Biblio

A randomly-selective encryption (RSE) algorithm is proposed for HEVC video bitstream in this paper. It is a pioneer algorithm with high efficiency and security. The encryption process is completely independent of video compression process. A randomly-selective sequence (RSS) based on the RC4 algorithm is designed to determine the extraction position in the video bitstream. The extracted bytes are encrypted by AES-CTR to obtain the encrypted video. Based on the high efficiency video coding (HEV C) bitstream, the simulation and analysis results show that the proposed RSE algorithm has low time complexity and high security, which is a promising tool for video cryptographic applications.

Data poisoning is a type of adversarial attack on training data where an attacker manipulates a fraction of data to degrade the performance of machine learning model. There are several known defensive mechanisms for handling offline attacks, however defensive measures for online learning, where data points arrive sequentially, have not garnered similar interest. In this work, we propose a defense mechanism to minimize the degradation caused by the poisoned training data on a learner's model in an online setup. Our proposed method utilizes an influence function which is a classic technique in robust statistics. Further, we supplement it with the existing data sanitization methods for filtering out some of the poisoned data points. We study the effectiveness of our defense mechanism on multiple datasets and across multiple attack strategies against an online learner.

With the advent of massive machine type of communications, security protection becomes more important than ever. Efforts have been made to impose security protection capability to physical-layer signal design, so called physical-layer security (PLS). The purpose of this paper is to evaluate the performance of PLS schemes for a multi-input-multi-output (MIMO) systems with space-time block coding (STBC) under imperfect channel estimation. Three PLS schemes for STBC schemes are modeled and their bit error rate (BER) performances are evaluated under various channel estimation error environments, and their performance characteristics are analyzed.

Co-salient object detection (CoSOD) has recently achieved significant progress and played a key role in retrieval-related tasks. However, it inevitably poses an entirely new safety and security issue, i.e., highly personal and sensitive content can potentially be extracting by powerful CoSOD methods. In this paper, we address this problem from the perspective of adversarial attacks and identify a novel task: adversarial co-saliency attack. Specially, given an image selected from a group of images containing some common and salient objects, we aim to generate an adversarial version that can mislead CoSOD methods to predict incorrect co-salient regions. Note that, compared with general white-box adversarial attacks for classification, this new task faces two additional challenges: (1) low success rate due to the diverse appearance of images in the group; (2) low transferability across CoSOD methods due to the considerable difference between CoSOD pipelines. To address these challenges, we propose the very first blackbox joint adversarial exposure and noise attack (Jadena), where we jointly and locally tune the exposure and additive perturbations of the image according to a newly designed high-feature-level contrast-sensitive loss function. Our method, without any information on the state-of-the-art CoSOD methods, leads to significant performance degradation on various co-saliency detection datasets and makes the co-salient objects undetectable. This can have strong practical benefits in properly securing the large number of personal photos currently shared on the Internet. Moreover, our method is potential to be utilized as a metric for evaluating the robustness of CoSOD methods.

VCB is an important component to ensure the safe and smooth operation of the power system. As an important driving part of the vacuum circuit breaker, the operating mechanism is prone to mechanical failure, which leads to power grid accidents. This paper offers an in-depth analysis of the mechanical faults of the operating mechanism of vacuum circuit breaker and their causes, extracts the current signal of the opening and closing coil strongly correlated with the mechanical faults of the operating mechanism as the characteristic information to build a Deep Belief Network (DBN) model, trains each data set via Restricted Boltzmann Machine(RBM) and updates the model parameters. The number of hidden layer nodes, the structure of the network layer, and the learning rate are determined, and the mechanical fault diagnosis system of vacuum circuit breaker based on the Deep Belief Network is established. The results show that when the network structure is 8-110-110-6 and the learning rate is 0.01, the recognition accuracy of the DBN model is the highest, which is 0.990871. Compared with BP neural network, DBN has a smaller cross-entropy error and higher accuracy. This method can accurately diagnose the mechanical fault of the vacuum circuit breaker, which lays a foundation for the smooth operation of the power system.

Swarm learning (SL) is an emerging promising decentralized machine learning paradigm and has achieved high performance in clinical applications. SL solves the problem of a central structure in federated learning by combining edge computing and blockchain-based peer-to-peer network. While there are promising results in the assumption of the independent and identically distributed (IID) data across participants, SL suffers from performance degradation as the degree of the non-IID data increases. To address this problem, we propose a generative augmentation framework in swarm learning called SL-GAN, which augments the non-IID data by generating the synthetic data from participants. SL-GAN trains generators and discriminators locally, and periodically aggregation via a randomly elected coordinator in SL network. Under the standard assumptions, we theoretically prove the convergence of SL-GAN using stochastic approximations. Experimental results demonstrate that SL-GAN outperforms state-of-art methods on three real world clinical datasets including Tuberculosis, Leukemia, COVID-19.

