Biblio

Static system configuration provides a significant advantage for the adversaries to discover the assets and launch attacks. Configuration-based moving target defense (MTD) reverses the cyber warfare asymmetry by mutating certain configuration parameters to disrupt the attack planning or increase the attack cost significantly. In this research, we present a methodology for the formal verification of MTD techniques. We formally modeled MTD techniques and verified them against constraints. We use Random Host Mutation (RHM) as a case study for MTD formal verification. The RHM transparently mutates the IP addresses of end-hosts and turns into untraceable moving targets. We apply the formal methodology to verify the correctness, safety, mutation, mutation quality, and deadlock-freeness of RHM using the model checking tool. An adversary is also modeled to validate the effectiveness of the MTD technique. Our experimentation validates the scalability and feasibility of the formal verification methodology.

Anomaly detection (AD) is very important across several real-world problems in the heavy industries and Internet-of-Things (IoT) domains. Traditional methods so far have categorized anomaly detection into (a) unsupervised, (b) semi-supervised and (c) supervised techniques. A relatively unexplored direction is the development of context aware anomaly detection systems which can build on top of any of these three techniques by using side information. Context can be captured from a different modality such as semantic graphs encoding grouping of sensors governed by the physics of the asset. Process flow diagrams of an operational plant depicting causal relationships between sensors can also provide useful context for ML algorithms. Capturing such semantics by itself can be pretty challenging, however, our paper mainly focuses on, (a) designing and implementing effective anomaly detection pipelines using sparse Gaussian Graphical Models with various statistical distance metrics, and (b) differentiating these pipelines by embedding contextual semantics inferred from graphs so as to obtain better KPIs in practice. The motivation for the latter of these two has been explained above, and the former in particular is well motivated by the relatively mediocre performance of highly parametric deep learning methods for small tabular datasets (compared to images) such as IoT sensor data. In contrast to such traditional automated deep learning (AutoAI) techniques, our anomaly detection system is based on developing semantics-driven industry specific ML pipelines which perform scalable computation evaluating several models to identify the best model. We benchmark our AD method against state-of-the-art AD techniques on publicly available UCI datasets. We also conduct a case study on IoT sensor and semantic data procured from a large thermal energy asset to evaluate the importance of semantics in enhancing our pipelines. In addition, we also provide explainable insights for our model which provide a complete perspective to a reliability engineer.

Recent disasters have shown that hazards can be unpredictable and can have catastrophic consequences. Emergency plans are key to dealing with these situations and communications play a key role in emergency management. In this paper, we provide a formalism to design resilient emergency plans in terms of communications. We exemplify how to use the formalism using a case study of a Nuclear Emergency Plan.

The growing size of modern datasets necessitates splitting a large scale computation into smaller computations and operate in a distributed manner for improving overall performance. However, adversarial servers in a distributed computing system deliberately send erroneous data in order to affect the computation for their benefit. Computing Boolean functions is the key component of many applications of interest, e.g., classification problem, verification functions in the blockchain and the design of cryptographic algorithm. In this paper, we consider the problem of computing a Boolean function in which the computation is carried out distributively across several workers with particular focus on security against Byzantine workers. We note that any Boolean function can be modeled as a multivariate polynomial which can have high degree in general. Hence, the recently proposed Lagrange Coded Computing (LCC) can be used to simultaneously provide resiliency, security, and privacy. However, the security threshold (i.e., the maximum number of adversarial workers that can be tolerated) provided by LCC can be extremely low if the degree of the polynomial is high. Our goal is to design an efficient coding scheme which achieves the optimal security threshold. We propose two novel schemes called coded Algebraic normal form (ANF) and coded Disjunctive normal form (DNF). Instead of modeling the Boolean function as a general polynomial, the key idea of the proposed schemes is to model it as the concatenation of some linear functions and threshold functions. The proposed coded ANF and coded DNF outperform LCC by providing the security threshold which is independent of the polynomial's degree.

Machine learning can be misled by adversarial examples, which is formed by making small changes to the original data. Nowadays, there are kinds of methods to produce adversarial examples. However, they can not apply non-differentiable models, reduce the amount of calculations, and shorten the sample generation time at the same time. In this paper, we propose a new black box attack generating adversarial examples based on reinforcement learning. By using deep Q-learning network, we can train the substitute model and generate adversarial examples at the same time. Experimental results show that this method only needs 7.7ms to produce an adversarial example, which solves the problems of low efficiency, large amount of calculation and inapplicable to non-differentiable model.

