Hybrid Hardware Research Articles

Hybrid neuromorphic hardware with sparing 2D synapse and CMOS neuron for character recognition

Neuromorphic computing enables efficient processing of data-intensive tasks, but requires numerous artificial synapses and neurons for certain functions, which leads to bulky systems and energy challenges. Achieving functionality with fewer synapses and neurons will facilitate integration density and computility. Two-dimensional (2D) materials exhibit potential for artificial synapses, including diverse biomimetic plasticity and efficient computing. Considering the complexity of neuron circuits and the maturity of complementary metal–oxide–semiconductor (CMOS), hybrid integration is attractive. Here, we demonstrate a hybrid neuromorphic hardware with 2D MoS2 synaptic arrays and CMOS neural circuitry integrated on board. With the joint benefit of hybrid integration, frequency coding and feature extraction, a total cost of twelve MoS2 synapses, three CMOS neurons, combined with digital/analogue converter enables alphabetic and numeric recognition. MoS2 synapses exhibit progressively tunable weight plasticity, CMOS neurons integrate and fire frequency-encoded spikes to display the target characters. The synapse- and neuron-saving hybrid hardware exhibits a competitive accuracy of 98.8% and single recognition energy consumption of 11.4 μW. This work provides a viable solution for building neuromorphic hardware with high compactness and computility.

A Multiformalism-Based Model for Performance Evaluation of Green Data Centres

Although the coexistence of ARM and INTEL technologies in green data centres is technically feasible, significant challenges exist that must be addressed. These challenges stem from the differences in instruction sets and power consumption between the two processor architectures. While ARM processors are known for their energy efficiency, INTEL processors tend to consume more power. Consequently, evaluating the performance of hybrid architectures can be a complex task. The contributions of this article consist of (i) a multiformalism-based model of a data centre, providing a natural and convenient approach to the specification process and performance analysis of a realistic scenario and (ii) a review of the performance indices, including the choice of one architecture over another, power consumption, the response time, and request loss, according to different policies. As a result, the model aims to address issues such as system underutilization and the need to estimate the optimal workload balance, thereby providing an effective solution for evaluating the performance of hybrid hardware architectures.

Embedded solar tracking system using arduino


 
 
 This paper describes a systematic approach to building an intelligent solar Tracking system (ISTS), for improving the performance of solar panels. The ISTS is a hybrid hardware / software prototype, which automatically provides best alignment of solar panel with the sun, to get maximum output (electricity). This method is increasing power collection efficiency by developing a device that tracts the sun to keep the panel at a right angle to its rays.The Embedded solar tracking instrumentation system by using microcontroller. The system consists of Light Dependent Resistor (LDR) sensor, DC motor .Arduino UNO microcontroller is the main component for controlling the system. The solar system will track the location of the sun to ensure the solar panel is always perpendicular with the sun therefore optimizing power output. The operation of the system on sunny and bad weather condition has been presented in this paper. The solar tracking prototype has been stated for future works.
 
 

Horizontal axis wind turbine


 
 
 This paper describes a systematic approach to building an intelligent solar Tracking system (ISTS), for improving the performance of solar panels. The ISTS is a hybrid hardware / software prototype, which automatically provides best alignment of solar panel with the sun, to get maximum output (electricity). This method is increasing power collection efficiency by developing a device that tracts the sun to keep the panel at a right angle to its rays.The Embedded solar tracking instrumentation system by using microcontroller. The system consists of Light Dependent Resistor (LDR) sensor, DC motor .Arduino UNO microcontroller is the main component for controlling the system. The solar system will track the location of the sun to ensure the solar panel is always perpendicular with the sun therefore optimizing power output. The operation of the system on sunny and bad weather condition has been presented in this paper. The solar tracking prototype has been stated for future works.
 
 

Hybrid FPGA-CPU-Based Architecture for Object Recognition in Visual Servoing of Arm Prosthesis.

The present paper proposes an implementation of a hybrid hardware–software system for the visual servoing of prosthetic arms. We focus on the most critical vision analysis part of the system. The prosthetic system comprises a glass-worn eye tracker and a video camera, and the task is to recognize the object to grasp. The lightweight architecture for gaze-driven object recognition has to be implemented as a wearable device with low power consumption (less than 5.6 W). The algorithmic chain comprises gaze fixations estimation and filtering, generation of candidates, and recognition, with two backbone convolutional neural networks (CNN). The time-consuming parts of the system, such as SIFT (Scale Invariant Feature Transform) detector and the backbone CNN feature extractor, are implemented in FPGA, and a new reduction layer is introduced in the object-recognition CNN to reduce the computational burden. The proposed implementation is compatible with the real-time control of the prosthetic arm.