An approach to substantiating the choice of a discipline for the maintenance of a redundant computer system, with the possible use of node resources saved after failures, is considered. The choice is aimed at improving the reliability and profitability of the system, taking into account the operational costs of restoring nodes. Models of reliability of systems with service disciplines are proposed, providing both the possibility of immediate recovery of nodes after failures, and allowing degradation of the system when using node resources stored after failures in it. The models take into account the conditions of the admissibility or inadmissibility of the loss of information accumulated during the operation of the system. The operating costs are determined, taking into account the costs of restoring nodes for the system maintenance disciplines under consideration

With the future 6G era, spiking neural networks (SNNs) can be powerful processing tools in various areas due to their strong artificial intelligence (AI) processing capabilities, such as biometric recognition, AI robotics, autonomous drive, and healthcare. However, within Cyber Physical System (CPS), SNNs are surprisingly vulnerable to adversarial examples generated by benign samples with human-imperceptible noise, this will lead to serious consequences such as face recognition anomalies, autonomous drive-out of control, and wrong medical diagnosis. Only by fully understanding the principles of adversarial attacks with adversarial samples can we defend against them. Nowadays, most existing adversarial attacks result in a severe accuracy degradation to trained SNNs. Still, the critical issue is that they only generate adversarial samples by randomly adding, deleting, and flipping spike trains, making them easy to identify by filters, even by human eyes. Besides, the attack performance and speed also can be improved further. Hence, Spike Probabilistic Attack (SPA) is presented in this paper and aims to generate adversarial samples with more minor perturbations, greater model accuracy degradation, and faster iteration. SPA uses Poisson coding to generate spikes as probabilities, directly converting input data into spikes for faster speed and generating uniformly distributed perturbation for better attack performance. Moreover, an objective function is constructed for minor perturbations and keeping attack success rate, which speeds up the convergence by adjusting parameters. Both white-box and black-box settings are conducted to evaluate the merits of SPA. Experimental results show the model's accuracy under white-box attack decreases by 9.2S% 31.1S% better than others, and average success rates are 74.87% under the black-box setting. The experimental results indicate that SPA has better attack performance than other existing attacks in the white-box and better transferability performance in the black-box setting,

Anomalous behaviour in subsystems of complex machines often affect overall performance even without failures. We devise unsupervised methods to detect times with degraded performance, and localize correlated signals, evaluated on a system with over 4000 monitored signals. From incidents comprising both downtimes and degraded performance, our approach localizes relevant signals within 1.2% of the parameter space.

Machine learning (ML) has been applied in prognostics and health management (PHM) to monitor and predict the health of industrial machinery. The use of PHM in production systems creates a cyber-physical, omni-layer system. While ML offers statistical improvements over previous methods, and brings statistical models to bear on new systems and PHM tasks, it is susceptible to performance degradation when the behavior of the systems that ML is receiving its inputs from changes. Natural changes such as physical wear and engineered changes such as maintenance and rebuild procedures are catalysts for performance degradation, and are both inherent to production systems. Drawing from data on the impact of maintenance procedures on ML performance in hydraulic actuators, this paper presents a simulation study that investigates how long it takes for ML performance degradation to create a difference in the throughput of serial production system. In particular, this investigation considers the performance of an ML model learned on data collected before a rebuild procedure is conducted on a hydraulic actuator and an ML model transfer learned on data collected after the rebuild procedure. Transfer learning is able to mitigate performance degradation, but there is still a significant impact on throughput. The conclusion is drawn that ML faults can have drastic, non-linear effects on the throughput of production systems.

