Search

Peer-to-peer (P2P) energy trading is one of the promising approaches for implementing decentralized electricity market paradigms. In the P2P trading, each actor negotiates directly with a set of trading partners. Since the physical network or grid is used for energy transfer, power losses are inevitable, and grid-related costs always occur during the P2P trading. A proper market clearing mechanism is required for the P2P energy trading between different producers and consumers. This paper proposes a decentralized market clearing mechanism for the P2P energy trading considering the privacy of the agents, power losses as well as the utilization fees for using the third party owned network. Grid-related costs in the P2P energy trading are considered by calculating the network utilization fees using an electrical distance approach. The simulation results are presented to verify the effectiveness of the proposed decentralized approach for market clearing in P2P energy trading. Agency for Science, Technology and Research (A*STAR) Accepted version This work was supported by the Agency for Science, Technology and Research (A*STAR) through its Singapore–Germany Academic-Industry (2+2) International Collaboration under Grant A1990b0060. Paper no. TSG-01805-2019.

If a disaster occurs or is predicted, the Korean government and local governments will deliver disaster information to the people based on the “Disaster and Safety Management Basic Law”[1]. Disaster information transmission media currently used in Korea are classified into mobile communication networks, broadcasting networks, civil defense alerts, and disaster warning and alerting systems. This paper deals with the implementation of an integrated system of disaster warning and alerting and managing disaster information in various types of disaster warning and alerting systems operated and managed by local governments. The proposed system utilizes the Common Alerting Protocol(CAP) established by the ITU-T standard[2] to integrate and issue and manage disaster information in the existing disaster forecast and warning system. In this paper, the CAP-based profile was implemented by modifying it to consider the environment of the Korean warning and alerting system. Based on the TTA standard, this CAP profile defined the method of using the parameter defined as an option according to the characteristics of each disaster. This paper includes interworking methods and test results to directly connect with the automatic voice notification system, the village broadcasting system, and the bus information system(BIS) among the existing disaster warning and alerting systems.

With the development of personal broadcasting such as YouTube, the demand for editing the video he filmed himself is steadily increasing. Traditional media video editing solutions help edit the results of the filming. However, there is a problem that requires a lot of editing time, as people usually edit videos that are from tens to hundreds of times longer than the final result of a personal broadcasting after filming. To overcome this problem, this paper automatically classifies images with specific scenes in the whole media image editing process, and secondly proposes automatic media editing solution technology in which people intervene. In particular, personal broadcasting focuses on the use of images that include characters, specific objects, and cue sign gestures among the entire. While the existing deep learning techniques such as faces, objects and gestures are advanced, integrated recognition technologies that simultaneously deal with special requirements for editing videos are still in the early stages of research. In this paper, the automatic composite recognition technology for editing video based on deep learning is proposed. The proposed technology was implemented with python and tensorflow software based on edge computing equipment. Using actual youtube videos, it took 0.1 second to process five-person recognition, 63-food recognition, or cue sign recognition using clapping or V poses at the same time. The recognized results are divided into timestamps of the entire movie, recognition results, and locations of objects on the screen, and are output to the json file. In addition, this solution was developed on an edge computing in order to increase real-time reliability. We expect to provide automatic video editing based on perceived json results as well as shorter editing times based on this implementation.

Cyber-physical Systems can be defined as a complex networked control system, which normally develop by combining several physical components with the cyber space. Cyber Physical System are already a part of our daily life. As its already being a part of everyone life, CPS also have great potential security threats and can be vulnerable to various cyber-attacks without showing any sign directly to component failure. To protect user security and privacy is a fundamental concern of any kind of system; either it's a simple web application or supplicated professional system. Digital Multifactor authentication is one of the best ways to make secure authentication. It covers many different areas of a Cyber-connected world, including online payments, communications, access right management, etc. Most of the time, Multifactor authentication is little complex as it requires extra step from users. This paper will discuss the evolution from single authentication to Multi-Factor Authentication (MFA) starting from Single-Factor Authentication (SFA) and through Two-Factor Authentication (2FA). This paper seeks to analyze and evaluate the most prominent authentication techniques based on accuracy, cost, and feasibility of implementation. We also suggest several authentication schemes which incorporate with Multifactor authentication for CPS.

Orthogonal defect classification (ODC) is a multi-dimensional measurement system with both qualitative and quantitative characteristics. And it is currently widely used in the software industry. However, its high level of abstraction leads to limited semantic information. Therefore, it seems to have a limited role in the process of software engineering of software-intensive systems (SISs). To solve this problem, this paper first analyzes software error lifetime from the perspective of knowledge-based software engineering and proposes an error generation model. Then, the paper proposes the concepts of software error pattern (SEP) and software requirements error pattern (SREP) based on the ODC. Then, according to an error generation mechanism, four types of software-hardware integrated error pattern (SHIEP) in the requirement stage, which is a sub-category of SREP, and corresponding ontology representation are given, focusing on “scenario”, “error manifestation” and “solution”. Finally, this paper takes a certain type of airborne radar software system as an example, uses protege to edit the SHIEPs and instances, and further introduces the application of software FMEA based on the above work. The results show that the prior information based on the SHIEPs is helpful to discover potential failures and failure modes that may adversely affect the function or performance of SISs. Therefore, the proposed SHIEP is of great significance for improving the quality of software development and verification.

