Search

Extensive integration of power electronics apparatuses complicates the modern power grid and consequently necessitates time-domain transients study for its planning and operation. In this work, a heterogeneous computing architecture utilizing the CPU and graphics processing unit (GPU) is proposed for the efficient study of interactions between a power grid network and massive utility-scale battery energy storage systems (BESSs). The device-level electromagnetic transient (EMT) simulation aiming at enhanced fidelity of the BESS is conducted simultaneously with electro-mechanical transient stability (TS) simulation which suffices system-level dynamic security assessment. Since the reservation of a large amount of energy storage units is computationally intensive for the CPU, the concurrent multi-streaming, multi-threading capability of GPU is exploited to achieve asynchronous sequential-parallel processing, so that the proposed EMT-TS co-simulation can flexibly harness all available computing resources. The multi-rate scheme is adopted for further computational burden alleviation in addition to achieving timely information exchange. It shows that the heterogeneous computation of an IEEE 118-bus system integrated with a substantial number of distributed batteries becomes feasible following the achievement of a remarkable speedup of over 200, and the device- as well as system-level accuracy are validated by MATLAB/Simulinkand DSATools/TSAT simulation, respectively.

Real-time dense SLAM techniques aim to reconstruct the dense three-dimensional geometry of a scene in real time with an RGB or RGB-D sensor. An indoor scene is an important type of working environment for these techniques. The planar prior can be used in this scenario to improve the reconstruction quality, especially for large low-texture regions that commonly occur in an indoor scene. This article fully explores the planar prior in a dense SLAM pipeline. First, we propose a novel plane detection and segmentation method that runs at 200 Hz on a modern graphics processing unit. Our algorithm for constructing global plane constraints is very efficient; hence, we use it in the process of each input frame for the camera pose estimation while maintaining the real-time performance. Second, we propose herein a plane-based map representation that greatly reduces the memory footprint of plane regions while keeping the geometric details on planes. The experiments reveal that our system yields superior reconstruction results with planar information running at more than 30 fps. Aside from speed and storage improvements, our technique also handles the low-texture problem in plane regions.

In this paper, a parallel and gradient-free strategy is presented for autonomous navigation of a quadrotor with a limited field of view (FOV). The FOV constraint is nonlinear and non-continuous in general and involves higher order dynamics of the vehicle. Planning problems with such constraints are difficult to solve, especially with the realtime requirement. The proposed method first constructs an environment map by integrating the measurement of a limited-FOV sensor in a parallel process. Based on the map, it then selects a feasible local goal from a global path and generates a constraint-free nominal control sequence. Finally, a stochastic optimal control problem is solved to satisfy the limited FOV constraint, the dynamic constraint, the dynamic constraint, and the collision-free conditions at the same time. The objective functions are non-linear and non-continuous and involves dynamic of MAV. A parallel gradient-free algorithm is used to address these problems. Both the sensing and planning problems are solved parallelly on a light-weight Graphics Processing Unit (GPU). The effectiveness of the proposed strategy is demonstrated through both simulation and real flight experiments.

Evolutionary Multi-Tasking (EMT), which shares knowledge across multiple tasks while the optimization progresses online, has demonstrated superior performance in terms of both optimization quality and convergence speed over its single-task counterpart in solving complex optimization problems. However, most of existing EMT algorithms only consider to handle two tasks simultaneously. As the computational cost incurred in the evolutionary search and knowledge transfer increased rapidly with the number of optimization tasks, these EMT algorithms cannot meet today’s requirements of optimization service on the cloud for many real-world applications, where hundreds or thousands of optimization requests (labelled as large-scale EMT) are often received simultaneously and require to be optimized in a short time. Recently, Graphics Processing Unit (GPU) computing has attracted extensive attention to accelerate the applications possessing large-scale data volume that are traditionally handled by the Central Processing Unit (CPU). Taking this cue, towards large-scale EMT, in this paper, we propose a new EMT paradigm based on the island model with the Compute Unified Device Architecture (CUDA), which is able to handle a large number of continuous optimization tasks efficiently and effectively. Moreover, under the proposed paradigm, we develop the GPU-based implicit and explicit knowledge transfer mechanisms for EMT. To evaluate the performance of the proposed paradigm, comprehensive empirical studies have been conducted against its CPU-based counterpart in large-scale EMT.

Model predictive control (MPC) using randomized optimization is expected to solve different control problems. However, it still faces various challenges for real-world applications. This paper attempts to solve those challenges and demonstrates a successful implementation of randomized MPC on the autonomous driving using a radio-controlled (RC) car. First of all, a sample generation technique in the frequency domain is discussed. This prevents undesirable randomness which affect the smoothness of the steering operation. Second, the proposed randomized MPC is implemented on a Graphics Processing Unit (GPU). The expected GPU acceleration in calculation speed at various problem sizes is also presented. The results show the improved control performance and computational speed that was not achievable using CPU based implementation. Besides, the selection of parameters for randomized MPC is discussed. The usefulness of the proposed scheme is demonstrated by both simulation and experiments. In the experiments, a 1/10 model RC car is used for collision avoidance task by autonomous driving.

The Deep Learning (DL) paradigm gained remarkable popularity in recent years. DL models are used to tackle increasingly complex problems, making the training process require considerable computational power. The parallel computing capabilities offered by modern GPUs partially fulfill this need, but the high costs related to GPU as a Service solutions in the cloud call for efficient capacity planning and job scheduling algorithms to reduce operational costs via resource sharing. In this work, we jointly address the online capacity planning and job scheduling problems from the perspective of cloud end-users. We present a Mixed Integer Linear Programming (MILP) formulation, and a path relinking-based method aiming at optimizing operational costs by (i) rightsizing Virtual Machine (VM) capacity at each node, (ii) partitioning the set of GPUs among multiple concurrent jobs on the same VM, and (iii) determining a due-date-aware job schedule. An extensive experimental campaign attests the effectiveness of the proposed approach in practical scenarios: costs savings up to 97% are attained compared with first-principle methods based on, e.g., Earliest Deadline First, cost reductions up to 20% are obtained with respect to a previously proposed Hierarchical Method and up to 95% against a dynamic programming-based method from the literature. Scalability analyses show that systems with up to 100 nodes and 450 concurrent jobs can be managed in less than 7 seconds. The validation in a prototype cloud environment shows a deviation below 5% between real and predicted costs.

This letter presents a ground-penetrating radar (GPR) forward model on dispersive media based on parallel conformal symplectic Euler algorithm. We developed a symplectic Euler algorithm combined with conformal meshes and graphics processing unit (GPU) acceleration technology, which proved to be an accurate and effective method to simulating GPR electromagnetic wave propagation in dispersive media.