Deep neural networks (DNNs) are known for their inability to utilize underlying hardware resources due to hard-ware susceptibility to sparse activations and weights. Even in finer granularities, many of the non-zero values hold a portion of zero-valued bits that may cause inefficiencies when executed on hard-ware. Inspired by conventional CPU simultaneous multithreading (SMT) that increases computer resource utilization by sharing them across several threads, we propose non-blocking SMT (NB-SMT) designated for DNN accelerators. Like conventional SMT, NB-SMT shares hardware resources among several execution flows. Yet, unlike SMT, NB-SMT is non-blocking, as it handles structural hazards by exploiting the algorithmic resiliency of DNNs. Instead of opportunistically dispatching instructions while they wait in a reservation station for available hardware, NB-SMT temporarily reduces the computation precision to accommodate all threads at once, enabling a non-blocking operation. We demonstrate NB-SMT applicability using SySMT, an NB-SMT-enabled output-stationary systolic array (OS-SA). Compared with a conventional OS-SA, a 2-threaded SySMT consumes 1.4× the area and delivers 2× speedup with 33% energy savings and less than 1% accuracy degradation of state-of-the-art CNNs with ImageNet. A 4-threaded SySMT consumes 2.5× the area and delivers, for example, 3.4× speedup and 39%×energy savings with 1% accuracy degradation of 40%-pruned ResNet-18.

Using Generative Adversarial Networks (GAN) to translate images is a significant field in computer vision. There are partial distortion, artifacts and detail loss in the images generated by current image translation algorithms. In order to solve this problem, this paper adds attention-based residual neural network to the generator of GAN. Attention-based residual neural network can improve the representation ability of the generator by weighting the channels of the feature map. Experiment results on the Facades dataset show that Attention Residual GAN can translate images with excellent quality.

With the development of wireless sensor networks (WSNs), WSNs have been widely used in various fields such as animal habitat detection, military surveillance, etc. This paper focuses on protecting the source location privacy (SLP) in WSNs. Existing algorithms perform poorly in non-uniform networks which are common in reality. In order to address the performance degradation problem of existing algorithms in non-uniform networks, this paper proposes a robust fixed path-based random routing scheme (RFRR), which guarantees the path diversity with certainty in non-uniform networks. In RFRR, the data packets are sent by selecting a routing path that is highly differentiated from each other, which effectively protects SLP and resists the backtracking attack. The experimental results show that RFRR increases the difficulty of the backtracking attack while safekeeping the balance between security and energy consumption.

Kotlin is a relatively new programming language from JetBrains: its development started in 2010 with release 1.0 done in early 2016. The Kotlin compiler, while slowly and steadily becoming more and more mature, still crashes from time to time on the more tricky input programs, not least because of the complexity of its features and their interactions. This makes it a great target for fuzzing, even the basic forms of which can find a significant number of Kotlin compiler crashes. There is a problem with fuzzing, however, closely related to the cause of the crashes: generating a random, non-trivial and semantically valid Kotlin program is hard. In this paper, we talk about type-centriccompilerfuzzing in the form of type-centricenumeration, an approach inspired by skeletal program enumeration [1] and based on a combination of generative and mutation-based fuzzing, which solves this problem by focusing on program types. After creating the skeleton program, we fill the typed holes with fragments of suitable type, created via generation and enhanced by semantic-aware mutation. We implemented this approach in our Kotlin compiler fuzzing framework called Backend Bug Finder (BBF) and did an extensive evaluation, not only testing the real-world feasibility of our approach, but also comparing it to other compiler fuzzing techniques. The results show our approach to be significantly better compared to other fuzzing approaches at generating semantically valid Kotlin programs, while creating more interesting crash-inducing inputs at the same time. We managed to find more than 50 previously unknown compiler crashes, of which 18 were considered important after their triage by the compiler team.

A physical layer security (PLS) technique called Gray Code Hopping (GCH) is presented offering simplistic implementation and no bit error rate (BER) performance degradation over the main channel. A synchronized transmitter and receiver "hop" to an alternative binary reflected Gray code (BRGC) mapping of bits to symbols between each consecutive modulation symbol. Monte Carlo simulations show improved BER performance over a similar technique from the literature. Simulations also confirm compatibility of GCH with either hard or soft decision decoding methods. Simplicity of GCH allows for ready implementation in adaptive 5th Generation New Radio (5G NR) modulation coding schemes.

Memristive nano crossbar array has paved the way for high density memories but in a very low power environment. But such high density circuits face multiple problems at the time of implementation. The sneak path problem in crossbar array is one such problem which causes difficulty in distinguishing the logical states of the memristors. On the other hand, hardware Trojan causes malfunctioning of the circuit or performance degradation. If any of these are present in the nano crossbar, it is difficult to identify whether the performance degradation is due to the sneak path problem or due to that of Hardware Trojan.This paper makes a comparative study of the sneak path problem and the hardware Trojan to understand the performance difference between both. It is observed that some parameters are affected by sneak path problem but remains unaffected in presence of Hardware Trojan and vice versa. Analyzing these parameters, we can classify whether the performance degradation is due to sneak path or due to Hardware Trojan. The experimental results well establish the proposed methods of detection of hardware Trojan in presence of sneak path in memristive nano crossbar circuits.