Online Social Network (OSN) services are very popular nowadays. In order to protect the data privacy, Decentralized Online Social Network (DOSN) services have been proposed. In DOSN, the data published by a user and the data replicas are only stored in the friend circle of the user. Although full replication can improve Data Availability (DA), pure DOSNs may not be able to deliver sustainable data availability. This paper proposes a Cloud-Assisted scheme, called Cadros, to improve the DA in DOSN. This paper conducts quantitative analysis about the storage capacity of Cadros as the result of integrating the Cloud into DOSN, and further models and predicts the level of DA that Cadros can achieve. Extensive simulation experiments have been conducted to verify the effectiveness of Cadros.

The number of residential proxies as a service has grown as an emerging gray-area business, in which a provider utilises (compromised) hosts inside residential networks in order to provide anonymity services, effectively hiding the IP addresses of their clients behind other, seemingly normal, Residential IP (RESIP) addresses. This paper investigates commercial RESIP proxy service providers and looks into their host recruitment practices, which are often suspicious or borderline legal. In turn, the paper proposes a detection mechanism for identifying RESIP proxy flows on compromised consumer electronic devices, whereby the proxy software may operate without the user’s knowledge or explicit consent.

Identification of test scenarios for Automated Driving Systems (ADSs) remains a key challenge for the Verification & Validation of ADSs. Various approaches including data based approaches and knowledge based approaches have been proposed for scenario generation. Identifying the conditions that lead to high severity traffic accidents can help us not only identify test scenarios for ADSs, but also implement measures to save lives and infrastructure resources. Taking a data based approach, in this paper, we introduce a novel accident data analysis method for generating test scenarios where we analyze UK's Stats19 accident data to identify trends in high severity accidents for test scenario generation. This paper first focuses on the severity of the accidents with the goal of relating it to static and time-dependent internal and external factors in a comprehensive way taking into account Operational Design Domain (ODD) properties, e.g. road, environmental conditions, and vehicle properties and driver characteristics. For this purpose, the paper utilizes a data grouping strategy (coarse-graining) and builds a logistic regression approach, derived from conventional regression models, in which emerging features become more pronounced, while uninteresting features and noise weaken. The approach makes the relationship between the factors and outcome variable more visible and hence well suited for the severity analysis. The method shows superior performance as compared to ordinary logistic models measured by goodness of fit and accounting for model variance $(R^{2}=0.05$ for the ordinary model, $R^{2}=0.85$ for the current model). The model is then used to solve the inverse problem of constructing high-risk pre-crash conditions as test scenarios for simulation based testing of ADSs.

The method of Support Vector Machine (SVM) based on Dissolved Gas Analysis (DGA) has been studied in the field of power transformer fault diagnosis. However, there are still some shortcomings, such as the fuzzy boundaries of DGA data, and SVM parameters are difficult to determine. Therefore, this paper proposes a power transformer fault diagnosis method based on Kernel Principal Component Analysis (KPCA) and a hybrid improved Seagull Optimization Algorithm to optimize the SVM (TISOA-SVM). Firstly, KPCA is used to extract features from DGA feature quantities. In addition, TISOA is further proposed to optimize the SVM parameters to build the optimal diagnosis model based on SVM. For the SOA, three improvement methods are proposed. An improved tent map is used to replace the original population initialization to improve population diversity. In addition, the nonlinear inertia weight and random double helix formula are proposed to improve the optimization accuracy and efficiency of the SOA. Then, benchmarking functions are used to test the optimization performance of TISOA and six algorithms, and the results show that TISOA has the best optimization accuracy and convergence speed. Finally, the fault diagnosis method based on KPCA and TISOA-SVM is obtained, and it is noteworthy that three examples are tested to verify the diagnostic performance of the proposed method. These results show that the proposed method has higher diagnostic accuracy, shorter diagnosis time, stronger significance and validity than other methods. Therefore, a research idea is provided for solving practical engineering problems in the field of fault diagnosis.

In the 5G era, the problem of data islands in various industries restricts the development of artificial intelligence technology, so data sharing is proposed. High-quality data sharing directly affects the effectiveness of machine learning models, but data leakage and abuse will inevitably occur in the process. As a consequence, in order to solve this problem, federated learning is proposed. This method uses the personalized data of multiple edge devices to train the model. The central server collects the training results of the edge devices and updates the global model, and then iteratively tests and updates the model through the edge devices. However, edge devices may have problems such as unbalanced load and exit from the training process, which makes the training time of the model long and the effect is poor. Therefore, in the process of federated learning, the selection of reliable and high-quality edge devices becomes crucial. On this basis, in this paper, we introduces reinforcement learning (RL) to pre-select edge devices and obtain a set of candidate devices, then determines reliable edge devices through social attribute perception. Simulation experiment data analysis demonstrate that this scheme can improve the reliability of federated learning and complete the training process in a shorter time, the efficiency of federated learning increased by approximately 10.3%.

Malware obfuscation can make both automatic and manual analysis of its binary code and the contained functionality significantly more time consuming. For malware research it would therefore be useful to be able to study the effects of different obfuscation methods on the resulting binary code. While some obfuscations are applied through rewriting of the binary, others have to be applied at source code level or during compile time. However, the source code of in-the-wild malware is often not available. For this paper, we collected the source code of eleven open source malware samples from the past 12 years and analyzed if they still compile on current systems. Furthermore, basic static analysis was performed to evaluate the usefulness of the resulting binaries for further malware obfuscation research. Our results indicate, that it is possible to compile available samples with moderate effort and the resulting binaries are very well suited for research purposes.

