Abstract Unprotected gradient exchange in federated learning (FL) systems may lead to gradient leakage-related attacks. CKKS is a promising approximate homomorphic encryption scheme to protect gradients, owing to its unique capability of performing operations directly on ciphertexts. However, configuring CKKS security parameters involves a trade-off between correctness, efficiency, and security. An evaluation gap exists regarding how these parameters impact computational performance. Additionally, the maximum vector length that CKKS can once encrypt, recommended by Homomorphic Encryption Standardization, is 16384, hampers its widespread adoption in FL when encrypting layers with numerous neurons. To protect gradients’ privacy in FL systems while maintaining practical performance, we comprehensively analyze the influence of security parameters such as polynomial modulus degree and coefficient modulus on homomorphic operations. Derived from our evaluation findings, we provide a method for selecting the optimal multiplication depth while meeting operational requirements. Then, we introduce an adaptive segmented encryption method tailored for CKKS, circumventing its encryption length constraint and enhancing its processing ability to encrypt neural network models. Finally, we present FedSHE , a privacy-preserving and efficient Fed erated learning scheme with adaptive S egmented CKKS H omomorphic E ncryption. FedSHE is implemented on top of the federated averaging (FedAvg) algorithm and is available at https://github.com/yooopan/FedSHE . Our evaluation results affirm the correctness and effectiveness of our proposed method, demonstrating that FedSHE outperforms existing homomorphic encryption-based federated learning research efforts in terms of model accuracy, computational efficiency, communication cost, and security level.

Neural language models have achieved superior performance.However, these models also suffer from the pathology of overconfidence in the out-of-distribution examples, potentially making the model difficult to interpret and making the interpretation methods fail to provide faithful attributions.In this paper, we explain the model pathology from the view of sentence representation and argue that the counter-intuitive bias degree and direction of the out-of-distribution examples' representation cause the pathology.We propose a Contrastive learning regularization method using Adversarial examples for Alleviating the Pathology (ConAAP), which calibrates the sentence representation of out-of-distribution examples.ConAAP generates positive and negative examples following the attribution results and utilizes adversarial examples to introduce direction information in regularization.Experiments show that ConAAP effectively alleviates the model pathology while slightly impacting the generalization ability on in-distribution examples and thus helps interpretation methods obtain more faithful results.

In the data-driven economy, data sharing is a key approach to unleashing the true value of data. Blockchain, as a decentralized ledger, can provide a trusted data sharing platform in an untrusted environment. However, existing blockchain-based data sharing schemes suffer from inefficiency and inadequate protection of security and privacy. To address the above issues, we propose a blockchain-based privacy-preserving data sharing scheme with security-enhanced access control. In the scheme, a secure data sharing architecture using dual-blockchain and the interplanetary file system (IPFS) is presented to provide decentralized and scalable storage. Based on the architecture, a blockchain-assisted multi-authority attribute-based encryption (BA-MA-ABE) algorithm with efficient attribute revocation and computation is designed in our work. Our BA-MA-ABE lever-ages blockchain to securely manage partial decryption keys and provides fine-grained access control over encrypted data. We also devise smart contracts that can support traceable access control over the flow of data while protecting user identity privacy with verifiable attribute credentials. In comparison with some existing work, our scheme shows more comprehensive security features with lower user computation overhead.

Microservice architecture is embraced by a growing number of enterprises due to the benefits of modularity and flexibility. However, being composed of numerous interdependent microservices, it is prone to cascading failures and afflicted by the arising problem of troubleshooting, which entails arduous efforts to identify the root cause node and ensure service availability. Previous works use call graph to characterize causality relation-ships of microservices but not completely or comprehensively, leading to an insufficient search of potential root cause nodes and consequently poor accuracy in culprit localization. In this paper, we propose ImpactTracer to address the above problems. ImpactTracer builds impact graph to provide a com-plete view of fault propagation in microservices and uses a novel backward tracing algorithm that exhaustively traverses the impact graph to identify the root cause node accurately. Extensive experiments on a real-world dataset demonstrate that ImpactTracer is effective in identifying the root cause node and outperforms the state-of-the-art methods by at least 72%, significantly facilitating troubleshooting in microservices.