In recent years, Deep Learning (DL) has been utilized for cyber-attack detection mechanisms as it offers highly accurate detection and is able to overcome the limitations of standard machine learning techniques. When applied in a Software-Defined Network (SDN) environment, a DL-based detection mechanism shows satisfying detection performance. However, in the case of adversarial attacks, the detection performance deteriorates. Therefore, in this paper, first, we outline a highly accurate flooding DDoS attack detection framework based on DL for SDN environments. Second, we investigate the performance degradation of our detection framework when being tested with two adversary traffic datasets. Finally, we evaluate three adversarial training procedures for improving the detection performance of our framework concerning adversarial attacks. It is shown that the application of one of the adversarial training procedures can avoid detection performance degradation and thus might be used in a real-time detection system based on continual learning.

Scientific computing sometimes involves computation on sensitive data. Depending on the data and the execution environment, the HPC (high-performance computing) user or data provider may require confidentiality and/or integrity guarantees. To study the applicability of hardware-based trusted execution environments (TEEs) to enable secure scientific computing, we deeply analyze the performance impact of general purpose TEEs, AMD SEV, and Intel SGX, for diverse HPC benchmarks including traditional scientific computing, machine learning, graph analytics, and emerging scientific computing workloads. We observe three main findings: 1) SEV requires careful memory placement on large scale NUMA machines (1×-3.4× slowdown without and 1×-1.15× slowdown with NUMA aware placement), 2) virtualization-a prerequisite for SEV- results in performance degradation for workloads with irregular memory accesses and large working sets (1×-4× slowdown compared to native execution for graph applications) and 3) SGX is inappropriate for HPC given its limited secure memory size and inflexible programming model (1.2×-126× slowdown over unsecure execution). Finally, we discuss forthcoming new TEE designs and their potential impact on scientific computing.

Physical layer authentication (PLA) has recently been discussed in the context of URLLC due to its low complexity and low overhead. Nevertheless, these schemes also introduce additional sources of error through missed detections and false alarms. The trade-offs of these characteristics are strongly dependent on the deployment scenario as well as the processing architecture. Thus, considering a feature-based PLA scheme utilizing channel-state information at multiple distributed radio-heads, we study these trade-offs analytically. We model and analyze different scenarios of centralized and decentralized decision-making and decoding, as well as the impacts of a single-antenna attacker launching a Sybil attack. Based on stochastic network calculus, we provide worst-case performance bounds on the system-level delay for the considered distributed scenarios under a Sybil attack. Results show that the arrival-rate capacity for a given latency deadline is increased for the distributed scenarios. For a clustered sensor deployment, we find that the distributed approach provides 23% higher capacity when compared to the centralized scenario.

In emergency situations like recent Australian bushfires, it is crucial for civilians and firefighters to receive critical information such as escape routes and safe sheltering points with guarantees on information quality attributes. Mobile Ad-hoc Networks (MANETs) can provide communications in bushfire when fixed infrastructure is destroyed and not available. Current MANET solutions, however, are mostly tested under static bushfire scenario. In this work, we investigate the impact of a realistic dynamic bushfire in a dry eucalypt forest with a shrubby understory, on the performance of data delivery solutions in a MANET. Simulation results show a significant degradation in the performance of state-of-the-art MANET quality of information solution. Other than frequent source handovers and reduced user usability, packet arrival latency increases by more than double in the 1st quartile with a median drop of 74.5 % in the overall packet delivery ratio. It is therefore crucial for MANET solutions to be thoroughly evaluated under realistic dynamic bushfire scenarios.

A human-swarm cooperative system, which mixes multiple robots and a human supervisor to form a mission team, has been widely used for emergent scenarios such as criminal tracking and victim assistance. These scenarios are related to human safety and require a robot team to quickly transit from the current undergoing task into the new emergent task. This sudden mission change brings difficulty in robot motion adjustment and increases the risk of performance degradation of the swarm. Trust in human-human collaboration reflects a general expectation of the collaboration; based on the trust humans mutually adjust their behaviors for better teamwork. Inspired by this, in this research, a trust-aware reflective control (Trust-R), was developed for a robot swarm to understand the collaborative mission and calibrate its motions accordingly for better emergency response. Typical emergent tasks “transit between area inspection tasks”, “response to emergent target - car accident” in social security with eight fault-related situations were designed to simulate robot deployments. A human user study with 50 volunteers was conducted to model trust and assess swarm performance. Trust-R's effectiveness in supporting a robot team for emergency response was validated by improved task performance and increased trust scores.