Recently, Infrastructure-as-a-Service (IaaS) clouds provide virtual machines (VMs) with a large amount of memory. To make the migration of such large-memory VMs flexible, split migration has been proposed. It divides the memory of a VM into smaller pieces and transfers them to multiple destination hosts. After the migration, the VM runs across multiple hosts and its memory data is exchanged between hosts by remote paging. There is often unused memory in a large-memory VM, but data of even unused memory is transferred via the network. This paper proposes FCtrans to achieve efficient split migration and remote paging by considering unused memory. FCtrans avoids transferring data of unused memory to destination hosts on split migration. Similarly, it does not perform remote paging for unused memory and immediately continues the execution of the VM. To enable this, FCtrans keeps track of the memory usage of a VM after starting split migration. In addition, it transparently and consistently reclaims memory released after used by the guest operating system using VM introspection and deals with it as unused. We have implemented FCtrans in KVM and conducted experiments using a VM with 352 GB of memory on the StarBED testbed. It is shown that split migration became up to 29x faster and the memory access performance of a VM across multiple hosts improved by up to 85 %., 2021 IEEE 14th International Conference on Cloud Computing (CLOUD), September 5-10, 2021, Chicago, IL, USA

Electromagnetic Transient (EMT) simulation tools are typically developed using conventional procedural programming languages. On the other hand, modern high-level and equation-based programming languages, such as Modelica, are currently available. Modelica allows formulating models that are easy to develop, maintain and understand by expressing what needs to be computed without stating how it should be computed. This paper presents a Modelica based simulator for electromagnetic transients. It is demonstrated that this approach offers significant advantages for developing sophisticated models. Computational performance and accuracy are compared to a conventional EMT-type simulation tool.

This paper demonstrates a Power Hardware-in-the-Loop (PHIL) implementation of a decentralized optimal power flow (D-OPF) algorithm embedded into the operations of two microgrids connected by a tie line. To integrate the static behavior of the optimization model, a two layer control architecture is introduced. Underneath the dispatch commands from the D-OPF, a primary control scheme provides instantaneous reaction to the load dynamics. This setup is tested in the PHIL environment of the CoSES Lab in TU Munich. In the experiment, the two microgrids cooperatively optimize their operation through an ADMM based unbalanced D-OPF. The operations is then benchmarked against the exclusive use of primary control, without D-OPF. The decentralized approach outperforms, but also shows minor inefficiencies of integrating optimization methods into the real-time operation of the system.

The Internet of Things (IoT) aims to connect everyday physical objects to the internet. These objects will produce a significant amount of data. The traditional cloud computing architecture aims to process data in the cloud. As a result, a significant amount of data needs to be communicated to the cloud. This creates a number of challenges, such as high communication latency between the devices and the cloud, increased energy consumption of devices during frequent data upload to the cloud, high bandwidth consumption, while making the network busy by sending the data continuously, and less privacy because of less control on the transmitted data to the server. Fog computing has been proposed to counter these weaknesses. Fog computing aims to process data at the edge and substantially eliminate the necessity of sending data to the cloud. However, combining the Service Oriented Architecture (SOA) with the fog computing architecture is still an open challenge. In this paper, we propose to decompose services to create linked-microservices (LMS). Linked-microservices are services that run on multiple nodes but closely linked to their linked-partners. Linked-microservices allow distributing the computation across different computing nodes in the IoT architecture. Using four different types of architectures namely cloud, fog, hybrid and fog+cloud, we explore and demonstrate the effectiveness of service decomposition by applying four experiments to three different type of datasets. Evaluation of the four architectures shows that decomposing services into nodes reduce the data consumption over the network by 10% - 70%. Overall, these results indicate that the importance of decomposing services in the context of fog computing for enhancing the quality of service., Comment: IEEE TRANSACTIONS SUSTAINABLE COMPUTING, 2019

Spatial co-location pattern mining (SCPM) aims to discover subsets of spatial features frequently located together in proximate areas. Previous studies of SCPM solely concern the inter-features association of a pattern, but neglect the interesting intra-feature behavior. In this paper, we propose the task of spatial co-location pattern mining with coupling relation consideration (SCPM-CR) to capture complex relations embedded in a co-location. Specifically, InterPCI measure is designed to capture the inter-features coupling by considering the comprehensive interaction of objects for the features in a pattern, and luckily it possesses the anti-monotone property. Another measure, IntraCAI, is proposed to capture the congregating behavior of intra-feature objects under the restriction of a co-location. A general framework is designed for SCPM-CR task and experiments show that a large fraction of computation time is devoted to identifying the participating objects. To tackle this calculation bottleneck, a novel candidate-and-search algorithm is suggested, CS-HBS, equipped with heuristic backtracking search. Extensive experiments are conducted on real and synthetic datasets to demonstrate the superiority of SCPM-CR compared with traditional SCPM methods, and also to validate the efficiency and scalability of CS-HBS. Experimental results show that CS-HBS outperforms the baselines by several times or even orders of magnitude.