Graphics processing units (GPUs) power today’s fastest supercomputers, are the dominant platform for deep learning, and provide the intelligence for devices ranging from self-driving cars to robots and smart cameras. They also generate compelling photorealistic images at real-time frame rates. GPUs have evolved by adding features to support new use cases. NVIDIA’s GeForce 256, the first GPU, was a dedicated processor for real-time graphics, an application that demands large amounts of floating-point arithmetic for vertex and fragment shading computations and high memory bandwidth. As real-time graphics advanced, GPUs became programmable. The combination of programmability and floating-point performance made GPUs attractive for running scientific applications. Scientists found ways to use early programmable GPUs by casting their calculations as vertex and fragment shaders. GPUs evolved to meet the needs of scientific users by adding hardware for simpler programming, double-precision floating-point arithmetic, and resilience.

We present a graphics processing unit (GPU)-based real time distributed dynamic strain sensing in optical frequency domain reflectometry (OFDR). GPU can accelerate the data processing rate distributed sensing in OFDR because it contains many parallel processing steps. We analyze parameters and performance of tunable laser source (TLS), data acquisition card (DAQ) and GPU. We present a choice method of the thread number in a block based on the GPU streaming multiprocessor utilization efficiency. We construct several kernel functions based on process steps of strain sensing and discuss the selection of blocks number for each kernel function. In the experiment, we find that total process time by the parallel computing in GPU is enhanced about 81 times compared with the serial computing in CPU. The measurement rate of this system is up to 60 Hz for real time distributed dynamic strain sensing. We achieve a dynamic strain sensing with a peak-to-peak variation of $2.5~\mu \varepsilon $ , a vibration frequency of 20 Hz and a sensing spatial resolution of 20 cm at a sensing range of 200 m. More importantly, two strain areas simultaneously loaded with a distance interval of 20 cm can be distinguished clearly.

We proposed a novel definition for thickness, termed “volumetric thickness” to clearly visualize thickness distribution in solid objects. Because our proposed method analyzes the thickness at internal points of an object, the computational complexity for its implementation is significantly higher than that of the conventional ray and sphere methods, which analyze do so only at points on the object's surface. This letter describes a fast analysis technique for volumetric thickness of 3D objects. To realize the computation in a realistic timeframe, we utilize a culling technique with hierarchical bounding boxes and a parallel processing function within the graphics processing unit (GPU). In the culling process, the necessary data is collected for appropriate utilization of the fast shared memory of GPU. We also present the numerical experiments conducted to verify the effectiveness of the proposed method.

Forward-looking correlated imaging plays an increasingly important role in modern radar imaging systems. It overcomes disadvantages of traditional side or squint synthetic aperture radar (SAR) which is dependent on specific relative motion between the radar and target scene. A new microwave forward-looking correlated 3-D imaging method based on random radiation field combined with sparse reconstruction is proposed in this article. Firstly, phased array radar (PAR) is adopted to form different and random antenna patterns. Then, combined with the compressed sensing (CS) theory, the target image can be recovered with very few samples which can break through Rayleigh resolution limitation. Furthermore, the proposed method can achieve resolution at least 5.5 times higher than real aperture imaging. To raise computation efficiency of sparse reconstruction, an improved quasi-Newton iteration method based on graphics processing unit (GPU) platform is developed. Meanwhile, a GPU-based (NVIDIA Tesla K40c) accelerated computing method can significantly reduce the processing time compared with the time given by a personal computer (PC). Both simulation and field experiment verify the validity of the proposed method.

The increasingly complex AC/DC network as a result of the massive integration of wind farms manifests the significance of a comprehensive transient study. In this work, the wind turbine (WT) and the DC grid are modeled in detail for the electromagnetic transient (EMT) simulation to maximize its fidelity, whilst the AC grid transient stability is analyzed by dynamic simulation (DS). An interactive EMT-DS interface is thus introduced to enable their concurrency and subsequently form a co-simulation. The CPU which is dominant in system study faces a tremendous challenge in handling a great number of components albeit they exhibit homogeneity. The many-core graphics processing unit (GPU) featuring massive parallelism is therefore exploited and following the definition of an adaptive computing boundary, a flexible heterogeneous sequential-parallel processing architecture is proposed for efficient analysis of the wind farm-integrated AC/DC grid. Topological reconfiguration of WTs is specifically carried out to reduce the numerical order whilst enhancing system homogeneity that enables the GPU to thoroughly utilize its peculiar property of single-instruction multiple-thread (SIMT) compute paradigm. Consequently, significant speedups can be attained by the proposed computing framework over pure CPU computation, while its accuracy is validated by the commercial EMT and dynamic security analysis tools PSCAD/EMTDC and DSATools, respectively.

Maximal clique enumeration (MCE) is a classic problem in graph theory to identify all complete subgraphs in a graph. In prior MCE work, the Bron-Kerbosch algorithm is one of the most popular solutions, and there are several improved algorithms proposed on CPU platforms. However, while few studies have focused on the related issue of parallel implementation, recently, there have been numerous explorations of the acceleration of general purpose applications using a graphics processing unit (GPU) to reduce the computing power consumption. In this article, we develop a GPU-based Bron-Kerbosch algorithm that efficiently solves the MCE problem in parallel by optimizing the process of subproblem decomposition and computing resource usage. To speed up the computations, we use coalesced memory accesses and warp reductions to increase bandwidth and reduce memory latency. Our experimental results show that the proposed algorithm can fully exploit the resources of GPU architectures, allowing for the vast acceleration of operations to solve the MCE problem.

High-fidelity electromagnetic transient (EMT) simulation plays a critical role in understanding the dynamic behavior and fast transients involved in operation, control, and protection of multiterminal dc (MTdc) grids. This article proposes a cost-effective high-performance real-time EMT simulation platform for large-scale cross-continental MTdc grids based on graphics processing unit (GPU). Fast dynamic transients from both ac and dc networks are captured in real time with 5 $\mu$ s time step, using advanced hybrid-discretized modular multilevel converter model, frequency-dependent transmission line model, and EMT-type model of synchronous generators. The proposed simulation platform i) assembles detailed EMT models of all components within an MTdc–ac grid into a single platform. This setup provides a complete simulation solution to capture fast transient signals required for high-bandwidth controller design and protection studies without any compromise; ii) implements the first GPU-based simulation architecture and corresponding algorithms for MTdc–ac grids with real-time performance at scales of 1s; iii) is highly efficient and balances the high utilization of GPU resources and low latency required for the simulation; and iv) outperforms the existing central processing unit- or digital signal processor (DSP)/field-programmable gate array-based simulators in terms of its higher scalability on large-scale MTdc–ac grids and superior price–performance ratio on the hardware. Accuracy and performance of the proposed platform are evaluated with respect to the reference results from power system computer aided design (PSCAD)/EMTdc environment.