Implementing evolutionary optimization on actual quantum processors

This paper introduces a new evolutionary algorithm with the support of an actual quantum processor, a computing device which uses phenomena from quantum mechanics to enable a considerable speed-up in computation. In particular, the proposed approach uses quantum superposition and entanglement to implement quantum evolutionary concepts such as quantum chromosome, entangled crossover, rotation mutation, and quantum elitism, to efficiently perform genetic evolution on quantum devices, and converge towards proper sub-optimal solutions of a given optimization problem. The proposed quantum genetic algorithm has been implemented by using a hybrid hardware architecture, where classical processors interact with the family of quantum processors provided by the IBM Q Experience® initiative. As shown in the experimental section, the proposed quantum genetic algorithm’s performance highlights that the synergy between quantum and evolutionary computation results in a new and promising bio-inspired optimization strategy.

Hybridization mechanism of the dual proximity effect in superconductor–topological insulator interfaces

In this communication we consider generalities of the proximity effect in a contact between a conventional s-wave superconductor (S) nano-island and a thin film of a topological insulator (TI). A local hybridization coupling mechanism is considered and a corresponding model is corroborated that captures not only the induced unconventional superconductivity in a TI, but also predicts the spreading of topologically protected surface states into the superconducting over-layer. This dual nature of the proximity effect leads specifically to a modified description of topological superconductivity in these systems. Experimentally accessible signatures of this phenomenon are discussed in the context of scanning tunneling microscopy measurements. For this purpose an effective density of states is computed in both the superconductor and topological insulator. As a guiding example, practical applications are made for Nb islands deposited on a surface of Bi2Se3. The obtained results are general and can be applied beyond the particular material system used. Possible implications of these results to proximity circuits and hybrid hardware devices for quantum computation processing are discussed.

A hybrid hardware oriented motion estimation algorithm for HEVC/H.265

High Efficiency Video Coding (HEVC) is the latest video coding standard that supports high resolution videos by providing approximately twice the compression efficiency as compared to its previous standard H.264. Motion Estimation (ME) in HEVC is the most computation-intensive block as a result it becomes a bottleneck in the design of the encoder while implementing video applications on various computing platforms such as general purpose and embedded processors. So developing computational efficient architectures on Field Programmable Gate Array (FPGA) and Application Specific Integrated Circuit (ASIC) platforms is inevitable. This paper proposes a fast hybrid search pattern algorithm and its hardware architecture for encoding UHD videos. The proposed Integer ME (IME) algorithm requires an average of 11.19% less encoding time than the default Test Zone Search (TZS) algorithm in HM reference software with compromising decrement in PSNR and increment in bit rate. The proposed architecture is implemented in both FPGA and ASIC platform with TSMC 90 nm technology library. It consumed 32-33% of resources in Virtex-7 FPGA and 2784.4 K equivalent gate count (in terms of NAND ) and 18 kB of memory, respectively. The results show that maximum frequency of the proposed architecture is 162 MHz and a total power consumption is 463.4 mW. The architecture provides a maximum throughput of 2.78 Gpixels/sec due to it process $$32\times 32$$ CU comparatively much less clock cycles (59.5) as compared to the state-of-the-art literature . Further, it supports 8K UHD $$(8192\times 4320)$$ @ 78 fps.

Algebraic Filtering of Surfaces from 3D Medical Images with Julia

In this paper we introduce a novel algebraic filter, based on algebraic topology methods, to extract and smooth the boundary surface of any subset of voxels arising from the segmentation of a 3D medical image. The input of the Linear Algebraic Representation (lar) Surface extraction filter (lar-surf) is defined as a chain, i.e., an element of a linear space of chains here subsets of voxels represented in coordinates as a sparse binary vector. The output is produced by a linear mapping between spaces of 3-and 2-chains, given by the boundary operator ∂3: C3 → C2. The only data structures used in this approach are sparse arrays with one or two indices, i.e., sparse vectors and sparse matrices. This work is based on lar algebraic methods and is implemented in Julia language, natively supporting parallel computing on hybrid hardware architectures.