Traditional service reliability evaluation involves queuing, calculating overflow, time-out, and other problems. With the development of Cloud computing and the emergence of new applications, parallel computation, redundant backup, and other complex environments shall be taken into consideration when evaluating reliability. Therefore, it's very important to carry out accurate, comprehensive, efficient reliability evaluation under the background of large-scale and complex development of Cloud computing. In view of the above problems, this paper puts forward an evaluation method of virtual machine quantity in cold and hot operation mode based on reliability guarantee, that is, to evaluate the reliability of the system by using MDD algorithm and adjust the quantity in cold and hot backup mode when the number of the virtual machine in operation mode changes under the background of meeting the reliability requirement R.

With the recent development of IoT technology, a variety of products and services that were previously not implemented due to technical limitations are gradually becoming feasible. Accordingly, in the agricultural field, smart farm implementation is gradually spreading through the use of IoT technology. In this paper, we analyse standard technologies for smart farms through implementation of smart farm devices using IoTivity, an open source software provided by OCF which is a middleware standard technology for logical interoperability between IoT devices.

In recent years, there are an increase in the mortality rate due to Chronic obstructive pulmonary disease (COPD) patients and it is estimated that it will increase in the coming years. Traditional methods take a long time to identify these diseases because a lot of clinical tests has to be performed for getting the confirmation, however with the advent of intelligent techniques as well as looking at the potential of the powerful techniques for predicting other critical diseases, it is believed that it would help to detect the chronic diseases at an early stage in a precise manner. In this paper, an attempt has been made to detect COPD patients and at the same time, it could distinguish the stages such as the early stage of chronic obstructive pulmonary disease patients (ESCP) and the Advanced stage of COPD patients (ASCP). We have used the Recursive Feature Elimination, Cross-Validated (RFECV) feature selection method to select features and then we consult the doctors, those are expert in the field to recommend the features among the features selected using RFECV method. The features selected using the doctor recommendation called features reduction with doctor recommendation (FRDR). After selecting two groups of features, we have used different machine learning algorithms to compare the performance of the algorithms as well as the importance of the features. It was found that the features selected using the RFECV method could able to provide accuracy of 96%, whereas the features selected with doctor recommendation name FRDR could able to provide accuracy of 90%. Although there is a difference of result in both the methods but overall, both set of features produces a good result. So, it is recommended that this approach would help distinguish different stages in real-life situations.

The Mirai botnet and its variants has made Internet of Things (IOT) devices a powerful amplifying platform for Low Rate Distributed Denial-of-Service (LDDoS) attacks. In this paper, we investigate and develop a novel semi-supervised Locality Sensitive Incremental Transductive Support Vector Machine (LS-ITSVM) method. The proposed method maximizes the margins of different network flows by incorporating local frequency-domain features from the autocorrelation sequence of network flow into the regularization time-domain framework of TSVM. And it saves training and detecting time by incremental training support vectors and new added samples. The result of simulation proves the proposed method can distinguish abnormal network flows with higher detection accuracy, faster training and response time, and prevent abnormal network flow groups with less impaction.

Operating System noise is a well-known problem that occurs mainly in large-scale systems running HPC application, and it also limits performance scalability. In order to reduce OS noise occurring in a general-purpose OS, different design approached kernels like micro-kernel and lightweight kernel are studied. Recently unikernel optimized for application program has been studied. The purpose of this paper is to measure OS noise in unikernel. We compared OS noise of Linux, general-purposed OS, and Azalea-unikernel, one of the unikernels. Well-known OS noise measurement benchmarks such as fixed work quanta (FTQ), fixed time quanta (FWQ), and Hourglass were used. As a result of the experiment, it was confirmed that Azalea-unikernel has less OS noise than Linux, and it also means unikernel is a kernel suitable for supporting scalability on manycore systems.

In this paper, using the time-series throughput data generated from a simulated factory scenario, we study throughput prediction using convolutional neural network (CNN). Different with image or numerical recognition using CNN, in which the distribution of the prediction target during training stage usually has the similar level, the distribution of the throughput data concentrates only on several values. This centralized distribution may degrade the prediction accuracy. Therefore, we will propose a new CNN prediction method employing target vectorization which can mitigate the centralization of distribution. This method makes training process of CNN hold more possibility and improves the prediction accuracy of the throughput.

Software system admins depend on log data for understanding system beliavior, monitoring anomalies, tracking software bugs, and malfunctioning detection. Log analysis based on machine learning techniques enables to transform of raw logs into meaningful information that helps the DevOps team and administrators to solve problems. AI ensures to group similar logs together and keeps periodic logs more organized and sorted, allowing us to get to where we need to look faster. In this paper, we present a log classification system on log data generated by VoIP (Voice over Internet Protocol) soft-switch product. In this way, we targeted to detect the problem, direct it to the relevant department, allocate resources, and solve software bugs faster and more efficiently. Machine learning algorithms such as Linear Classifiers, Support Vector Machines, Decision Tree, Random Forest, Boosting, K-Nearest Neighbors, and Multilayer Perceptron are used for log classification.