This article presents ViRa 1 1 Available online: https://vira.aau.at/ . , a real-time open-source millimeter-wave radar simulation framework for perception-driven robotic applications. ViRa provides $(i)$ raw data of radar sensors in real-time, simulation of $(ii)$ multi-antenna configurations for spatial estimation of objects, $(iii)$ wave penetration of non-conductive objects to infer information in occluded situations, $(iv)$ different radar beam patterns, and, $(v)$ configurations of radar sensors as given by real-world radars. By using ViRa, researchers can simulate radar sensors in different robotic scenarios and investigate radars prior to the installation. This allows an acceleration in the development of radar sensors for robotic applications without the need of real hardware. Contrary to simple model abstractions, which only output loose features, ViRa generates raw radar data using computer graphics techniques on graphics processing unit (GPU) level embedded inside a game engine environment. ViRa allows to feed data directly into machine learning frameworks, which enables further improvement in novel research directions, such as safe human-robot interaction or agile drone flights in obstacle-rich environments. The proposed simulation framework is validated with data from different scenarios in robotics such as human tracking for human-robot interaction. The obtained results are compared with a reference simulation framework and show significantly higher correlation when compared to real-world measurement data.

This paper presents benchmark tests of various visual(-inertial) odometry algorithms on NVIDIA Jetson platforms. The compared algorithms include mono and stereo, covering Visual Odometry (VO) and Visual-Inertial Odometry (VIO): VINS-Mono, VINS-Fusion, Kimera, ALVIO, Stereo-MSCKF, ORB-SLAM2 stereo, and ROVIO. As these methods are mainly used for unmanned aerial vehicles (UAVs), they must perform well in situations where the size of the processing board and weight is limited. Jetson boards released by NVIDIA satisfy these constraints as they have a sufficiently powerful central processing unit (CPU) and graphics processing unit (GPU) for image processing. However, in existing studies, the performance of Jetson boards as a processing platform for executing VO/VIO has not been compared extensively in terms of the usage of computing resources and accuracy. Therefore, this study compares representative VO/VIO algorithms on several NVIDIA Jetson platforms, namely NVIDIA Jetson TX2, Xavier NX, and AGX Xavier, and introduces a novel dataset 'KAIST VIO dataset' for UAVs. Including pure rotations, the dataset has several geometric trajectories that are harsh to visual(-inertial) state estimation. The evaluation is performed in terms of the accuracy of estimated odometry, CPU usage, and memory usage on various Jetson boards, algorithms, and trajectories. We present the {results of the} comprehensive benchmark test and release the dataset for the computer vision and robotics applications., 8 pages, 5 figures

Computing performance and hardware integrity are tightly linked to adequate thermal management in datacenters. In addition, a large share of its electricity bills comes from cooling systems, so there is a pressing need for enhanced thermal management systems. This paper proposes a fiber Bragg grating (FBG) quasi-distributed sensing device as an alternative to electronic sensors. The main focus is to demonstrate that a FBG-based solution allows thermal measurement at critical non-monitored regions (blind spots) within a server in an acquisition rate of 1 Hz, with thermal sensitivity of 8.757 pm/°C, a thermal resolution smaller than 1 °C and spatial resolution and multiplexing needed to monitoring diverse blind spots where there is virtually no room for adapting conventional sensors. In addition, we also monitor spots where there are already internal sensors next to them (referential spots), such as central processing unit (CPU) and graphics processing unit (GPU). As a result, a simple FBG sensor calibration technique is proposed using those referential spots. The technique proposed can be scaled out to monitor a large number of elements across large areas in a datacenter environment, since quasi-distributed FBG sensing signal can be transmitted over longer optical fibers than those employed in optical distributed temperature systems (DTS).

Combined relay selection only requires two relays to forward signals transmitted on multiple subcarriers, but the optimal outage performance is almost surely achievable in the high signal-to-noise ratio (SNR) region. However, because combined relay selection involves the generation of the full set of two-relay combinations, the selection complexity of combined relay selection is much higher than that of per-subcarrier relay selection when the number of relays goes large. This drawback restricts the implementation of combined relay selection in dense networks. To overcome this drawback, we propose to enable combined relay selection by supervised machine learning (ML). Because the training procedure is off-line, the proposed implementation scheme can considerably reduce the selection complexity and the processing latency. We carry out extensive experiments on TensorFlow 2.1 over a graphics processing unit (GPU) aided computing cloud server to validate the effectiveness of the proposed scheme. The experimental results confirm that supervised ML can provide near-optimal performance with lower computing latency that well matches that provided by brute-force search and the optimal relay selection in a per-subcarrier manner.

$Q$ -compensated reverse-time migration ( $Q$ -RTM) has been proven as an efficient method for seismic imaging with high fidelity. However, the source (forward) and receiver (backward) wavefields propagate along the opposite direction of time, and the recursive computation with the out-of-order access requires that all the wavefields of source propagation should be stored on the hard disk. For massive amounts of seismic data, saving the source wavefield from the central processing unit (CPU) [or graphics processing unit (GPU)] device to the disk and loading these data from the hard disk to the CPU (or GPU) device become extremely intensive in time and storage, which has been a bottleneck of $Q$ -RTM. Several methods have been developed to reduce the huge wavefield storage in acoustic media, but are not applicable in the attenuated media. In this letter, we present a reversible hybrid absorbing boundary condition for $Q$ -RTM, which is implemented by mixing the reversible attenuation and the random boundary conditions. Based on our developed new boundary, we just need to save the wavefield at the last one or two time steps in the forward process and then reconstruct the source wavefield in the time-reversal order. Numerical results demonstrate that the method can avoid the huge seismic data input and output (I/O) requirement and improve the computational efficiency dramatically.

The consultative committee for space data system (CCSDS)-123 is a standard for lossless compression of multispectral and hyperspectral images with applications in on-board power-constrained systems, such as satellites and military drones. This letter explores the low-power heterogeneous architecture of the Nvidia Jetson TX2 by proposing a parallel solution to the CCSDS-123 compressor on embedded systems, reducing development effort compared with the production of dedicated circuits, while maintaining low energy consumption. This solution parallelizes the predictor on a low-power graphics processing unit (GPU) while the encoders exploit the heterogeneous multiple cores of the CPUs and GPU concurrently. We report more than 16.6 Gb/s for the predictor and 1.4-Gb/s for the whole system, requiring less than 6.3 W and providing an efficiency of 245.6 Mb/s/W.

