstacle of handling the discrete constraints on binary codes for optimization, which may generate large quantization error; some of them are not robust to noise. All these problems may degrade the performance of a model. To consider these problems, in this paper, we propose a novel supervised hashing framework for cross-modal retrieval, i.e., Supervised Robust Discrete Multimodal Hashing (SRDMH). Specifically, SRDMH tries to make final binary codes preserve label information as same as that in original data so that it can leverage more label information to supervise the binary codes learning. In addition, it learns hashing functions and binary codes directly instead of relaxing the binary constraints so as to avoid large quantization error problem. Moreover, to make it robust and easy to solve, we further integrate a flexible l2,p loss with nonlinear kernel embedding and an intermediate presentation of each instance. Finally, an alternating algorithm is proposed to solve the optimization problem in SRDMH. Extensive experiments are conducted on three benchmark data sets. The results demonstrate that the proposed method (SRDMH) outperforms or is comparable to several state-of-the-art methods for cross-modal retrieval task.

Development of the Smart City has produced much data with attributions of timestamp and location, but in some applications like investigation of the large bomb explosion in New York, the government takes precedence to investigate the relation data from New York city rather than the whole Country, which prompts us to do some research works in computing partial graph more fast. So we propose SpatialGraphx, a graph parallel computing framework supporting direct and fast partial graph construction and partial graph computation. Leveraging the spatial and temporal attributions of data, SpatialGraphx presents two extensions on the partial graph construction by building a spatio-temporal tree index and on the computation by a new location-based partition strategy. Using mobile network's data with hundred million edges, we demonstrate SpatialGraphx can support direct and fast partial graph construction and enables efficient partial graph analysis for spatial and temporal data. And compared to original Graphx, the improvement of SpatialGraphx is 3x to several orders of magnitude for large enough dataset.

Many service providers including large enterprises have released their own applications (apps) that incorporate HTTP clients to facilitate the communications with their servers. The workflows of and APIs used by a web app and its corresponding mobile app are not always the same. We call the APIs found in apps private web APIs in that they are only supposed to be invoked by apps that developed by the service providers themselves. However, checking the origin of an HTTP request is very difficult, and private web APIs can be easily invoked by other entities. Hence, it is imperative to study if private web APIs provide the same level of security checks and validations as their public counterparts. To automatically discover the undocumented private APIs in Android apps, we design a system that uses static analysis to find the activities that invoke web APIs. Our system then runs the discovered activities on a customized Android system to monitor its HTTP requests and responses. We evaluated our system on 76 popular apps on the Google Play market. Our system successfully run 48 apps and discovered many private server-side APIs from more than 30 apps. Further manual investigation discovered that 9 of the apps have vulnerabilities that would enable API misuse and session hijacking.

Today, there are many hybrid apps in which both native Android app UI and WebView UI are used. To protect the security and privacy of the communications, these hybrid apps all use HTTPS by WebView, a key component in modern web browser. In this paper, we show there is another type of SSL vulnerability that stems from the error-handling code in the hybrid mobile web apps. At a high level, this error-handling code should have stopped the communication but it still proceeds regardless of certificate errors, thereby leading to the MITM attacks. To automatically identify these vulnerable apps, we present a hybrid approach that combines both static analysis and dynamic analysis. We have implemented our approach and evaluated with 13,820 real world mobile web apps from a third party market, of which 645 are confirmed truly vulnerable, with an average overhead of 60.8 seconds per app.

Security of authentication protocols heavily relies on the confidentiality of credentials (or authenticators) like passwords and session IDs. However, unlike browser-based web applications for which highly evolved browsers manage the authenticators, Android apps have to construct their own management. We find that most apps simply locate their authenticators into the persistent storage and entrust underlying Android OS for mediation. Consequently, these authenticators can be leaked through compromised backup channels. In this work, we conduct the first systematic investigation on this previously overlooked attack vector. We find that nearly all backup apps on Google Play inadvertently expose backup data to any app with internet and SD card permissions. With this exposure, the malicious apps can steal other apps' authenticators and obtain complete control over the authenticated sessions. We show that this can be stealthily and efficiently done by building a proof-of-concept app named AuthSniffer. We find that 80 (68.4%) out of the 117 tested top-ranked apps which have implemented authentication schemes are subject to this threat. Our study should raise the awareness of app developers and protocol analysts about this attack vector.