True Random Number Generator (TRNG) is indispensable in cryptographic algorithms and protocols, and the quality of randomness directly influences the security of cryptographic applications. Multiple theoretical or offline entropy estimation methods have been proposed to evaluate the security of TRNGs, while their ideal assumptions commonly cannot be satisfied due to the perturbation of operating conditions at runtime, which makes it difficult to achieve sufficient entropy for the output of TRNGs in practice. Moreover, the output bitrate of TRNG is another fundamental concern during TRNG practical applications, while popular elementary oscillator-based structure commonly has relatively low output bitrate due to the inherent low sensitivity of entropy extraction to jitter (source of randomness). In this paper, we aim to design a TRNG satisfying both practical security (i.e., on-chip entropy assurance) and high output bitrate simultaneously. In particular, an improved stochastic model and a measurement method are established to quantify the entropy of coherent sampling based TRNG. Moreover, an on-chip entropy assurance module is provided to realize the robustness of the proposed design under various operating conditions. We implement the proposed TRNG in a simulation platform and ASIC chips (with SMIC 130 nm CMOS technology). Experimental results indicate that the generated data has sufficient entropy ($\geq 0.999$per bit) under various operating conditions. In addition, all the output can pass the NIST SP800-22 and AIS 31 statistical tests with an output bitrate of 4.2 Mbps, which is equivalent to 2 orders of magnitude faster than that of the elementary oscillator-based TRNG.

Neural networks are vulnerable to adversarial examples. The adversary can successfully attack a model even without knowing model architecture and parameters, i.e., under a black-box scenario. Previous works on word-level attacks widely use word importance ranking (WIR) methods and complex search methods, including greedy search and heuristic algorithms, to find optimal substitutions. However, these methods fail to balance the attack success rate and the cost of attacks, such as the number of queries to the model and the time consumption. In this paper, We propose PAthological woRd Saliency sEarch (PARSE) that performs the search under dynamic search space following the subarea importance. Experiments show that PARSE can achieve comparable attack success rates to complex search methods while saving numerous queries and time, e.g., saving at most 74% of queries and 90% of time compared with greedy search when attacking the examples from Yelp dataset. The adversarial examples crafted by PARSE are also of high quality, highly transferable, and can effectively improve model robustness in adversarial training.

In the insider threat detection domain, data imbalance is a well-known problem. Most existing solutions, including rebalancing datasets and anomaly detection, have problems such as model overfitting, high cost, and high False Positive Rate (FPR). Therefore, how to effectively detect insider threats on an imbalanced dataset is a challenge. This paper proposes a new Siamese-architecture Insider Threat Detection (SITD) method, which detects insider threat by judging whether the input sample pairs belong to the same category instead of directly classifying a sample while avoiding the abovementioned problems. In addition, we improve the contrastive loss function to make the model pay more attention to the samples pairs of different categories, which significantly enhances the detection performance. Experimental results show that SITD outperforms other insider detection methods on the imbalanced CERT dataset. Moreover, SITD can achieve a good result no matter how imbalanced the dataset is.

Textual adversarial examples crafted with well-designed perturbation can mislead state-of-the-art natural language models. Most previous works on word-level black-box attacks propose different text substituting strategies based on the word saliency determined by Leave One Out (LOO) methods, while the attack effectiveness is actually limited due to the model pathology. The word saliency determined by LOO methods can be severely affected by the model pathology, and unconscious bias is introduced. In this paper, we propose a word saliency method called Second-Order Enhanced Word Saliency (SOEWS), which considers the overfitting information obtained from the second-order model pathological behavior that helps to reduce the bias of LOO methods. Extensive experiments show that our method outperforms current black-box word saliency methods and can boost the effectiveness of previous attack frameworks. Performing adversarial training utilizing our method best improves model robustness, and human evaluation shows that the adversarial examples crafted by our method are almost imperceptible to humans. We also conduct case study to intuitively demonstrate the superiority of our method.

Random number generator (RNG) is a fundamental and important cryptographic element, which has made an outstanding contribution to guaranteeing the network and communication security of cryptographic applications in the Internet age. In reality, if the random number used cannot provide sufficient randomness (unpredictability) as expected, these cryptographic applications are vulnerable to security threats and cause system crashes. Min-entropy is one of the approaches that are usually employed to quantify the unpredictability. The NIST Special Publication 800-90B adopts the concept of min-entropy in the design of its statistical entropy estimation methods, and the predictive model-based estimators added in the second draft of this standard effectively improve the overall capability of the test suite. However, these predictors have problems on limited application scope and high computational complexity, e.g., they have shortfalls in evaluating random numbers with long dependence and multivariate due to the huge time complexity (i.e., high-order polynomial time complexity). Fortunately, there has been increasing attention to using neural networks to model and forecast time series, and random numbers are also a type of time series. In our work, we propose several new and efficient approaches for min-entropy estimation by using neural network technologies and design a novel execution strategy for the proposed entropy estimation to make it applicable to the validation of both stationary and nonstationary sources. Compared with the 90B’s predictors officially published in 2018, the experimental results on various simulated and real-world data sources demonstrate that our predictors have a better performance on the accuracy, scope of applicability, and execution efficiency. The average execution efficiency of our predictors can be up to 10 times higher than that of the 90B’s for 10 6 sample size with different sample spaces. Furthermore, when the sample space is over 2 2 and the sample size is over 10 8 , the 90B’s predictors cannot give estimated results. Instead, our predictors can still provide accurate results. Copyright© 2019 John Wiley & Sons, Ltd.