Highly-detailed equipment models for electromagnetic transient simulation provide an accurate insight into the system characteristics and behavior. In this article, a coupled field-circuit co-simulation employing detailed component-level models is proposed for the solid-state transformer. To reveal comprehensive thermo-electromagnetic information of the equipment, a high-order nonlinear insulated-gate bipolar transistor (IGBT) model is utilized for the modular multilevel converter, while the finite element method (FEM) is adopted in modeling the transformer. The heavy computational challenge posed by the complexity of these models is alleviated by exploiting model parallelism and the subsequent processing by massively parallel architecture of the graphics processing unit, e.g., a pair of coupled voltage-current sources is adopted for reducing the order of the matrix equation in the circuit part, while in the FEM-based models, a matrix-free nodal domain decomposition solution is utilized to parallelize the overall system to the maximum. A multi-rate scheme is applied for a further computational burden reduction of the co simulation due to a large disparity in the appropriate time-steps between power semiconductor switches and the magnetic component. Simulation of a multiterminal dc system including the SST is carried out, and the accuracy of proposed models are validated by offline tools such as SaberRD, ANSYS, and PSCAD/EMTDC

In this paper, a three-dimensional universal cooperative localizer (3D UCL) is proposed for vehicular ad-hoc networks (VANETs) in 3D space under varied types of ranging measurements including time-of-arrival (TOA), received signal strength (RSS), angle-of-arrival (AOA), and Doppler frequency. Its core idea is to exploit generalized approximate message passing (GAMP) to resolve the 3D cooperative positioning problem after converting it as a generalized linear mixing problem. Unfortunately, the positioning performance of 3D UCL is severely degraded by the inaccurate ranging measurements from the non-line-of-sight (NLOS) links. Therefore, a 3D geographical information enhanced UCL (3D GIE-UCL) is developed by combining 3D UCL with a NLOS identification mechanism assisted by geographical information. Finally, 3D UCL is accelerated by graphics processing unit (GPU) parallelization, particle reduction and message censoring. 3D GIE-UCL is accelerated by particle reduction and anchor upgrading. Simulation results validate state-of-the-art positioning performances and cooperative gains of both 3D UCL and 3D GIE-UCL after comparing them with existing cooperative localizers. 3D GIE-UCL approaches to its performance upper bound provided by its correspondence with oracle link-type information. 3D UCL and GIE-UCL show $241\times $ and $3.3\times $ speedup after adopting the acceleration techniques, respectively.

Orbital angular momentum-shift keying (OAM-SK), which is the rapid switching of OAM modes, is vital but seriously impeded by the deficiency of OAM demodulation techniques, particularly when videos are transmitted over the system. Thus, in this paper, 3D chaotic interleaved multi-coded video frames (VFs) are conveyed via an N-OAM-SK free-space optical (FSO) communication system to enhance the reliability and efficiency of video communication. To tackle the defects of the OAM-SK-FSO mechanism, two efficient deep learning (DL) techniques, namely convolution recurrent neural network (CRNN) and 3D convolution neural network (3DCNN) are used to decode OAM modes with a low bit error rate (BER). Moreover, a graphics processing unit (GPU) is used to accelerate the training process with slight power consumption. The utilized datasets for OAM states are generated by applying different scenarios using a trial-and-error method. The simulation results imply that LDPC-coded VFs achieve the largest peak signal-to-noise ratios (PSNRs) and the lowest BERs using the 16-OAM-SK model. Both 3DCNN and CRNN techniques have nearly the same performance, but this performance deteriorates in the case of larger dataset classes. Moreover, the GPU accelerates the performance by almost 67.6% and 36.9% for the CRNN and 3DCNN techniques, respectively. These two DL techniques are more effective in evaluating the classification accuracy than the other traditional techniques by almost 10 – 20%.

A digitized analog signal often encounters a high-frequency noisy background which degrades the signal-to-noise ratio (SNR) particularly in case of low signal strength. Despite quite a lot of hardware- and software-based approaches have been reported to date to deal with the noise issue, it is still a challenging task to real-time retrieve the noise-contaminated low-frequency information efficiently without degrading the original bandwidth. In this paper, we report a modified unsharp-masking (UM)-based Graphics Processing Unit (GPU)-accelerated algorithm to efficiently suppress a high-frequency noisy background in a digitized two-dimensional image. The proposed idea works effectively even if noise-density is high and signal of interest is comparable or weaker than the maximum noise level. While suppressing the noisy background, the original resolution remains least compromised. We first explore the effectiveness of the algorithm by means of simulated images and subsequently extend our demonstration towards a real-world life-science imaging application. Securing a potential for real-time applicability, we implement the algorithm via Compute Unified Device Architecture (CUDA)-acceleration and preserve a $ < 300~\mu \text{s}$ processing time for a $1000\times 1000$ -sized 8-bit data set.

Deep learning (DL) algorithms have been demonstrated to have great potential in hyperspectral image (HSI) classification. However, most DL methods mainly focus on improving classification performance, neglecting the computational cost. In order to broaden the application scenarios of DL-based HSI classification methods, it is necessary to develop lightweight and fast models to fit the deployment needs of computational-resource-limited platforms. Taking into account the efficiency and accuracy, this article designs a lightweight network architecture based on spectral and channel-wise attention modules, namely LiteSCANet, for HSI classification. It contains a residual double-branch structure, which makes the model effectively extract spectral-spatial features and achieve good performance with fast inference speed and low computational consumption (i.e., floating-point operations, number of parameters, and graphics processing unit memory usage). The experiment results on four benchmark datasets show that our proposed model achieves an excellent trade-off between efficiency and accuracy compared with the other six existing networks.

This article presents an automatic real-time object detection method using sidescan sonar (SSS) and an onboard graphics processing unit (GPU). The detection method is based on a modified convolutional neural network (CNN), which is referred to as self-cascaded CNN (SC-CNN). The SC-CNN model segments SSS images into object-highlight, object-shadow, and seafloor areas, and it is robust to speckle noise and intensity inhomogeneity. Compared with typical CNN, SC-CNN utilizes crop layers which enable the network to use local and global features simultaneously without adding convolution parameters. Moreover, to take the local dependencies of class labels into consideration, the results of SC-CNN are postprocessed using Markov random field. Furthermore, the sea trial for real-time object detection via the presented method was implemented on our autonomous underwater vehicle (AUV) named SAILFISH via its GPU module at Jiaozhou Bay Bridge, Qingdao, China. The results show that the presented method for SSS image segmentation has obvious advantages when compared with the typical CNN and unsupervised segmentation methods, and is applicable in real-time object detection task.

This article demonstrates a spectroscopic diffuse optical tomography (DOT) system for the noninvasive continuous imaging of tissue. The system was built on an embedded system architecture along with an algorithm that performs entire system control, data acquisition, noise-free lock-in detection, and wireless data transmission to a host computer, equipped with a graphics processing unit (GPU). The GPU system executes a fast 3-D DOT spectroscopic image reconstruction algorithm. The algorithm computes Jacobians and solves four different depth-sensitive inverse problems of image reconstruction. The host computer sends commands to the DOT server system and collects data, and simultaneously performs high-speed 3-D spectroscopic image reconstruction to display images in real-time. We designed a configurable optical fiber patch to collect depth-resolved diffused light exiting the imaging media by placing the patch on the surface. The simulation and experimental results from two setups showed the spectroscopic imaging capability of the system to map the concentration of oxyhemoglobin, deoxyhemoglobin, and water, and to characterize the tumor located deep inside the tissue. The proposed system uses low-cost LEDs, photodiode detectors, embedded system, and algorithms to perform high-speed 3-D functional imaging and can be further miniaturized. Our promising results pave the way for the development of a low-cost hand-held spectroscopic DOT system for the real-time onsite functional imaging of biological tissue.