Two-Dimensional barcodes are used as data authentication storage tool on several cryptographic architectures. This article describes a novel meaningful image authentication method for data validation using the Meaningless Reversible Degradation concept and QR Codes. The system architecture use the Meaningless Reversible Degradation algorithm, systematic Reed-Solomon error correction codes, meaningful images, and QR Codes. The encoded images are the secret key for visual validation. The proposed work encodes any secret image file up to 3.892 Bytes and is decoded using data stored in a QR Code and a digital file retrieved through a wireless connection on a mobile device. The QR Code carries partially distorted and stream ciphered bits. The QR Code version is defined in conformity with the secret image file size. Once the QR Code data is decoded, the authenticating party retrieves a previous created Reed-Solomon redundancy file to correct the QR Code stored data. Finally, the secret image is decoded for user visual identification. A regular QR Code reader cannot decode any meaningful information when the QR Code is scanned. The presented cryptosystem improves the redundancy download file size up to 50% compared to a plaintext image transmission.

Lightweight virtualization represented by container technology provides a virtual environment for cloud services with more flexibility and efficiency due to the kernel-sharing property. However, the shared kernel also means that the system isolation mechanisms are incomplete. Attackers can scan the shared system configuration files to explore vulnerabilities for launching attacks. Previous works mainly eliminate the problem by fixing operating systems or using access control policies, but these methods require significant modifications and cannot meet the security needs of individual containers accurately. In this paper, we present ConfigRand, a moving target defense framework to prevent the information leakages due to the shared kernel in the container-based cloud. The ConfigRand deploys deceptive system configurations for each container, bounding the scan of attackers aimed at the shared kernel. In design of ConfigRand, we (1) propose a framework applying the moving target defense philosophy to periodically generate, distribute, and deploy the deceptive system configurations in the container-based cloud; (2) establish a model to formalize these configurations and quantify their heterogeneity; (3) present a configuration movement strategy to evaluate and optimize the variation of configurations. The results show that ConfigRand can effectively prevent the information leakages due to the shared kernel and apply to typical container applications with minimal system modification and performance degradation.

Moving Target Defense (MTD) has been emerged as a promising countermeasure to defend systems against cyberattacks asymmetrically while working well with legacy security and defense mechanisms. MTD provides proactive security services by dynamically altering attack surfaces and increasing attack cost or complexity to prevent further escalation of the attack. However, one of the non-trivial hurdles in deploying MTD techniques is how to handle potential performance degradation (e.g., interruptions of service availability) and maintain acceptable quality-of-service (QoS) in an MTD-enabled system. In this paper, we derive the service performance metrics (e.g., an extent of failed jobs) to measure how much performance degradation is introduced due to MTD operations, and propose QoS-aware service strategies (i.e., drop and wait) to manage ongoing jobs with the minimum performance degradation even under MTD operations running. We evaluate the service performance of software-defined networking (SDN)-based web services (i.e., Apache web servers). Our experimental results prove that the MTD-enabled system can minimize performance degradation by using the proposed job management strategies. The proposed strategies aim to optimize a specific service configuration (e.g., types of jobs and request rates) and effectively minimize the adverse impact of deploying MTD in the system with acceptable QoS while retaining the security effect of IP shuffling-based MTD.

The complexity and ubiquity of modern computing systems is a fertile ground for anomalies, including security and privacy breaches. In this paper, we propose a new methodology that addresses the practical challenges to implement anomaly detection approaches. Specifically, it is challenging to define normal behavior comprehensively and to acquire data on anomalies in diverse cloud environments. To tackle those challenges, we focus on anomaly detection approaches based on system performance signatures. In particular, performance signatures have the potential of detecting zero-day attacks, as those approaches are based on detecting performance deviations and do not require detailed knowledge of attack history. The proposed methodology leverages an analytical performance model and experimentation, and allows to control the rate of false positives in a principled manner. The methodology is evaluated using the TPCx-V workload, which was profiled during a set of executions using resource exhaustion anomalies that emulate the effects of anomalies affecting system performance. The proposed approach was able to successfully detect the anomalies, with a low number of false positives (precision 90%-98%).