Global darknet monitoring provides an effective way to observe cyber-attacks that are significantly threatening network security and management. In this paper, we present a study on characterization of cyberattacks in the big stream data collected in a large scale distributed darknet using association rule learning. The experiment shows that association rule learning in the darknet stream data can support strategic cyberattack countermeasure in the following ways. First, statistics computed from malware-specific rules can lead to better understanding of the global trend of cyberattacks in the Internet. Second, strong association rules can lead to further insights into the nature of the attacking tools and hence expedite the diagnosis. Then, the discovery of emerging new attacks may lead to early detection and prompt prevention of pandemic incidents, preventing damage to the IT infrastructure and extensive financial loss. Finally, exploring the knowledge in the frequent attacking patterns can enable accurate prediction of future attacks from analyzed hosts, which could improve the performance of honeypot systems to collect more pertinent malware information using limited system and network resources.

An enormous number of applications have been developed for Android in recent years, making it one of the most popular mobile operating systems. However, it is obvious that more vulnerabilities would appear along with the booming amounts of applications. Poorly designed applications may contain security vulnerabilities that can dramatically undermine users' security and privacy. In this paper, we studied a kind of recently reported application vulnerability named confused deputy - a specific type of privilege escalation vulnerability, which can result in unauthorized operations, and so on. We proposed a novel system with code-level static analysis to analyze the applications and automatically detect possible confused deputy vulnerabilities. To tackle analysis challenges imposed by Android's component-based programming paradigm, we employed special control flow graph construction techniques to build call relations among components and function call graph within components. We developed a prototype of this system named PaddyFrog and evaluated with 7190 real world Android applications from two of the most popular markets in China. We found 1240 applications with confused deputy vulnerability and proved to be exploitable. The median execution time of this system on an application is 14.4s, which is fast enough to be used in volumes of applications testing scenarios. Copyright © 2015 John Wiley & Sons, Ltd.

Packing is among the most popular obfuscation techniques to impede anti-virus scanners from successfully detecting malware. In this paper we propose a string-kernel-based support vector machine classifier to identify the packer that is used to create a given malware program. Our approach is featured by the following characteristics. First, the adoption of a string-kernel-based method bridges the gap between signature-based and machine-learning-base approaches. Second, the kernel function derived from the Levenshtein distance integrates important domain knowledge in the learning process. Then, application of support vector machine, a state-of-the-art classifier, enables an automated packer identification scheme with high generalization ability and time efficiency. Finally, selection of the code segment with the most essential packer relevant information further boosts the classification performance. Experiments on a dataset of 3228 binary programs composed of packed files created by 25 packers show that the proposed approach outperforms PEiD and previous machine-learning-based approaches in prediction accuracy with a large margin. This method can help to improve the scanning efficiency of anti-virus products and promote efficient back-end malware research.

Characterization of Peer-to-Peer (P2P) traffic is an essential step to develop workload models towards capacity planning and cyber-threat countermeasure over P2P networks. In this paper, we present a new scheme for characterizing P2P file sharing hosts based on transport layer statistical features. The proposed scheme is featured by its tunability over monitoring cost, system response time, and prediction accuracy. We further employ feature selection to identify the most essential discriminators for the analysis. Experimental results show that an equally accurate system could be obtained using only 3 out of the 18 defined discriminators, which further enhances the adaptability and reduces the monitoring cost of the system.

Recent statistical studies on telecommunication networks outline that peer-to-peer (P2P) file-sharing is keeping increasing and it now contributes about 50-80% of the overall Internet traffic. Moreover, more and more network applications such as streaming media, internet telephony, and instant messaging are taking a form of P2P telecommunication. The bandwidth intensive nature of P2P applications suggests that P2P traffic can have significant impact on the underlying network. Therefore, analyzing and characterizing this kind of traffic is an essential step to develop workload models towards efficient amelioration in network traffic engineering and capacity planning. In this paper, we first introduce an adaptive system for handy P2P trace capturing and analysis. By using virtualization technology, the system can efficiently organize limited resources to build a reliable and tractable network that supports adjustable experimental study and practical performance tuning. Then the proposed system is applied to traffic characterization of File Sharing P2P (FSP2P) applications. To avoid excessive computing cost of payload information inspection, we proposed a more light-weighted analytical scheme which makes use of meta features extracted from packet headers. With carefully selected system parameters, we show that satisfactory prediction accuracy on differentiating FSP2P applications from ordinary network applications could be achieved with acceptable computing costs. The proposed scheme supports performance tuning between monitoring cost and the system response time, which enables its adaption to network environments with different specifications.