Composite PEEK/carbon fiber rods in the treatment for bone tumors of the cervical spine: a case series

Radiotherapy (RT) is frequently applied as an adjuvant therapy during spinal tumors treatment. Metallic implants can interfere with RT planning and execution, as it is known that metallic implants produce a backscattering effect that can limit RT accuracy and their presence can be associated with unwanted dose increase. PEEK/carbon fiber implants are designed to reduce these problems but their application in the cervical spine is limited, due to the reduced number and types of implants, the screw dimensions and the absence of lateral mass screws. We propose a hybrid system made of carbon rods and screws coupled with subliminal polyester bands with titanium clamps. We designed this hybrid construct to enclose the cervical region in the area of instrumentation without limiting the application of postoperative radiotherapy. Six patients in which the hybrid hardware was implanted were retrospectively examined. Data on demographics, intraoperative and postoperative events, tumor details and staging and cervical alignment were collected pre- and postoperatively. No intraoperative complications occurred. En bloc resection was performed in two patients, while the remaining four received an intralesional resection. Three out of six patients received postoperative RT, without any alteration in its planning and administration. Hybrid implants made of composite PEEK/carbon fiber screws and rods and sublaminar bands are a helpful solution for spinal reconstruction in the cervical and cervico-thoracic regions after spine tumor surgery. The implants do not produce artifacts at postoperative images, easing the planning and execution of postoperative radiotherapy.

An efficient hybrid digital architecture for space vector PWM method for multilevel VSI

This paper presents an efficient, cost effective design implementation of a hybrid digital architecture for space vector pulse width modulation (SVPWM) method for multilevel inverters (MLIs). The SVPWM method is one of the most popular real time PWM method for three phase voltage source inverter (VSI). The implementation of SVPWM method becomes complex with an increase in the number of levels in a multilevel inverter. The SVPWM method for multilevel inverter is a multitask system. The main constraint when it comes to implementing SVPWM for multilevel inverters is the processing of dwell time computation and the generation of PWM gate signals for all of the switches with an accurate delay. A hybrid hardware structure consisting of a simple low-cost, low-power dsPIC micro controller (dsPIC 30F4011) and a state of the art Field Programmable Gate Array (FPGA) (Cyclone V 5CGXFC5C6F27C7N) is used to implement SVPWM. The proposed hybrid digital architecture utilizes the advantages and resources of the dsPIC and FPGA. The hybrid digital architecture meets the timing constraints of multitasking through synchronization and parallelism. A communication interface between the dsPIC and the FPGA reduces the design complexity. The software overhead for the communication interface remains fixed for any number of levels. The hybrid structure of the digital architecture provides scalability for the SVPWM method with more number of levels in multilevel inverter. The operation of the proposed hybrid digital architecture is experimentally validated with an optimized SVPWM method for a five level VSI. An optimized region identification algorithm and simple dwell time expressions are described for a five level SVPWM. The input DC of the five level VSI is obtained from a differential power processing (DPP) based PV system. Experimental results under different operating conditions are presented.

Testing a hybrid hardware quantum multi-agent system architecture that utilizes the quantum speed advantage for interactive computer music

This paper introduces MIq (Multi-Agent Interactive qgMuse), which builds on the single agent quantum system qgMuse using teleportation. MIq is the first attempt at a real-time interactive quantum computer music algorithm that utilises the quantum advantage. Previous interactive or real-time quantum music algorithms running on quantum computers have been mappings of classical computing algorithms, with no quantum advantage obtained. MIq provides a quadratic speed-up over classical methods. It is a Quantum Hybrid Multi-agent System architecture implemented on 5 and 14 qubit quantum hardware. Classical agents and quantum agents connect via a classical/quantum hybrid agent that enables communication using quantum teleportation.