Widely used as fundamental security components in most cryptographic applications, random number generators (RNGs) rely mainly on randomness provided by entropy sources. If the provided randomness is less than expected, RNGs may be compromised and thus impair the security of the whole cryptographic applications. However, the common assumptions (e.g., outputs are independent and identically distributed, i.e., IID) may not always hold. For example, many entropy sources are based on some physical phenomena that are fragile and sensitive to external factors (e.g., temperature), which means the distributions of these entropy sources' outputs are continuously changing. As important tools to measure the quality of entropy sources, existing entropy estimation methods may provide false estimations against these time-varying data, because they cannot detect the changes of data distributions. In this paper, we firstly review and analyze the existing typical entropy estimators including the NIST SP 800-90B (90B for short) estimators and the lately proposed neural network based (NN-based) estimators, especially, their limitations on the aforementioned time-varying data. Second, we propose an entropy estimation framework adopting change detection techniques to address this problem. In contrast to the NN-based estimators, the proposed estimator under this framework employs a change detection method to preprocess the tested data and adds additional distribution features to each data sample, which makes it possible to learn the distribution changes and estimate the entropy more accurately. Finally, we evaluate the performance of our estimator using various kinds of simulated data and real world data, and compare our estimator with the 90B estimators and the NN-based estimators. Extensive evaluations demonstrate that the proposed estimator provides similar or more accurate entropy estimation than the other estimators, especially for time-varying data.

Trust mechanisms based on trusted authorities are used to establish trust between two unfamiliar entities in conventional authentication, which cannot meet the higher security requirements of emerging technologies for authentication. Blockchain achieves a decentralized trust mechanism between a group of mutually distrusted nodes without a trusted authority by means of data encryption, timestamp and consensus mechanisms. In this work, we therefore propose an innovative blockchain-based authentication method with one-time password (OTP). Specifically, we propose an OTP scheme and apply it to the blockchain, where blockchain serves as an OTP verifier. We then develop a list of security criteria to analyze the security of the proposed authentication protocol. This paper shows that our scheme can resist replay attacks, brute force attacks, OTP forgery attacks and so on. We finally compare the proposed OTP scheme with the representative OTP schemes according to the designed evaluation criteria and conduct a series of comparative experiments on Hyperledger Fabric. The results show that the proposed scheme is superior in terms of performance and security.

Recent studies have shown that co-resident attacks have aroused great security threat in cloud. Since hardware is shared among different tenants, malicious tenants can launch various co-resident attacks, such as side channel attacks, covert channel attacks and resource interference attacks. Existing countermeasures have their limitations and can not provide comprehensive defense against co-resident attacks. This paper combines the advantages of various countermeasures and proposes a complete co-resident threat defense solution which consists of co-resident-resistant VM allocation (CRRVA), analytic hierarchy process-based threat score mechanism (AHPTSM) and attack-aware VM reallocation (AAVR). CRRVA securely allocates VMs and also takes load balance and power consumption into consideration to make the allocation policy more practical. According to the intrinsic characteristics of co-resident attacks, AHPTSM evaluates VM’s threat score which denotes the probability that a VM is suffering or conducting co-resident attacks based on analytic hierarchy process. And AAVR further migrates VMs with extremely high threat scores and separates VM pairs which are likely to be malicious to each other. Extensive experiments in CloudSim have shown that CRRVA can greatly reduce the allocation co-resident threat as well as balancing the load for both CSPs and tenants with little impact on power consumption. In addition, guided by threat score distribution, AAVR can effectively guarantee runtime co-resident security by migrating high threat score VMs with less migration cost.

True random number generators (TRNGs) are widely used in cryptographic systems, and their security is the base of many cryptographic algorithms and protocols. At present, entropy estimation based on a stochastic model is a well-recommended approach to evaluate the security of a specific TRNG structure. Besides, the generation speed is also an important property for TRNGs. For this purpose, an analog-to-digital converter (ADC) can be employed to sample the noisy signal to achieve high bit rate. However, no research focuses on the entropy estimation on the basis of the stochastic model toward ADC sampling. In this paper, we propose an entropy estimation for the ADC sampling-based TRNG through extending an existing model. In particular, we present an equivalent model to estimate the entropy of any single bit in the converted sample obtained by the ADC sampling. Furthermore, we propose a method of the entropy estimation for the multi-bit ADC output, which provides the lower bound of the entropy. By conducting simulations and hardware experiments on this type of TRNG, we confirm the correctness of the proposed entropy estimation theory. The prototype chip is fabricated in the SMIC 65-nm process, and the consumed power is 34 mW. The random bit sequences compatible with the AIS 31 standard are generated at a speed of 132.3 Mb/s. The sequences are able to pass the rigorous statistical test suites, including NIST SP 800-22, Diehard, and TestU01 (containing the Big Crush test), after simple post-processing at a bit rate of around 33 Mb/s.