Identifying heavy-hitter flows in high-speed network link is important for some applications. This paper studied the approach of measuring various heavy-hitter flows simultaneously. We proposed a novel scheme, named TS-LRU (Two-Stage Least Recently Used), which process arriving packets through two stages to extract heavy-hitter flows. New packets are aggregated into FGFs (Fine-Grained Flow) and preserved in Stage1. The FGFs with no arrival packets for a relative long time are evicted from Stage1 using LRU replacement. The replaced FGFs are added into Stage2 and aggregated into RGFs (Rough-Grained Flow) further. The replacement scheme used in Stage2 is based on LRU with considering RGF size, named LRU-Size. There could be several similar data structures in Stage2 to extract different types of RGFs concurrently. Mathematical analysis indicates that this algorithm can save memory space and improve processing speed efficiently through exploiting the distribution characteristics of flows. We also examined TS-LRU with simulated experiments on real packet traces. Other than the proportional increasing of common approaches, the average processing time per packet of TS-LRU increases more slowly when measure multiple types of flows concurrently. Compared to the well-known multi-stage filters algorithm, TS-LRU achieves superior performance in terms of measurement accuracy in constrained memory space.

Wireless sensor networks (WSNs) deployed in hostile environments are vulnerable to clone attacks. In such attack, an adversary compromises a few nodes, replicates them, and inserts arbitrary number of replicas into the network. Consequently, the adversary can carry out many internal attacks. Previous solutions on detecting clone attacks have several drawbacks. First, some of them require a central control, which introduces several inherent limits. Second, some of them are deterministic and vulnerable to simple witness compromising attacks. Third, in some solutions the adversary can easily learn the critical witness nodes to start smart attacks and protect replicas from being detected. In this paper, we first show that in order to avoid existing drawbacks, replica-detection protocols must be non-deterministic and fully distributed (NDFD), and fulfill three security requirements on witness selection. To our knowledge, only one existing protocol, Randomized Multicast, is NDFD and fulfills the requirements, but it has very high communication overhead. Then, based on random walk, we propose two new NDFD protocols, RAndom WaLk (RAWL) and Table-assisted RAndom WaLk (TRAWL), which fulfill the requirements while having only moderate communication and memory overheads. The random walk strategy outperforms previous strategies because it distributes a core step, the witness selection, to every passed node of random walks, and then the adversary cannot easily find out the critical witness nodes. We theoretically analyze the required number of walk steps for ensuring detection. Our simulation results show that our protocols outperform an existing NDFD protocol with the lowest overheads in witness selection, and TRAWL even has lower memory overhead than that protocol. The communication overheads of our protocols are higher but are affordable considering their security benefits.

Clustering has been widely used in wireless ad hoc networks for various purposes such as routing, broadcasting and Qos. Many clustering algorithms have been proposed. However, most of them implicitly assume that nodes behave honestly in the clustering process. In practice, there might be some malicious nodes trying to manipulate the clustering process to make them serve as clusterheads, which can obtain some special power, e.g., eavesdropping more messages. In this paper we present a secure weighted clustering algorithm (SWCA). SWCA uses the weighted clustering algorithm (WCA) for clustering and TELSA for efficiently authenticating packets. We propose a novel neighbor verification scheme to check whether the values of election-related features (e.g., node degree) are forged by malicious nodes. Also, we theoretically analyze the probability for a malicious node to tamper node degree without being detected and derive a lower bound on the probability. Finally, simulation results show that SWCA is secure but still has comparable performance with WCA. To the best of our knowledge, SWCA is the first algorithm considering the security of 1-hop type clustering (in this type only the clusterhead can communicate with ordinary members directly) in ad hoc networks.

Many secure localization algorithms have been proposed. In these algorithms, collusion attack is usually considered as the strongest attack when evaluating their performance. Also, for ensuring correct localization under the collusion attack, a necessary number of normal beacons are needed and a lower bound on this number has been established (assuming the errors of distance measurements are ignorable). In this paper, we introduce pollution attack, a more powerful attack which can succeed even when the number of normal beacons is more than the lower bound. In this attack, victim node is misled to a special chosen location, which results in a confusion of compromised beacon with normal beacon. We propose a new metric to measure the vulnerability of a normal location reference set to pollution attack, and develop two algorithms to efficiently compute the value of the proposed metric. We also present a method to judge whether the output of the localization algorithm is credible under pollution attack. Simulation results show that the pollution attack can succeed with high probability.