Due to the rapid spread of mutated variants of coronavirus, and the lack of complete mass vaccination all over the world, everybody needs to strict adherence to personal hygiene and social distancing. And also, contamination of infected individuals for preventing coronavirus spreading plays a vital role. Home quarantine seems to be the best isolation solution in terms of cost and psychological health of individuals who are infected or close contacts. However, the number of quarantine violations increases day by day in many parts of the world because of the long duration of the pandemic. In this paper, a potential solution is presented for monitoring the violation of home quarantine based on the Internet of Things (IoT) and cloud computing technologies. The deep learning convolutional neural network is trained on the cloud with images of quarantine subjects for face recognition. Monitoring IoT nodes expect individuals to scan their faces at regular intervals to prevent them from leaving quarantine areas. Moreover, by measuring their body temperature regularly, unexpected rises can be detected, and this situation can be reported to health authorities. IEEE

A key aspect of multi-energy microgrids (MEMGs) is the capability to efficiently convert and store energy in order to reduce the costs and environmental impact. Peer-to-peer (P2P) energy trading is a novel paradigm for decentralised energy market designs. In this paper, we investigate the external P2P energy trading problem and internal energy conversion problem within interconnected residential, commercial and industrial MEMGs. These two problems are complex decision-making problems with enormous high-dimensional data and uncertainty, so a multi-agent deep reinforcement learning approach combining the multi-agent actor-critic algorithm with the twin delayed deep deterministic policy gradient algorithm is proposed. The proposed approach can handle the high-dimensional continuous action space and aligns with the nature of P2P energy trading with multiple MEMGs. Simulation results based on three real-world MG datasets show that the proposed approach significantly reduces each MG’s average hourly operation cost. The impact of carbon tax pricing is also considered.

People are prone to developing opioid dependence and other health problems due to regular non-medical use, prolonged use, misuse, and use without medical supervision. In this paper, opioid-related healthcare data from Froedtert Health Medical System in Wisconsin are analyzed and machine learning models are proposed to predict the length of stay (LOS) of opioid overdose admitted patients. We also determine important features that impact the LOS. To explore the factors that significantly influence the LOS, we implement machine learning algorithms, namely, Random Forest and XGBoost, to select important features from the data. Predictive models such as Random Forest regressor, Gradient Boost regressor, Support Vector Machine and k-Neighbors regressor are conducted and trained on the top-50, 100, 300, 650, and 1000 important features. We propose to evaluate the regression models using Mean Squared Error (MSE) and R-squared.

JSON is a popular data interchange format for many web, cloud, and IoT systems due to its simplicity, human readability, and widespread support. However, applications must first parse and convert the data to a native in-memory format before being able to perform useful computations. Many big data applications with high performance requirements convert JSON data to Apache Arrow RecordBatches, the latter being a widely-used columnar in-memory format for large tabular data sets used in data analytics. In this paper, we analyze the performance characteristics of such applications and show that JSON parsing represents a bottleneck in the system. Various strategies are explored to speed up JSON parsing on CPU and GPU as much as possible. Due to performance limitation of the CPU and GPU implementations, we furthermore present an FPGA accelerated implementation. We explain how hardware components that can parse variable-sized and nested structures can be combined to produce JSON parsers for any type of JSON document. Several fully integrated FPGA-accelerated JSON parser implementations are presented using the Intel Arria 10 GX and Xilinx VU37P devices, and compared to the performance of their respective host systems; an Intel Xeon and an IBM POWER9 system. Result show the accelerators achieve an end-to-end throughput close to 7 GB/s with the Arria 10 GX using PCIe, and close to 20 GB/s with the VU37P using OpenCAPI 3. Depending on the complexity of the JSON data to parse, the bandwidth is limited by the host-to-accelerator interface or available FPGA resources. Overall, this provides a throughput increase of up to 6x, compared to the baseline application. Also, we observe a full system energy efficiency improvement of up to 59x more JSON data parsed per joule.

The microarchitectural state held by predictors in modern processors can leak sensitive information. This is the first work to analyze the security of a special type of predictor, the value predictor, and demonstrate new security attacks. The new attacks bypass all the existing predictor defenses which have not yet considered value predictors as sources of vulnerabilities. This work further shows there are many value predictor attack variants, as derived using our new attack model. This paper highlights the importance of security analysis of processor features before they are realized in silicon, so the security is understood at the design time.

Implementing hardware secure against side-channel attacks (SCA) demands significant time and expertise in hardware design. In this paper, we propose a simple and fast approach for synthesizing masked block cipher hardware from a C-code exploiting High-Level-Synthesis (HLS), which allows a very short design time. Compared to previous approaches, our proposal provides a systematic and general flow based on state-of-the-art Domain-Oriented Masking (DOM). We also present a fast security-validation flow for the synthesized circuits at the early design stages using commercial-off-the-shelf CAD tools. Efficacy of the proposed design-flow has been established over a set of representative benchmarks including a masked S-Box and a lightweight block-cipher.

This paper presents MobileSwap, a cross-device memory swapping scheme for mobile devices. It exploits the unbalanced utilization of memory resources across devices. MobileSwap achieves comparable-to-local swapping performance based on existing network infrastructure. This is realized by two novel approaches: resource dedicated swapping for fast swapping among devices and app aware swapping for network connectivity considerations. MobileSwap is implemented and deployed on real mobile devices. Experimental results show that MobileSwap can enhance app caching capability by 2x compared with no swapping, and improve performance by 2.3x compared with state-of-the-art remote swapping. More importantly, local swapping induced read-write conflicts are largely removed.

Real-world e-commerce systems need large cache capacities. Persistent memory (PM) can be employed to enlarge JVMs’ cache capacities, meanwhile they incur heavy write slowdowns and garbage collection overheads. This paper proposes JPDheap, a JVM heap design for PM-DRAM memories. A JPDheap is composed of a standard Java heap on DRAM and another heap on PM. The core insight is to separate heap objects and store them on DRAM or PM, allowing objects to be accessed much more efficiently. Our evaluation shows that JPDheap outperforms state-of-the-art heap designs by up to 115.96% in increasing applications’ throughput and by up to 87.03% in decreasing the average latency.