Using multiple unmanned aerial vehicles (UAVs) improves the efficiency of reconnaissance, surveillance, and search and rescue missions. This article presents a path-planning software for a team of UAVs utilizing graphics processing units (GPUs). The UAVs are tasked to visit multiple points of interest (POIs) in a 3-D environment, and the software finds an optimized solution that assigns the POIs to the UAVs, selects the order in which the POIs are visited, and calculates the paths between the POIs. The software uses a multistep approach using a single-source-shortest-path algorithm to find the optimal paths between all combinations of POIs followed by a genetic algorithm to solve the multitraveling salesperson problem. The algorithm can minimize distance, time, or energy consumption depending on the setting selected by the user. The proposed GPU implementation is tested on eight different maps from around the world and executes in just 0.6 s, a $48.3\times $ speedup compared to a sequential execution on CPU. This performance improvement is a real asset in a mission-changing environment.

Over the past few years, 2-D convolutional neural networks (CNNs) have demonstrated their great success in a wide range of 2-D computer vision applications, such as image classification and object detection. At the same time, 3-D CNNs, as a variant of 2-D CNNs, have shown their excellent ability to analyze 3-D data, such as video and geometric data. However, the heavy algorithmic complexity of 2-D and 3-D CNNs imposes a substantial overhead over the speed of these networks, which limits their deployment in real-life applications. Although various domain-specific accelerators have been proposed to address this challenge, most of them only focus on accelerating 2-D CNNs, without considering their computational efficiency on 3-D CNNs. In this article, we propose a unified hardware architecture to accelerate both 2-D and 3-D CNNs with high hardware efficiency. Our experiments demonstrate that the proposed accelerator can achieve up to 92.4% and 85.2% multiply-accumulate efficiency on 2-D and 3-D CNNs, respectively. To improve the hardware performance, we propose a hardware-friendly quantization approach called static block floating point (BFP), which eliminates the frequent representation conversions required in traditional dynamic BFP arithmetic. Comparing with the integer linear quantization using zero-point, the static BFP quantization can decrease the logic resource consumption of the convolutional kernel design by nearly 50% on a field-programmable gate array (FPGA). Without time-consuming retraining, the proposed static BFP quantization is able to quantize the precision to 8-bit mantissa with negligible accuracy loss. As different CNNs on our reconfigurable system require different hardware and software parameters to achieve optimal hardware performance and accuracy, we also propose an automatic tool for parameter optimization. Based on our hardware design and optimization, we demonstrate that the proposed accelerator can achieve 3.8-5.6 times higher energy efficiency than graphics processing unit (GPU) implementation. Comparing with the state-of-the-art FPGA-based accelerators, our design achieves higher generality and up to 1.4-2.2 times higher resource efficiency on both 2-D and 3-D CNNs.

This article presents for the first time an OpenACC (Open accelerators)-aided Graphics Processing Unit (GPU)-based approach adopting a 3D moving window finite difference time domain (MW-FDTD) method for calculating lightning electromagnetic fields over large-scale terrains. The efficacy and accuracy of the proposed method are evaluated by comparing its results with the results obtained from a conventional FDTD method. The results show that the GPU-based MW-FDTD method is capable of providing highly accurate results 42 times faster than the conventional CPU-based FDTD method. Moreover, the proposed method required only $\left({\frac {2{M}}{N_{x}}}\right)\times 100=40\%$ . of the memory needed for carrying out the calculation compared to a conventional FDTD method, where ${M}$ and ${ N}_{ x}$ are the size of the entire domain in the conventional FDTD method and size of the divided blocks in the MW-FDTD method domain, respectively. The results show the applicability of the MW-FDTD-GPU method for dealing with lightning electromagnetic problems as well as large-scale electromagnetic compatibility (EMC) applications where the efficiency and minimization of the computational time and memory requirements are of great importance.

Flight simulators based on wrap-around displays provide an immersive experience for pilot training, but not only are they expensive and bulky, they typically only simulate a single aircraft configuration. Virtual reality (VR) is an interesting alternative, providing an immersive experience with lower cost and space requirements, as well as the ability to simulate arbitrary aircraft. Unfortunately, current VR systems do not provide natural tactile interaction with instruments and avionics in the cockpit environment. This article explores augmented reality (AR) as an alternative to VR and wrap-around display flight simulators, delivering an immersive experience for pilot training and tactile interaction with aircraft controls. The concept is demonstrated through a basic general aviation simulator setup, showing that a relatively low-cost personal computer and graphics processing unit can provide AR simulations at useful frame rates.

Despite the approaching Internet of Things (IoT) era, most smart devices cannot directly execute complicated deep neural networks (DNNs) due to their limited memory sizes and computing power. Even with model compression, compressing and executing a non-binarized DNN model often needs a codebook for weight calculation, which typically requires acceleration from a Graphics Processing Unit (GPU). Parameter binarization is one commonly used method to reduce computing time and space, but it often seriously compromises the accuracy of the binarized DNN (BNN), which tends to adopt single-stage non-modularized design. To address these issues, we propose a modularized multi-stage BNN model compression with flexible combination of parameter and structure compression to minimize model size while maintaining maximum accuracy possible. The compressed BNNs do not need any codebooks so they better fit into embedded platforms without any GPU acceleration. From our evaluations, the size of a BNN model can be reduced to 1/57 size at the cost of only about a 5% decrease in accuracy, which outperforms the state-of-the-art single-stage compression on BNN models. Furthermore, the proposed method can recover the accuracy of a BNN by 5% even after its size has been further compressed to approximately half its size, demonstrating its potentials in leveraging edge intelligence.

Recent support for unified memory and demand paging has improved graphics processing unit (GPU) programmability and enabled memory oversubscription. However, this support introduces high overhead when page faults occur. Therefore, when the GPU memory fills to capacity, an important issue is how to select eviction candidates. The widely used policy, LRU, and the advanced replacement policies, RRIP and CLOCK-Pro, suffer from inefficiency when dealing with thrashing access patterns. They also incur significant overhead due to managing metadata at page level. In this article, we propose hierarchical page eviction (HPE), a new replacement policy for GPUs with unified memory. Aided by page walk hit information, HPE manages a page set chain dynamically. It uses statistics to classify applications into three categories and selects an appropriate eviction strategy for each category. It also applies dynamic adjustment to switch the eviction strategy when necessary. The simulation results show that, on average, HPE achieves $1.34\times $ and $1.16\times $ speedup (up to $2.81\times $ ) over LRU when the oversubscription rate is 75% and 50%, respectively. HPE also outperforms RRIP and CLOCK-Pro.

