Search

Abstract Spiking neural network algorithms require fine-tuned neuromorphic hardware to increase their effectiveness. Such hardware, mainly digital, is typically built on mature silicon nodes. Future artificial intelligence applications will demand the execution of tasks with increasing complexity and over timescales spanning several decades. The multi-timescale requirements for certain tasks cannot be attained effectively enough through the existing silicon-based solutions. Indium-Gallium-Zinc-Oxide thin-film transistors can alleviate the timescale-related shortcomings of silicon platforms thanks to their bellow atto-ampere leakage currents. These small currents enable wide timescale ranges, far beyond what has been feasible through various emerging technologies. Here we have estimated and exploited these low leakage currents to create a multi-timescale neuron that integrates information spanning a range of 7 orders of magnitude and assessed its advantages in larger networks. The multi-timescale ability of this neuron can be utilized together with silicon to create hybrid spiking neural networks capable of effectively executing more complex tasks than their single-technology counterparts.

The development of future full-scale quantum computers (QCs) not only comprises the design of good quality qubits, but also entails the design of classical complementary metal–oxide semiconductor (CMOS) control circuitry and optimized operation protocols. The construction and implementation of quantum error correction (QEC) protocols, necessary for correcting the errors that inevitably occur in the physical qubit layer, form a crucial step in this design process. The steadily rising numbers of qubits in a single system make the development of small-scale quantum architectures that are able to execute such protocols a pressing challenge. Similar to classical systems, optimized simulation tools can greatly improve the efficiency of the design process. We propose an automated simulation framework for the development of qubit microarchitectures, in which the effects of design choices in the physical qubit layer on the performance of QEC protocols can be evaluated, whereas the focus in the current state-of-the-art design tools only lies on the simulation of the individual quantum gates. The hybrid Hamiltonian framework introduces the innovative combination of a hybrid nature that allows to incorporate several levels throughout the QC stack, with optimized embedded solvers. This provides the level of detail required for an in-depth analysis of the QEC protocol's stability.

Probabilistic forecasting models have been developed over the past years in order to aid in the estimation of the uncertainty of the predictive results. A hybrid, bootstrapping long-short term memory (Boot-LSTM)-based model is proposed in this paper, in order to construct accurate prediction intervals (PIs) for short-term solar power generation. A novel approach that introduces an improved chicken swarm optimization (ICSO) algorithm along with a prey-predator (PP) mechanism is developed in order to optimize the predictive accuracy. Exploiting the ICSO’s ability to optimize the position of the swarm’s particles as well as the PP’s ability to further improve the particles’ searching possibilities, the weights and biases of the neurons of the neural network (NN) of the model are optimized and the predictive accuracy is further improved. The accuracy of the PIs is evaluated by minimizing the coverage width criterion (CWC) cost function. The efficiency and the accuracy of the proposed hybrid Boot-LSTM-ICSO-PP model is confirmed via comparing the predictive outputs with state-of-the-art methodologies considering probabilistic evaluation metrics. The proposed model was applied on two datasets of existing solar parks and was further analyzed from a seasonal perspective, in order to prove its efficiency with real-life cases. In terms of CWC minimization, the proposed model, for the first PV park, achieves a 60.3% and 46.94% average improvement compared to the base BELM and LSTM models, respectively, while for the second PV park, achieves a 50.64% and 37.87% average improvement to the respective base models as well.

Implementing a compact optical integrated circuit by utilizing subwavelength plasmonic devices requires the design of compact and efficient photonic to plasmonic mode converters. Especially for plasmonic multiple-input devices such as logic gates that require multiple converters, the footprint can be largely penalized by the photonic waveguides, which should be considered in the design. In this work, we simulate and benchmark five photonic to plasmonic mode converter topologies for the application of multiple-input plasmonic devices. Our design includes both directional and end coupling schemes of plasmonic waveguide and Si photonic waveguides of wire and slot configurations. We optimize the performance of the converters considering the pitch mismatch between the photonic and plasmonic waveguides, the total footprint, and mode conversion efficiency.