Analog layout design still relies heavily on manual efforts. Current fully automated flows are not yet able to satisfy the demands of versatile customization and not compatible to the existing manual flows. Interactive layout editing has the potential to bridge the gap between the manual flows and fully automated flows shooting for both performance and productivity. In this paper, we propose an interactive editing framework with instructions for both topological editing and detailed customization. We also propose an effective instant legalization algorithm for fast layout update during the real-time interaction with users.

The pursuit of manufacturing cost reduction reshaped the conventional system-on-chip (SoC) design and manufacturing flow into the horizontal business model. In this model, the design house loses control of the design during the manufacturing process. Therefore, this shift has introduced potential vulnerabilities at each stage of the flow and provides adversaries ample opportunities to cause piracy, security, and trust concerns. Further, SoCs deployed in IoT, smart, and mission-critical devices contain sensitive assets to perform security-critical applications, requiring an on-chip security engine (SE) to ensure protecting assets and secure operation throughout the lifecycle. In this paper, we present an end-to-end secure SoC lifecycle management flow that establishes trust at each stage of the manufacturing process, prevents potential security threats, provides secure provisioning schemes, and protects the chip from in-field and supply chain vulnerabilities.

Arbitration logic is extensively used in modern computer architectures to dynamically determine how shared hardware resources are allocated or accessed. Recent work has shown that machine learning techniques can learn non-obvious yet effective arbitration policies, which in simulation demonstrate superior performance over human-designed heuristics. However, existing methods based on deep learning are too expensive to be directly implemented as an arbitration unit in hardware. While some prior efforts managed to manually analyze and reduce a deep learning model into relatively small circuits in certain cases, such ad hoc and labor-intensive approaches cannot easily generalize. In this work, we propose a new methodology to automatically “distill” the arbitration logic from simulation traces. Starting by training a deep learning model, we leverage tree-based models as a bridge to convert the more complex model to a compact logic implementation. This paper presents a case study of the proposed methodology on a network-on-chip port arbitration task. Compared with an array of combinational multipliers that exactly computes the neural network output, our arbitration logic achieves up to 282x area reduction without significant performance degradation. Under the training traffic, our arbitration logic achieves up to 64x reduction in average packet latency and up to 5% increase in network throughput over the FIFO arbitration policy. The distilled arbitration policy is also able to generalize to different injection rates and traffic patterns.

Graphics processing units (GPUs) are being widely used as co-processors for performance acceleration in many autonomous embedded systems such as robotics and autonomous vehicles. However, current GPU hardware and systems software, including GPU device drivers, compilers, and operating systems, do not implement proper memory protection mechanisms due to performance and proprietary reasons, causing severe vulnerabilities such as information leakage. In this paper, we present gGuard, a leakage-resilient GPU memory management system with strong isolation. Based on the intrinsic characteristics of information leakage vulnerabilities on GPUs, gGuard develops a set of efficient and accurate data shredding techniques implemented at the compiler, library, and operating system levels, with the core idea of exploring the data access patterns and dependencies for efficient application-aware data shredding. Our implementation and evaluation show that gGuard can provide effective mitigation on GPU data leakage issues through efficient GPU data shredding while introducing less than 6% overhead in all tested scenarios.

Implantable Cardioverter Defibrillator (ICD) is an ultra-low-power device which monitors heart rate and delivers in-time defibrillation on detected ventricular arrhythmias (VAs). The parameters of VAs detection mechanism on each recipient’s ICD are supposed to be fine-tuned to obtain accurate detection due to the individual’s unique rhythm features. However, the process extremely relies on clinical expertise and thus must be conducted manually and routinely by cardiologists diagnosing massive amount of rhythm data. In this paper, we introduce a novel self-supervised on-device personalization of convolutional neural network (CNNs) for VAs detection. We first propose a computing framework consisting of an edge device and an ICD to enable efficient on-device CNNs personalization and real-time inference respectively. Then, we propose a generative model that learns to synthesize patient-specific intracardiac EGMs signals, which can then be used as personalized training data to improve patient-specific VAs detection performance on ICDs. Evaluations on three detection models show that the self-supervised on-device personalization significantly improve VAs detection performance under a patient-specific setting.

For safety-critical| computer systems, time-predictability and performance are usually required simultaneously in I/O virtualization. However, both requirements are challenging to achieve due to complex I/O access path and resource management at system level and lack of support from preemptive scheduling at I/O hardware level. In this paper, we propose a new framework, I/O-GUARD, which reconstructs the system architecture of I/O virtualization, bringing a dedicated hardware hypervisor to handle resource management throughout the system. The hypervisor improves system real-time performance by enabling preemptive scheduling in I/O virtualization with both analytical and experimental real-time guarantees. Specifically, I/O-GUARD is a First-of-Its-Kind framework for multi-/many-core I/O virtualization.