To fully exploit hardware resources and maximize the throughput of the heterogeneous systems, a graphics processing unit (GPU)+CPU collaborative computing method for the finite-difference time-domain (FDTD) method scattering simulation is proposed in this article. For the workload imbalance caused by different computational performances and device-module affinity of different processors in the heterogeneous systems, a dynamic load balancing method is proposed, which can distribute parallel tasks intelligently according to the captured runtime performance index of processors. The method proposed herein can be applied to different heterogeneous hardware environments adaptively, owing to its online adjustment of workload distribution. Simulation results show that the proposed dynamic load balancing-based heterogeneous parallel FDTD method does not introduce additional errors in solving the scattering problems and has a great improvement in efficiency compared to a parallel implementation using a single kind of processor. It shows up to $56.76\times $ and $5.85\times $ speedups over the multicore CPU and multi-GPU parallel execution, respectively, which is quite an attractive improvement for practical applications.

Deepfakes, or artificially generated audiovisual renderings, can be used to defame a public figure or influence public opinion. With the recent discovery of generative adversarial networks, an attacker using a normal desktop computer fitted with an off-the-shelf graphics processing unit can make renditions realistic enough to easily fool a human observer. Detecting deepfakes is thus becoming important for reporters, social media platforms, and the general public. In this work, we introduce simple, yet surprisingly efficient digital forensic methods for audio spoof and visual deepfake detection. Our methods combine convolutional latent representations with bidirectional recurrent structures and entropy-based cost functions. The latent representations for both audio and video are carefully chosen to extract semantically rich information from the recordings. By feeding these into a recurrent framework, we can detect both spatial and temporal signatures of deepfake renditions. The entropy-based cost functions work well in isolation as well as in context with traditional cost functions. We demonstrate our methods on the FaceForensics++ and Celeb-DF video datasets and the ASVSpoof 2019 Logical Access audio datasets, achieving new benchmarks in all categories. We also perform extensive studies to demonstrate generalization to new domains and gain further insight into the effectiveness of the new architectures.

NIST (National Institute of Standards and Technology) statistical test recognized as the most authoritative is widely used in verifying the randomness of binary sequences. The Non-overlapping Template Matching Test as the 7th test of the NIST Test Suit is remarkably time consuming and the slow performance is one of the major hurdles in the testing process. In this paper, we present an efficient bit-parallel matching algorithm and segmented scan-based strategy for execution on Graphics Processing Unit (GPU) using NVIDIA Compute Unified Device Architecture (CUDA). Experimental results show the significant performance improvement of the parallelized Non-overlapping Template Matching Test, the running speed is 483 times faster than the original NIST implementation without attenuating the test result accuracy.

Memory access latencies and low data transfer bandwidth limit the processing speed of many data intensive applications such as Convolutional Neural Networks (CNNs) in conventional Von Neumann architectures. Processing in Memory (PIM) is envisioned as a potential hardware solution for such applications as the data access bottlenecks can be avoided in PIM by performing computations within the memory die. However, PIM realizations with logic-based complex processing units within the memory present complicated fabrication challenges. In this letter, we propose to leverage the existing memory infrastructure to implement a programmable PIM (pPIM), a novel Look-Up-Table (LUT)-based PIM where all the processing units are implemented solely with LUTs, as opposed to prior LUT-based PIM implementations that combine LUT with logic circuitry for computations. This enables pPIM to perform ultra-low power & low-latency operations with minimal fabrication complications. Moreover, the complete LUT-based design offers simple ‘memory write’ based programmability in pPIM. Enabling precision scaling further improves the performance and the power consumption for CNN applications. The programmability feature potentially makes it easier for online training implementations. Our preliminary simulations demonstrate that our proposed pPIM can achieve 2000x, 657.5x and 1.46x improvement in inference throughput per unit power consumption compared to state-of-the-art conventional processor architecture, Graphics Processing Unit (GPUs) and a prior hybrid LUT-logic based PIM respectively. Furthermore, precision scaling improves the energy efficiency of the pPIM approximately by 1.35x over its full-precision operation.

Data completion is a problem of filling missing or unobserved elements of partially observed datasets. Data completion algorithms have received wide attention and achievements in diverse domains including data mining, signal processing, and computer vision. We observe a ubiquitous tubal-sampling pattern in big data and Internet of Things (IoT) applications, which is introduced by many reasons such as high data acquisition cost, downsampling for data compression, sensor node failures, and packet losses in low-power wireless transmissions. To meet the time and accuracy requirements of applications, data completion methods are expected to be accurate as well as fast. However, the existing methods for data completion with the tubal-sampling pattern are either accurate or fast, but not both. In this article, we propose high-performance graphics processing unit (GPU) tensor completion for data completion with the tubal-sampling pattern. First, by exploiting the convolution theorem, we split a tensor least-squares minimization problem into multiple least-squares sub-problems in the frequency domain. In this way, massive parallelisms are exposed for many-core GPU architectures while still preserving high recovery accuracy. Second, we propose computing slice-level and tube-level tasks in batches to improve GPU utilization. Third, we reduce the data transfer cost by eliminating the accesses to the CPU memory inside algorithm loop structures. The experimental results show that the proposed tensor completion is both fast and accurate. Using synthetic data of varying sizes, the proposed GPU tensor completion achieves maximum 248.18×, 7, 403.27×, and 33.27× speedups over the CPU MATLAB implementation, GPU element-sampling tensor completion in the cuTensor-tubal library, and GPU high-performance matrix completion, respectively. With a 50 percent sampling rate, the proposed GPU tensor completion achieves a recovery error of 1.40e-5, which is comparable with that of the GPU element-sampling tensor completion and three orders of magnitude better than that of the GPU high-performance matrix completion. To utilize multiple GPUs in servers, we design a multi-GPU scheme for tubal-sampling tensor completion. The multi-GPU tensor completion achieves maximum 1.89× speedup on two GPUs versus on a single GPU for medium or big tensors. We further evaluate the performance of the proposed GPU tensor completion in three real applications, namely, video transmission in wireless camera networks, RF fingerprint-based indoor localization, and seismic data completion, and it achieves maximum speedups of 448.68×, 24.63×, and 311.54×, respectively. We integrate this high-performance GPU tensor completion implementation into the cuTensor-tubal library to support various applications.