The evolution of Service-Oriented Computing (SOC) provides more efficient software development methods for building and engineering new value-added service-based applications. SOC is a computing paradigm that relies on Web services as fundamental elements. Research and technical advancements in Web services composition have been considered as an effective opportunity to develop new service-based applications satisfying complex requirements rapidly and efficiently. In this paper, we present a novel approach enhancing the composition of semantic Web services. The novelty of our approach, as compared to others reported in the literature, rests on: i) mapping user's/organization's requirements with Business Process Modeling Notation (BPMN) and semantic descriptions using ontologies, ii) considering functional requirements and also different types of non-functional requirements, such as quality of service (QoS), quality of experience (QoE), and quality of business (QoBiz), iii) using Formal Concept Analysis (FCA) technique to select the optimal set of Web services, iv) considering composability levels between sequential Web services using Relational Concept Analysis (RCA) technique to decrease the required adaptation efforts, and finally, v) validating the obtained service-based applications by performing an analytical technique, which is the monitoring. The approach experimented on an extended version of the OWLS-TC dataset, which includes more than 10830 Web services descriptions from various domains. The obtained results demonstrate that our approach allows to successfully and effectively compose Web services satisfying different types of user's functional and non-functional requirements.

We address energy-efficient placement of data and analytics components of composite analytics services on a wireless network to minimize execution-time energy consumption (computation and communication) subject to compute, storage and network resource constraints. We introduce an expressive analytics service hypergraph model for representing k-ary composability relationships (k ≥ 2) between various analytics and data components and leverage binary quadratic programming (BQP) to minimize the total energy consumption of a given placement of the analytics hypergraph nodes on the network subject to resource availability constraints. Then, after defining a potential energy functional Φ(·) to model the affinities of analytics components and network resources using analogs of attractive and repulsive forces in physics, we propose a decentralized Metropolis Monte Carlo (MMC) sampling method which seeks to minimize Φ by moving analytics and data on the network. Although Φ is non-convex, using a potential game formulation, we identify conditions under which the algorithm provably converges to a local minimum energy equilibrium placement configuration. Trace-based simulations of the placement of a deep-neural-network analytics service on a realistic wireless network show that for smaller problem instances our MMC algorithm yields placements with total energy within a small factor of BQP and more balanced workload distributions; for larger problems, it yields low-energy configurations while the BQP approach fails.

The acceleration of extending popularity of Online Social Networks (OSNs) thanks to various services with which they provide people, is inevitable. This is why in OSNs security as a way to protect private data of users to be abused by unauthoritative people has a vital role to play. Trust evaluation is the security approach that has been utilized since the advent of OSNs. Graph-based approaches are among the most popular methods for trust evaluation. However, graph-based models need to employ limitations in the search process of finding trusted paths. This contributes to a reduction in trust accuracy. In this investigation, a learning-based model which with no limitation is able to find reliable users of any target user, is proposed. Experimental results depict 12% improvement in trust accuracy compares to models based on the graph-based approach.

The location information of a node is one of the essential attributes used in most underwater communication routing algorithms to identify a candidate forwarding node by any of the sources. The exact location information of a node exchanged with its neighbours' in plain text and the absence of node authentication results in some of the attacks such as Sybil attack, Blackhole attack, and Wormhole attack. Moreover, the severe consequence of these attacks is Denial of Service (DoS), poor network performance, reduced network lifetime, etc. This paper proposes an anti-Spoof (a-Spoof) algorithm for mitigating localization and neighbour spoofing attacks in UASN. a-Spoof uses three pre-shared symmetric keys to share the location. Additionally, location integrity provided through the hash function. Further, the performance of a-Spoof demonstrated through its implementation in UnetStack with reference to end-to-end packet delay and the number of hops.