It becomes more and more common for people to have multiple mobile devices. This opens the opportunity of multi-surface computing in which users interact with an app using multiple devices simultaneously. Recently, a system called FLUID was developed, which can distribute User Interface (UI) elements of an app to multiple devices to support multi-surface computing. FLUID enables general, flexible and transparent multi-device interaction, which cannot be achieved by previous approaches such as screen mirroring, app migration, and customized app development on multiple devices. However, the practicality of FLUID is still severely limited because it requires that (1) the app source codes must be available and (2) the same app is pre-installed on all devices. This paper presents PRUID, a UI distribution system that is free from the above-mentioned limitations of FLUID. PRUID captures and extracts relevant information about UI elements to be distributed completely at run time, without requiring the app source code. An app-independent UI agent is designed to dock and render the UI components distributed to the guest device, so pre-installation of the app on guest devices is not required. We developed representative use cases to demonstrate the usage and evaluate the performance of PRUID. The evaluation results show that the extra overhead incurred due to the UI information extraction at run time is marginal and PRUID provides a smooth user experience.

The combination of deep learning techniques and commercial 3D sensors reveal a bright future as they provide a low cost and convenient method to collect and analyze depth information from the environment for various applications ranging from industrial modeling to mobile face recognition. Despite the abundant research devoted to the development of more accurate, flexible and efficient machine learning schemes as well as 3D sensors, security concerns related to these techniques remain largely untouched. In this paper, we propose a novel adversarial attack against this combination by showing that deep learning models with popular 3D sensors may misclassify real objects in the physical environment. Comparing to the existing attack algorithms against deep learning models developed for 3D data analysis that only consider digital point cloud data and single deep learning model, our attacks target popular commercial 3D sensors combined with various deep learning schemes in the black-box setting. Experimental results demonstrate that our 3D printed adversarial objects stay effective after scanned by a 3D sensor.

The scalable and extendable RISC-V ISA introduced a new level of flexibility in designing highly customizable processors. This flexibility in processor designs adds to the complexity of already complex functional verification process. Although formal methods are increasingly used to exhaustively verify the processors, the required manual effort and verification expertise become the major hurdles in industrial flows. Furthermore, efficient ISA modeling techniques are required that are scalable to multiple ISA extensions and to different architectural variants of a processor. This paper proposes a trace notation for ISA definition to capture the implicit execution behavior of a processor and specific characteristics. The proposed trace notation can be annotated with timing and hierarchy information to adapt the trace to any kind of processor architecture. From this trace notation, a complete set of properties are generated to detect all functional bugs in a processor implementation. The approach requires significantly less manual effort compared to the contemporary techniques. Its industry strength has been demonstrated by formally verifying a wide variety of RISC-V processor implementations with one or more ISA extensions (RV32I, C, Zicsr, M and custom extensions supporting AI acceleration and safety features).

The data imbalance problem often occurs in nanometer VLSI applications, where normal cases far outnumber error ones. Many imbalanced data handling methods have been proposed, such as oversampling minority class samples and downsampling majority class samples. However, existing methods focus on improving the quality of minority classes while causing quality deterioration of majority ones. In this paper, we propose a two-stage classifier to handle the data imbalance problem. We first develop an iterative neural network framework to reduce false alarms. Then the oversampling method on a final classification network is applied to predict the two classes better. As a result, the data imbalance problem is well handled, and the quality deterioration of majority classes is also reduced. Since the iterative stage does not change any existing network structure, any convolutional neural network can be used in the framework. Compared with the state-of-the-art imbalanced data handling methods, experimental results on the hotspot detection problem show that our two-stage classification method achieves the best prediction accuracy and reduces false alarms significantly.

Hyperdimensional Computing (HDC) is introduced as a promising solution for robust and efficient learning on embedded devices with limited resources. Since HDC often runs in a distributed way, edge devices need to share their model with other parties. However, the learned model by itself may expose information of the train data, resulting in a serious privacy concern. This paper is the first effort to show the possibility of a model inversion attack in HDC and provide solutions to overcome the challenges. HDC performs learning tasks after mapping data points into high-dimensional space. We first show the vulnerability of the HDC encoding module by introducing techniques that decode the high-dimensional data back to the original space. Then, we exploit this invertibility to extract the HDC model’s information and reconstruct the train data just by accessing the model. To address the privacy challenges we propose two iterative techniques which scrutinize HDC model from a privacy perspective: (i) intelligent noise injection that identifies and randomizes insignificant features of the model in the original space, and (ii) model quantization that removes model’s recoverable information while teaches the model iteratively to compensate the possible quality loss. Our evaluation over a wide range of classification problems indicates that our solution reduces the information leakage by 92 %(66 %) while having less than 5 % (3%) impact on the learning accuracy.

In this paper we propose to combine the safe C dialect Checked C with concolic testing to obtain an effective methodology for attaining safer C code. Checked C is a modern and backward compatible extension to the C programming language which provides facilities for writing memory-safe C code. We utilize incremental conversions of unsafe C software to Checked C. After each increment, we leverage concolic testing, an effective test generation technique, to support the conversion process by searching for newly introduced and existing bugs.Our RISC-V experiments using the RIOT Operating System (OS) demonstrate the effectiveness of our approach. We uncovered 4 previously unknown bugs and 3 bugs accidentally introduced through our conversion process.

Due to their crucial role in many decision-making tasks, Deep Neural Networks (DNNs) are common targets for a large array of integrity breaches. In this paper, we propose AID, a novel methodology to Attest the Integrity of DNNs. AID generates a set of test cases called edge-points that can reveal whether a model has been compromised. AID does not require access to parameters of the DNN and can work with a restricted black-box access to the model, which makes it applicable to most real life scenarios. Experimental results show that AID is highly effective and reliable. With at most four edge-points, AID is able to detect eight representative integrity breaches including backdoor, poisoning, and compression attacks, with zero false-positive.