In this paper, we present a graphics processing unit (GPU)-based implementation of a weighted bit-reliability based (wBRB) decoder for non-binary LDPC (NB-LDPC) codes. To achieve coalesced memory accesses, an efficient data structure for the wBRB algorithm is proposed. Based on the Single-Instruction Multiple-Threads (SIMT) programming model, a novel mapping strategy with high intra-frame parallelism is presented to improve the latency and throughput performance. Moreover, by using Single-Instruction Multiple-Data (SIMD) intrinsics, four 8-bit message elements are packed into a 32-bit unit and simultaneously processed. Experimental results show that the proposed wBRB decoder provides good tradeoff between error performance and throughput for the codes with relatively large column degrees or high rates.

An efficient facet-based composite scattering model (FBCSM) is developed for calculating the time-evolving electromagnetic (EM) scattering field (TESF) to study the normalized radar cross section and Doppler spectrum characteristics from dynamic sea surfaces. The dynamic sea surface comprises two-scale profiles: small-scale capillary ripples modulated by large-scale gravity waves, which are modeled by millions of small facets. In microwave bands, two scattering mechanisms, quasi-specular scattering with respect to gravity waves and Bragg scattering with respect to ripples, are taken into account in the FBCSM for computation of the time-evolving EM scattering field under diverse polarizations. However, it may be very time-consuming and difficult to calculate the TESF due to the high resolution and dynamic complexity of the large dynamic sea surface. In this paper, the NVIDIA Tesla K80 graphics processing unit (GPU) with the compute unified device architecture is utilized to improve the computational performance of the TESF. The whole GPU-based TESF computation includes the optimal use of temporary variables, shared memory, constant memory and register, fast-math compiler options, asynchronous data transfer, and the most suitable block size and number of registers. By utilizing the proposed five improvement strategies, a significant speedup of $1200 \times $ can be achieved for computation of TESF from large dynamic sea surfaces for microwave bands compared with the single-threaded C program executed on the Intel(R) Core(TM) i5-3450 CPU.

In past robotics applications, Model Predictive Control (MPC) has often been limited to linear models and relatively short time horizons. In recent years however, research in optimization, optimal control, and simulation has enabled some forms of nonlinear model predictive control which find locally optimal solutions. The limiting factor for applying nonlinear MPC for robotics remains the computation necessary to solve the optimization, especially for complex systems and for long time horizons. This letter presents a new solution method which addresses computational concerns related to nonlinear MPC called nonlinear Evolutionary MPC (NEMPC), and then we compare it to several existing methods. These comparisons include simulations on torque-limited robots performing a swing-up task and demonstrate that NEMPC is able to discover complex behaviors to accomplish the task. Comparisons with state-of-the-art nonlinear MPC algorithms show that NEMPC finds high quality control solutions very quickly using a global, instead of local, optimization method. Finally, an application in hardware (a 24 state pneumatically actuated continuum soft robot) demonstrates that this method is tractable for real-time control of high degree of freedom systems.

Most lightning location networks are based on real-time analytical solutions of certain simplified models, while the reality is much more complicated. In this paper, we introduce a graphics processing unit (GPU)-based parallel computing algorithm that can extensively benefit lightning geolocation networks. For a network running this GPU-based algorithm, one can build up a geolocation database based on numerical solutions of certain complete models in advance, and lightning geolocations can then be easily determined with a grid-searching technique in real time. One such grid-searching technique, is the grid traverse algorithm (GTA) for the traditional time of arrival technique. By running GPU-based GTA in a six-station two-dimensional (2-D) and a five-station 3-D networks, we show that extremely high network performance can be achieved, with a processing speed of about 2700 times faster than CPU-based GTA. The location accuracy of GPU-GTA is examined with Monte Carlo simulations, showing that GPU-GTA can locate a lightning source in real time with high accuracy. We also find that when the grid step is comparable with the inherent time uncertainty of a network, the location accuracy cannot be improved further with a finer grid step.

Proton beam therapy has the advantages over other forms of radiation therapy because it does less damage to healthy tissue while treating cancerous tissue. These advantages arise from the high dose deposition at the end of the proton beam which is known as the Bragg peak. This paper proposes a new architecture for graphics processing unit (GPU)-accelerated stochastic origin ensemble (SOE) algorithm, eventually achieves a great acceleration and allows to reconstruct images of the gamma radiation produced by proton beam in real time. The object of this paper is to improve the run time of SOE algorithm for improving the accuracy of the proton therapy. In previous studies, it is difficult to reconstruct images of prompt gamma (PG) ray in real time in proton therapy. After analyzing the structure of the SOE algorithm, we correspond each gamma ray detected by Compton camera (CC) to each thread in GPU, initialize the origin cones in GPU, and use data structures which are suitable for parallelization, shared memory, and random number lists to speed up parallel program. To evaluate the proposed method, we use the parallel SOE algorithm to reconstruct the sources with the CC imaging system, which is simulated by Geant4. The parallel program on the GPU eventually achieves a $25\times $ speedup over serial CPU run, reconstructing images of gamma radiation in 0.4 s. In the end, we also accelerate the Monte Carlo-based back projection (MC-BP) algorithm, the parallel program achieves a $79\times $ speedup over serial CPU run.

The extended use of mobile multimedia devices in applications like gaming, 3D video and audio reproduction, immersive teleconferencing, or virtual and augmented reality, is demanding efficient algorithms and methodologies. All these applications require real-time spatial audio engines with the capability of dealing with intensive signal processing operations while facing a number of constraints related to computational cost, latency and energy consumption. Most mobile multimedia devices include a Graphics Processing Unit (GPU) that is primarily used to accelerate video processing tasks, providing high computational capabilities due to its inherent parallel architecture. This paper describes a scalable parallel implementation of a real-time binaural audio engine for GPU-equipped mobile devices. The engine is based on a set of head-related transfer functions (HRTFs) modelled with a parametric parallel structure, allowing efficient synthesis and interpolation while reducing the size required for HRTF data storage. Several strategies to optimize the GPU implementation are evaluated over a well-known kind of processor present in a wide range of mobile devices. In this context, we analyze both the energy consumption and real-time capabilities of the system by exploring different GPU and CPU configuration alternatives. Moreover, the implementation has been conducted using the OpenCL framework, guarantying the portability of the code. This work was supported in part by the Spanish Ministry of Economy and Competitiveness under Grant RTI2018-097045-B-C21, Grant RTI2018-097045-B-C22, Grant TIN2017-82972-R, and Grant ESP2015-68245-C4-1-P, and in part by the Universitat Jaume I Project UJI-B2019-36.