Super-Resolution Channel Estimation for Arbitrary Arrays in Hybrid Millimeter-Wave Massive MIMO Systems

This paper develops efficient channel estimation techniques for millimeter-wave (mmWave) massive multiple-input multiple-output (MIMO) systems under practical hardware limitations, including an arbitrary array geometry and a hybrid hardware structure. Taking on an angle-based approach, this work adopts a generalized array manifold separation approach via Jacobi-Anger approximation, which transforms a non-ideal, non-uniform array manifold into a virtual array domain with a desired uniform geometric structure to facilitate super-resolution angle estimation and channel acquisition. Accordingly, structure-based optimization techniques are developed to effectively estimate both the channel covariance and the instantaneous channel state information (CSI) within a short sensing time. In particular, the difference in time-variation of channel path angles and path gains is capitalized to design a two-step CSI estimation scheme that can quickly sense fading channels. Theoretical results are provided on the fundamental limits of the proposed technique in terms of sample efficiency. For computational efficiency, a fast iterative algorithm is developed via the alternating direction method of multipliers. Other related issues such as spurious-peak cancellation in nonuniform linear arrays and extensions to higher-dimensional cases are also discussed. Simulations testify the effectiveness of the proposed approaches in hybrid mmWave massive MIMO systems with arbitrary arrays.

Multi-scale high-performance computing in astrophysics: simulating clusters with stars, binaries and planets.

The demand on simulation software in astrophysics has dramatically increased over the last decades. This increase is driven by improvements in observational data and computer hardware. At the same time, computers have become more complicated to program due to the introduction of more parallelism and hybrid hardware. To keep up with these developments, much of the software has to be redesigned. In order to prevent the future need to rewrite again when new developments present themselves, the main effort should go into making the software maintainable, flexible and scalable. In this paper, we explain our strategy for coupling elementary solvers and how to combine them into a high-performance multi-scale environment in which complex simulations can be performed. The elementary parts can remain succinct while supporting the aggregation to more satisfactory functionality by coupling them on a higher level. The advanced code-coupling strategies we present here allow such a hierarchy and support the development of complex codes. A library of simple elementary solvers subsequently stimulates the rapid development of more complex code that can co-evolve with the latest advances in computer hardware. We demonstrate how to combine several of these elementary solvers in a hierarchical and generic system, and how the resulting complex codes can be applied to multi-scale problems in astrophysics. Our aim is to achieve the best of several worlds with respect to performance, flexibility and maintainability while reducing development time. We succeeded in the development of the hierarchical coupling strategy and the general framework, but a comprehensive library of minimal fundamental-physics solvers is still unavailable. This article is part of the theme issue 'Multiscale modelling, simulation and computing: from the desktop to the exascale'.

An Efficient DSP–FPGA-Based Implementation of Hybrid PWM for Electric Rail Traction Induction Motor Control

Low switching frequency is always used in an electric rail traction induction motor control system, in order to reduce switching losses and increase system reliability. But under low switching frequency, the conventional single pulse width modulation (PWM) technique for the whole speed region will produce a large number of harmonic components, which in turn cause motor heating, torque ripple, and other adverse effects. In this paper, a hybrid PWM technique formed by merging carrier-based asynchronous and synchronous modulation and optimal synchronous modulation is proposed to achieve low losses and total harmonics based on very low switching frequency. A simple low-cost and low-power digital signal processor (DSP) and field programmable gate array (FPGA) hybrid hardware structure is then adopted for implementing the technique, which fully utilizes advantages and resources of each processor. Total harmonic distortion of the system can hence be reduced by operating with the high-precision hybrid PWM, while not compromising the multirate and multitask processing requirements of traction control. Detailed implementations in both DSP and FPGA with high accuracy, low usage of resources, and computational burden are then described, focusing particularly at the smooth transitions among different PWM schemes. Simulations and experiments are eventually performed with results captured for validating the presented hybrid PWM technique.

A GPU‐aware mixed‐precision solver for low‐rank algebraic Riccati equations

SummaryWe investigate different alternatives for the solution of algebraic Riccati equations on hybrid hardware platforms (ie, CPUs+GPUs). We evaluate a mixed‐precision approach that uses single‐precision arithmetic to obtain an approximation to the solution and later improve it to the desired precision, applying some steps of an economic iterative refinement. This method exploits the higher performance of the hardware to accelerate the solver when single‐precision arithmetic is employed and simultaneously obtains a high‐accuracy solution with the iterative refinement. We extend this approach to exploit the low‐rank property of the equation, when possible, to further improve its efficiency. The experimental evaluation shows that the mixed‐precision approach reports time and energy savings and provides similar or even more accurate solutions than well‐known methods like the sign function iteration or the structure‐preserving doubling algorithm.