The wide expansion of the Internet of Things is pushing the growth of vehicular ad-hoc networks (VANETs) into the Internet of Vehicles (IoV). Secure data communication is vital to the success and stability of the IoV and should be integrated into its various operations and aspects. In this paper, we present a framework for secure IoV communications by utilizing the High Performance Blockchain Consensus (HPBC) algorithm. Based on a previously published communication model for VANETs that uses an efficient routing protocol for transmitting packets between vehicles, we describe in this paper how to integrate a blockchain model on top of the IoV communications system. We illustrate the method that we used to implement HPBC within the IoV nodes. In order to prove the efficiency of the proposed model, we carry out extensive simulations that test the proposed model and study its overhead on the IoV network. The simulation results demonstrated the good performance of the HPBC algorithm when implemented within the IoV environment.

Application of the attention model hybrid building based on the artificial intelligence in the course of the intelligent architecture history is studied in this article. A Hadoop distributed architecture using big data processing technology which combines basic building information with the building energy consumption data for the data mining research methods, and conduct a preliminary design of a Hadoop-based public building energy consumption data mining system. The principles of the proposed model were summarized. At first, the intelligent firewall processes the decision data faster, when the harmful information invades. The intelligent firewall can monitor and also intercept the harmful information in a timelier manner. Secondly, develop a problem data processing plan, delete and identify different types of problem data, and supplement the deleted problem data according to the rules obtained by data mining. The experimental results have reflected the efficiency of the proposed model.

The burgeoning success of deep learning has raised the security and privacy concerns as more and more tasks are accompanied with sensitive data. Adversarial attacks in deep learning have emerged as one of the dominating security threats to a range of mission-critical deep learning systems and applications. This paper takes a holistic view to characterize the adversarial examples in deep learning by studying their adverse effect and presents an attack-independent countermeasure with three original contributions. First, we provide a general formulation of adversarial examples and elaborate on the basic principle for adversarial attack algorithm design. Then, we evaluate 15 adversarial attacks with a variety of evaluation metrics to study their adverse effects and costs. We further conduct three case studies to analyze the effectiveness of adversarial examples and to demonstrate their divergence across attack instances. We take advantage of the instance-level divergence of adversarial examples and propose strategic input transformation teaming defense. The proposed defense methodology is attack-independent and capable of auto-repairing and auto-verifying the prediction decision made on the adversarial input. We show that the strategic input transformation teaming defense can achieve high defense success rates and are more robust with high attack prevention success rates and low benign false-positive rates, compared to existing representative defense methods.

In this paper, a new intrusion detection mechanism is proposed based on Trust and Packet Loss Rate at Sensor Node in WSNs. To find the true malicious nodes, the proposed mechanism performs a deep analysis on the packet loss. Two independent metrics such as buffer capacity metric and residual energy metric are considered for packet loss rate evaluation. Further, the trust evaluation also considers the basic communication interactions between sensor nodes. Based on these three metrics, a new composite metric called Packet Forwarding Probability (PFP) is derived through which the malicious nodes are identified. Simulation experiments are conducted over the proposed mechanism and the performance is evaluated through False Positive Rate (FPR) and Malicious Detection Rate (MDR). The results declare that the proposed mechanism achieves a better performance compared to the conventional approaches.

Battery energy storage systems are facing risks of unreliable battery sensor data which might be caused by sensor faults in an embedded battery management system, communication failures, and even cyber-attacks. It is crucial to evaluate the trustworthiness of battery sensor data since inaccurate sensor data could lead to not only serious damages to battery energy storage systems, but also threaten the overall reliability of their applications (e.g., electric vehicles or power grids). This paper introduces a battery sensor data trust framework enabling detecting unreliable data using a deep learning algorithm. The proposed sensor data trust mechanism could potentially improve safety and reliability of the battery energy storage systems. The proposed deep learning-based battery sensor fault detection algorithm is validated by simulation studies using a convolutional neural network.

The trust of a robot in its human partner is a significant issue in human-robot interaction, which is seldom explored in the field of robotics. This study addresses a critical issue of robots' trust in humans during the human-robot collaboration process based on the data of human motions, past interactions of the human-robot pair, and the human's current performance in the co-carry task. The trust level is evaluated dynamically throughout the collaborative task that allows the trust level to change if the human performs false positive actions, which can help the robot avoid making unpredictable movements and causing injury to the human. Experimental results showed that the robot effectively assisted the human in collaborative tasks through the proposed computational trust model.