Modern Graphics Processing Units (GPUs) require large hardware resources for massive parallel thread executions. In particular, modern GPUs have a large register file composed of Static Random Access Memory (SRAM). Due to the high leakage current of SRAM, the register file consumes approximately 20% of the total GPU energy. The energy efficiency of the register file becomes more critical as the throughput of GPUs increases. For more energy-efficient GPUs, the usage of non-volatile memory such as Spin-Transfer Torque Magnetic Random Access Memory (STT-MRAM) as the GPU register file has been studied extensively. STT-MRAM requires a lower leakage current compared to SRAM and provides an appropriate read performance. However, using STT-MRAM directly in the GPU register file causes problems in performance and endurance because of complicated write procedures and material characteristics. To overcome these challenges, we propose a novel register file architecture and its management system for GPUs, named Hi-End, which exploits the data locality and compressibility of the register file. For STT-MRAM-based GPU register files, Hi-End increases the data write performance and endurance by caching and data compression, respectively. In our evaluation, Hi-End enhances the energy efficiency of a GPU register file by 70.02% and reduces the write operations by up to 95.98% with negligible performance degradation compared to SRAM-based register files.

The traditional surgical navigation system combining preoperative CT and intraoperative ultrasound is widely used in open hepatectomy, but the problem of this system is that there are some errors in terms of the precision of the time and the space during the surgery. In order to solve the above problems, this paper introduces augmented reality technology into the surgical navigation system. In order to accurately describe the biomechanical characteristics of the liver and let the navigation system perform more accurately, this paper uses Tetgen to perform tetrahedral partition on the triangle mesh data obtained after the three-dimensional CT was reconstructed, and then the surface mesh data and the internal tetrahedral data were obtained and they were used to describe a whole liver model. The surface triangle mesh data is used to render graphics and describe the surface topology change, and the internal tetrahedral data combined with mass-spring theory is used to simulate deformation. Subsequently, the ex vivo pig liver was used to experimentally verify the accuracy of gravity deformation of the liver model, and the results show that the error was mainly distributed between −2mm and −2.5mm. At the same time, this paper uses the NDI Polaris infrared tracking system to carry out precision experiments on the augmented reality module, and the measured error is 1.55± 0.29mm. Finally, various modules of the system are integrated to finish the experiment in which the ring-shaped lesions are cut from the ex vivo pig liver with the aid of augmented reality. The experimental error is 0± 1.26mm, and with the assistance of the general purpose graphics processing unit (GPGPU), the refresh rate is above 200FPS. The results prove that the liver surgery navigation system proposed in this paper is excellent in terms of real-time performance and accuracy, which can help doctors accurately locate the tumor during surgery and perform ideal resection.

The energy efficiency of accelerating hundreds of MB-large deep neural networks (DNNs) in a mobile environment is less than that of a server-class big chip accelerator because of the limited power budget, silicon area, and smaller buffer size of static random access memory associated with mobile systems. To address this challenge and provide powerful computing capability for processing large DNN models in power/resource-limited mobile systems, we propose McDRAM v2, which is a novel in-dynamic random access memory (DRAM) systolic array accelerator architecture. McDRAM v2 makes the best use of large in-DRAM bandwidths for accelerating various DNN applications. It can handle large DNN models without off-chip memory accesses, in a fast and efficient manner, by exposing the large DRAM capacity and large in-DRAM bandwidth directly to an input systolic array of a processing element matrix. Additionally, it maximizes data reuse using a systolic multiply-accumulate (MAC) structure. The proposed architecture maximizes the utilization of large-scale MAC units by judiciously exploiting the DRAM's internal bus and buffer structure. An evaluation of large DNN models in the fields of image classification, natural language processing, and recommendation systems shows that it achieves 1.7 times tera operations per second (TOPS), 3.7 times TOPS/watt, and 8.6 times TOPS/mm2 improvements over a state-of-the-art mobile graphics processing unit accelerator, and 4.1 times better energy efficiency over a state-of-the-art server-class accelerator. Moreover, it incurs a minimal overhead, i.e., a 9.7% increase in area, and uses less than 4.4 W of peak operating power.

N-1 static security analysis (SSA) is an important method for power system stability analysis that requires solving N alternating-current power flows (ACPF) for a system with N elements to obtain strictly accurate results. Past researches have shown the potential of accelerating these calculations using an iterative solver with graphics processing unit (GPU). This paper proposes a GPU-based N-1 SSA algorithm with the preconditioned conjugate gradient (PCG) method. First, a shared preconditioner is selected to accelerate preprocessing of the iterative method for fast decoupled power flow (FDPF) in N-1 SSA. Second, it proposes a GPU-based batch-PCG solver, which packages a massive number of PCG subtasks into a large-scale problem to achieve a higher degree of parallelism and better coalesced memory accesses. Finally, the paper presents a novel GPU-accelerated batch-PCG solution for N-1 SSA. Case studies on a practical 10828-bus system show that the GPU-based N-1 SSA algorithm with the batch-PCG solver is 4.90 times faster than a sequential algorithm on an 8-core CPU. This demonstrates the potential of the GPU-based high-performance SSA solution with the PCG method under a batch framework.

The demand for on-line analyzing of internet traffic for both security and QoS consideration directly increases as a function of using diverse applications and as malicious attacks increase. This paper presents a new fast parallel packet classification algorithm based on entropy hashing. The algorithm uses a one-level hashing data structure and enables partitioning a very large rules-set into smaller uniformly distributed sub-rules look-up tables based on maximum entropy and Most Significant Bit (MSB) pattern hash keys. This minimizes the classifier searches only to the relevant appropriate look-up table using the same hash key, and therefore significantly shortens the classification process. A further speed-up factor is achieved by parallelizing the classification algorithm using Nvidia Graphics Processing Unit (GPU). The proposed algorithm is applied to both ACL and FW applications using common synthetic rules-sets of size up to 500k rules. The simulation results show that the proposed algorithm outperforms existing classifiers in terms of both speed up and memory utilization. The required memory size is significantly reduced, and a classification speed-up factor of up to 200 is demonstrated compared to a similar serial approach.

Partial registration for point clouds plays an important role in various fields such as 3D mapping reconstruction, remote sensing, unmanned driving, and cultural heritage protection. Unfortunately, partial registration is challenging due to difficulties such as the low overlap ratio of two point clouds and the perturbation in the orderless and sparse 3D point clouds. Thus, a variety of the 3D shape context descriptors are introduced for finding the optimal matching. However, extracting geometric features and descriptors are time consuming and easily degenerated by noise. To overcome these problems, we introduce a parallel coarse-to-fine partial registration method. Our contributions can be summarized as: Firstly, a robust coarse trimmed method is proposed to estimate the coarse overlap area and the initial transformation via fast bilateral denoising and parallel point feature histogram (PPFH) descriptor aligning. Secondly, an accelerated fine registration procedure is conducted by a parallel trimmed iterative closest point (PTrICP) method. Moreover, most parts of our coarse-to-fine workflow are accelerated under the Graphics Processing Unit (GPU) parallel execution mode for efficiency. Thirdly, we extend our method from the rigid registration to the isotropic scaling registration, which improves its applicability. Experiments have demonstrated that our method is feasible and robust in various situations, including the low overlap ratio, outlier, noise and scaling.