This paper presents the open-source high-level synthesis (HLS) research framework Bambu. Bambu provides a research environment to experiment with new ideas across HLS, high-level verification and debugging, FPGA/ASIC design, design flow space exploration, and parallel hardware accelerator design. The tool accepts as input standard C/C++ specifications and compiler intermediate representations (IRs) coming from the well-known Clang/LLVM and GCC compilers. The broad spectrum and flexibility of input formats allow the electronic design automation (EDA) research community to explore and integrate new transformations and optimizations. The easily extendable modular framework already includes many optimizations and HLS benchmarks used to evaluate the QoR of the tool against existing approaches [1]. The integration with synthesis and verification backends (commercial and open-source) allows researchers to quickly test any new finding and easily obtain performance and resource usage metrics for a given application. Different FPGA devices are supported from several different vendors: AMD/Xilinx, Intel/Altera, Lattice Semiconductor, and NanoXplore. Finally, integration with the OpenRoad open-source end-to-end silicon compiler perfectly fits with the recent push towards open-source EDA.

The brain-inspired hyperdimensional computing (HDC) gains attention as a light-weight and extremely parallelizable learning solution alternative to deep neural networks. Prior research shows the effectiveness of HDC-based learning on less powerful systems such as edge computing devices. However, the many-class classification problem is beyond the focus of mainstream HDC research; the existing HDC would not provide sufficient quality and efficiency due to its coarse-grained training. In this paper, we propose an efficient many-class learning framework, called CascadeHD, which identifies latent high-dimensional patterns of many classes holistically while learning a hierarchical inference structure using a novel meta-learning algorithm for high efficiency. Our evaluation conducted on the NVIDIA Jetson device family shows that CascadeHD improves the accuracy for many-class classification by up to 18% while achieving 32% speedup compared to the existing HDC.

With feature size scaling and complexity increase of circuit designs, hotspot detection has become a significant challenge in the very-large-scale-integration (VLSI) industry. Traditional detection methods, such as pattern matching and machine learning, have been made a remarkable progress. However, the performance of classifiers relies heavily on reference layout libraries, leading to the high cost of lithography simulation. Querying and sampling qualified candidates from raw datasets make active learning-based strategies serve as an effective solution in this field, but existing relevant studies fail to take sufficient sampling criteria into account. In this paper, embedded in pattern sampling and hotspot detection framework, an entropy-based batch mode sampling strategy is proposed in terms of calibrated model uncertainty and data diversity to handle the hotspot detection problem. Redundant patterns can be effectively avoided, and the classifier can converge with high celerity. Experiment results show that our method outperforms previous works in both ICCAD2012 and ICCAD2016 Contest benchmarks, achieving satisfactory detection accuracy and significantly reduced lithography simulation overhead.

Rowhammer is a memory-based attack that leverages capacitive-coupling to induce faults in modern dynamic random-access memory (DRAM). Over the last decade, a significant number of Rowhammer attacks have been presented to reveal that it is a severe security issue capable of causing privilege escalations, launching distributed denial-of-service (DDoS) attacks, and even runtime attack such as control flow hijacking. Moreover, the Rowhammer vulnerability has also been identified and validated in both cloud computing and data center environments, threatening data security and privacy at a large scale. Various solutions have been proposed to counter Rowhammer attacks but existing methods lack a circuit-level explanation of the capacitive-coupling phenomenon in modern DRAMs, the key cause of Rowhammer attacks.In this paper, we develop an analytical model of capacitive-coupling vulnerabilities in DRAMs. We thoroughly analyze all parameters in the mathematical model contributing to the Rowhammer vulnerability and quantify them through real DRAM measurements. We validate the model with different attributions on a wide range of DRAM brands from various manufacturers. Through our model we re-evaluate existing Rowhammer attacks on both DDR3 and DDR4 memory, including the recently developed TRRespass attack. Our analysis presents a new Rowhammer attack insight and will guide future research in this area.

This paper proposes to explain the black-box feature of data-driven machine learning (ML) models used for controlling power electronic converters for the first time. As the name suggests, their “black box” feature prevents a clear understanding of the physical insights behind these ML models. It remains a fundamental aspect, if one plans to take action based on a prediction, or deploy a new ML model. Moreover, leaked and corrupted data during the training process can easily augment unexplainable actions from them. To address these issues, we first interpret the actions of the black box models by calculating a conditional entropy for each input with respect to an output. Using this metric, the averaged relationships between each input-output can be mapped and representative conclusions are firstly drawn on identifying erroneous data. Finally, these abnormal data are then removed from the training database to improve the interpretability & classification abilities of the ML model. We illustrate our findings on the performance of a regression based learning tool used for controlling a grid-connected voltage source inverter (VSI).

This paper studies the demand-capacity balancing (DCB) problem in air traffic flow management (ATFM) with collaborative multi-agent reinforcement learning (MARL). To attempt the proper ground delay for resolving airspace hotspots, a multi-agent asynchronous advantage actor-critic (MAA3C) framework is firstly constructed with the long short-term memory network (LSTM) for the observations, in which the number of agents varies across training steps. The unsupervised learning and supervised learning are then introduced for better collaboration and learning among the agents. Experimental results demonstrate the scalability and generalization of the proposed frameworks, by means of applying the trained models to resolve different simulated and real-world DCB scenarios, with various flights number, sectors number and capacity settings. 10.13039/501100004543-China Scholarship Council