A Survey to Predict the Trend of AI-able Server Evolution in the Cloud

About a decade ago, people concerned about the risks of adopting cloud computing. It was an unproven new thing that raised more questions than it answered. Nowadays, we hear more about the risks of not adopting the cloud. Three of the leading cloud players, Amazon Web Services, Microsoft Azure, and Google Cloud Platform, and other participants have developed complex cloud platforms that are driving the cloud agenda and launching innovative new products to meet the needs of modern businesses. When looking at processors, core components of the cloud, there is a trend for hyperscale data centers is to move beyond the CPUs and turn to dedicated chips, such as graphics processing units, field programmable gating arrays, and application specific integrated circuits. We think it is an artificial intelligence (AI) realization process and provide a detailed survey about hardware server design in this process. After discussing and summarizing various disclosed techniques and platforms, we conceived a hybrid hardware structure for efficient AI applications.

Computerized Solar Radiation Tracker in Home Appliances using Mathematical Morphological Operator

Objective: The proposed topic is implemented for the development of an automatic computerized solar radiation tracker and using power in home appliances. Other resource of energy are short term, thus solar energy is an emerging new source fuel, that can be applied to any source and it is been widely used. Methods: The aim of this research work is to monitor feasible approachto maximize the efficiency of solar array systems. Proposed in this report is a system that controls the movement of a solar array so that it is constantly aligned towards the direction of the sun and that there is a constant power supply without any hindrance. This is achieved by applying the measurements using Mathematical Morphological Operators. Existing tracking systems focus on directing the sun rays towards the solar panel and it lands up in manual operation. To overcome this problem a computerized system is proposed to enhance the proposed model to have a vital role in tracking the solar system used to get the energy. To find the region or the path in the solar panel morphological operator is applied to have a calculated process. Thus this proposed system helps to locate and redirect the deviated and lost energy rays. Findings: Solar Energy is a vital resource nowadays to operate any kind of appliances. In countries like India solar energy is a big boon as the summer is long and this paves for more solar power. In this abstract development of an automatic computerized solar radiation tracker and using power in home appliances is discussed. Other resource of energy are short term, thus solar energy is an emerging new source fuel, that can be applied to any source and it is been widely used. Solar panels or modules are devices that convert sunlight into electricity and offer a practical solution to the problem of power generation in remote areas. The Proposed Computerized solar tracker designed and constructed offers a reliable and affordable method of aligning a solar module with the sun in order to maximize its energy output. Automatic Sun Tracking System is a combination of both hardware and software (hybrid) hardware prototype, which automatically provides best alignment of solar panels according to the sunrays position, to get maximum output. Application: Thus this system can be implemented to enhance the working technology of the methods available for a better usage of the solar power. Keywords: Measurements, Morphological Operator, Solar Panels, Stepper Motor, Tracker

Increasing the Resilience of Cyber Physical Systems in Smart Grid Environments using Dynamic Cells

Resilience is an important system property that relies on the ability of a system to automatically recover from a degraded state so as to continue providing its services. Resilient systems have the means of detecting faults and failures with the added capability of automatically restoring their normal operations. Mastering resilience in the domain of Cyber-Physical Systems is challenging due to the interdependence of hybrid hardware and software components, along with physical limitations, laws, regulations and standards, among others. In order to overcome these challenges, this paper presents a modeling approach, based on the concept of Dynamic Cells, tailored to the management of Smart Grids. Additionally, a heuristic algorithm that works on top of the proposed modeling approach, to find resilient configurations, has been defined and implemented. More specifically, the model supports a flexible representation of Smart Grids and the algorithm is able to manage, at different abstraction levels, the resource consumption of individual grid elements on the presence of failures and faults. Finally, the proposal is evaluated in a test scenario where the effectiveness of such approach, when dealing with complex scenarios where adequate solutions are difficult to find, is shown.

Validation of short-circuit current calculations in electric power system

The determining factor for the selection of electrical equipment and switching devices is the short-circuit currents (SCC), for which the electrodynamics and thermal resistance is checked for the first, and for the second the switching capacity and wear-resistance. Also, the fault current is calculated for setting the relay protection and automation threshold. As is known, different digital simulation tools are used for calculating the SCC, in which inevitably use simplifications and limitations of mathematical models of equipment and EPS in whole. In this connection, the calculated values of SCC comprise, in some cases, significant error, so the validation is needed for used simulation tools. Proposed to use specialized hybrid software and hardware system for validation as the source of data needed to compare the simulation results of verified tools. It helps to determine their actual properties and capabilities, as well as to increase the reliability of decisions made on the basis of the calculated SCC, by understanding the causes and possible level of errors that arise.