In today's technological climate, users require fast automation and digitization of results for large amounts of data at record speeds. Especially in the field of medicine, where each patient is often asked to undergo many different examinations within one diagnosis or treatment. Each examination can help in the diagnosis or prediction of further disease progression. Furthermore, all produced data from these examinations must be stored somewhere and available to various medical practitioners for analysis who may be in geographically diverse locations. The current medical climate leans towards remote patient monitoring and AI-assisted diagnosis. To make this possible, medical data should ideally be secured and made accessible to many medical practitioners, which makes them prone to malicious entities. Medical information has inherent value to malicious entities due to its privacy-sensitive nature in a variety of ways. Furthermore, if access to data is distributively made available to AI algorithms (particularly neural networks) for further analysis/diagnosis, the danger to the data may increase (e.g., model poisoning with fake data introduction). In this paper, we propose a federated learning approach that uses decentralized learning with blockchain-based security and a proposition that accompanies that training intelligent systems using distributed and locally-stored data for the use of all patients. Our work in progress hopes to contribute to the latest trend of the Internet of Medical Things security and privacy.

Anomaly detection is critically important for intelligent surveillance systems to detect in a timely manner any malicious activities. Many video anomaly detection approaches using deep learning methods focus on a single camera video stream with a fixed scenario. These deep learning methods use large-scale training data with large complexity. As a solution, in this paper, we show how to use pre-trained convolutional neural net models to perform feature extraction and context mining, and then use denoising autoencoder with relatively low model complexity to provide efficient and accurate surveillance anomaly detection, which can be useful for the resource-constrained devices such as edge devices of the Internet of Things (IoT). Our anomaly detection model makes decisions based on the high-level features derived from the selected embedded computer vision models such as object classification and object detection. Additionally, we derive contextual properties from the high-level features to further improve the performance of our video anomaly detection method. We use two UCSD datasets to demonstrate that our approach with relatively low model complexity can achieve comparable performance compared to the state-of-the-art approaches.

We propose a new approach to creating a subject-oriented distributed computing environment. Such an environment is used to support decision-making in solving relevant problems of ensuring energy systems resilience. The proposed approach is based on the idea of advancing and integrating the following important capabilities in supercomputer engineering: continuous integration, delivery, and deployment of the system and applied software, high-performance computing in heterogeneous environments, multi-agent intelligent computation planning and resource allocation, big data processing and geo-information servicing for subject information, including weakly structured data, and decision-making support. This combination of capabilities and their advancing are unique to the subject domain under consideration, which is related to combinatorial studying critical objects of energy systems. Evaluation of decision-making alternatives is carrying out through applying combinatorial modeling and multi-criteria selection rules. The Orlando Tools framework is used as the basis for an integrated software environment. It implements a flexible modular approach to the development of scientific applications (distributed applied software packages).

In STOC 1988, Ben-Or, Goldwasser, and Wigderson (BGW) established an important milestone in the fields of cryptography and distributed computing by showing that every functionality can be computed with perfect (information-theoretic and error-free) security at the presence of an active (aka Byzantine) rushing adversary that controls up to n/3 of the parties. We study the round complexity of general secure multiparty computation in the BGW model. Our main result shows that every functionality can be realized in only four rounds of interaction, and that some functionalities cannot be computed in three rounds. This completely settles the round-complexity of perfect actively-secure optimally-resilient MPC, resolving a long line of research. Our lower-bound is based on a novel round-reduction technique that allows us to lift existing three-round lower-bounds for verifiable secret sharing to four-round lower-bounds for general MPC. To prove the upper-bound, we develop new round-efficient protocols for computing degree-2 functionalities over large fields, and establish the completeness of such functionalities. The latter result extends the recent completeness theorem of Applebaum, Brakerski and Tsabary (TCC 2018, Eurocrypt 2019) that was limited to the binary field.

FPGAs are becoming a common sight in cloud environments and new usage paradigms, such as FPGA-as-a-Service, have emerged. This development poses a challenge to traditional FPGA security models, as these are assuming trust between the user and the hardware owner. Currently, the user cannot keep bitstream nor data protected from the hardware owner in an FPGA-as-a-service setting. This paper proposes a security model where the chip manufacturer takes the role of root-of-trust to remedy these security problems. We suggest that the chip manufacturer creates a Public Key Infrastructure (PKI), used for user bitstream protection and data encryption, on each device. The chip manufacturer, rather than the hardware owner, also controls certain security-related peripherals. This allows the user to take control over a predefined part of the programmable logic and set up a protected enclave area. Hence, all user data can be provided in encrypted form and only be revealed inside the enclave area. In addition, our model enables secure and concurrent multi-tenant usage of remote FPGAs. To also consider the needs of the hardware owner, our solution includes bitstream certification and affirming that uploaded bitstreams have been vetted against maliciousness.

In the face of increasingly serious firmware attacks, it is of great significance to analyze the vulnerability security of UEFI. This paper first introduces the commonly used trusted authentication mechanisms of UEFI. Then, aiming at the loopholes in the process of UEFI trust verification in the startup phase, combined with the state transition diagram, PageRank algorithm and Bayesian network theory, the analysis model of UEFI trust verification startup vulnerability is constructed. And according to the example to verify the analysis. Through the verification and analysis of the data obtained, the vulnerable attack paths and key vulnerable nodes are found. Finally, according to the analysis results, security enhancement measures for UEFI are proposed.

Coreference resolution is an important task in the field of natural language processing. Most existing methods usually utilize word-level representations, ignoring massive information from the texts. To address this issue, we investigate how to improve Chinese coreference resolution by using span-level semantic representations. Specifically, we propose a model which acquires word and character representations through pre-trained Skip-Gram embeddings and pre-trained BERT, then explicitly leverages span-level information by performing bidirectional LSTMs among above representations. Experiments on CoNLL-2012 shared task have demonstrated that the proposed model achieves 62.95% F1-score, outperforming our baseline methods.

IoT devices provide a rich data source that is not available in the past, which is valuable for a wide range of intelligence applications, especially deep neural network (DNN) applications that are data-thirsty. An established DNN model provides useful analysis results that can improve the operation of IoT systems in turn. The progress in distributed/federated DNN training further unleashes the potential of integration of IoT and intelligence applications. When a large number of IoT devices are deployed in different physical locations, distributed training allows training modules to be deployed to multiple edge data centers that are close to the IoT devices to reduce the latency and movement of large amounts of data. In practice, these IoT devices and edge data centers are usually owned and managed by different parties, who do not fully trust each other or have conflicting interests. It is hard to coordinate them to provide end-to-end integrity protection of the DNN construction and application with classical security enhancement tools. For example, one party may share an incomplete data set with others, or contribute a modified sub DNN model to manipulate the aggregated model and affect the decision-making process. To mitigate this risk, we propose a novel blockchain based end-to-end integrity protection scheme for DNN applications integrated with an IoT system in the edge computing environment. The protection system leverages a set of cryptography primitives to build a blockchain adapted for edge computing that is scalable to handle a large number of IoT devices. The customized blockchain is integrated with a distributed/federated DNN to offer integrity and authenticity protection services.

With the introduction of new technologies such as cloud computing and big data, the security issues of industrial cyber-physical systems (ICPSs) have become more complicated. Meanwhile, a lot of current security research lacks adaptation to industrial system upgrades. In this paper, an edge-cloud synergy framework for security decision-making is proposed, which takes advantage of the huge convenience and advantages brought by cloud computing and edge computing, and can make security decisions on a global perspective. Under this framework, a combination of Bayesian network-based risk assessment and stochastic game model-based security decision-making is proposed to generate an optimal defense strategy to minimize system losses. This method trains models in the clouds and infers at the edge computing nodes to achieve rapid defense strategy generation. Finally, a case study on the hardware-in-the-loop simulation platform proves the feasibility of the approach.

Human action is naturally compositional: humans can easily recognize and perform actions with objects that are different from those used in training demonstrations. In this paper, we study the compositionality of action by looking into the dynamics of subject-object interactions. We propose a novel model which can explicitly reason about the geometric relations between constituent objects and an agent performing an action. To train our model, we collect dense object box annotations on the Something-Something dataset. We propose a novel compositional action recognition task where the training combinations of verbs and nouns do not overlap with the test set. The novel aspects of our model are applicable to activities with prominent object interaction dynamics and to objects which can be tracked using state-of-the-art approaches; for activities without clearly defined spatial object-agent interactions, we rely on baseline scene-level spatio-temporal representations. We show the effectiveness of our approach not only on the proposed compositional action recognition task but also in a few-shot compositional setting which requires the model to generalize across both object appearance and action category.