Accurate simulation techniques are indispensable to efficiently propose new memory or architectural organizations. As implementing new hardware concepts in real systems is often not feasible, cycle-accurate simulators employed together with certain benchmarks are commonly used. However, detailed simulators may take too much time to execute these programs until completion. Therefore, several techniques aimed at reducing this time are usually employed. These schemes select fragments of the source code considered as representative of the entire application’s behaviour–mainly in terms of performance, but not plenty considering the behaviour of cache memory levels–and only these intervals are simulated. Our hypothesis is that the different simulation windows currently employed when evaluating microarchitectural proposals, especially those involving the last level cache (LLC), do not reproduce the overall cache behaviour during the entire execution, potentially leading to wrong conclusions on the real performance of the proposals assessed. In this work, we first demonstrate this hypothesis by evaluating different cache replacement policies using various typical simulation approaches. Consequently, we also propose a simulation strategy, based on the applications’ LLC activity, which mimics the overall behaviour of the cache much closer than conventional simulation intervals. Our proposal allows a fairer comparison between cache-related approaches as it reports, on average, a number of changes in the relative order among the policies assessed – with respect to the full simulation – more than 30% lower than that of conventional strategies, maintaining the simulation time largely unchanged and without losing accuracy on performance terms, especially for memory-intensive applications.

This paper proposes a method based on Computational Fluid Dynamics (CFD) and the detection of Wind Energy Extraction Latency for a given wind turbine (WT) designed for ultra-short-term (UST) wind energy forecasting over complex terrain. The core of the suggested modeling approach is the Wind Spatial Extrapolation model (WiSpEx). Measured vertical wind profile data are used as the inlet for stationary CFD simulations to reconstruct the wind flow over a wind farm (WF). This wind field reconstruction helps operators obtain the wind speed and available wind energy at the hub height of the installed WTs, enabling the estimation of their energy production. WT power output is calculated by accounting for the average time it takes for the turbine to adjust its power output in response to changes in wind speed. The proposed method is evaluated with data from two WTs (E40-500, NM 750/48). The wind speed dataset used for this study contains ramp events and wind speeds that range in magnitude from 3 m/s to 18 m/s. The results show that the proposed method can achieve a Symmetric Mean Absolute Percentage Error (SMAPE) of 8.44% for E40-500 and 9.26% for NM 750/48, even with significant simplifications, while the SMAPE of the persistence model is above 15.03% for E40-500 and 16.12% for NM 750/48. Each forecast requires less than two minutes of computational time on a low-cost commercial platform. This performance is comparable to state-of-the-art methods and significantly faster than time-dependent simulations. Such simulations necessitate excessive computational resources, making them impractical for online forecasting.

High-resolution wind datasets are crucial for ultra-short-term wind forecasting. Penetration of WT installations near urban areas that are constantly changing will motivate researchers to understand how to adapt their models to terrain changes to reduce forecasting errors. Although CFD modelling is not widely used for ultra-short-term forecasting purposes, it can overcome such difficulties. In this research, we will spatially extrapolate vertical profile LIDAR wind measurements into a 3D wind velocity field over a large and relatively complex terrain with the use of stationary CFD simulations. The extrapolated field is validated with measurements at a hub height of three WTs located in the area. The accuracy of the model increases with height because of the terrain anomalies and turbulence effects. The maximum MAE of wind velocity at WT hub height is 0.81 m/s, and MAPE is 7.98%. Our model remains accurate even with great simplifications and scarce measurements for the complex terrain conditions of our case study. The models’ performance under such circumstances establishes it as a promising tool for the evolution of ultra-short-term forecasting as well as for the evaluation of new WT installations by providing valuable data for all models.

Today, video analytics are becoming extremely popular due to the increasing need for extracting valuable information from videos available in public sharing services through camera-driven streams in IoT environments. To avoid data communication overheads, a common practice is to have computation close to the data source rather than Cloud offloading. Typically, video analytics are organized as separate tasks, each with different resource requirements (e.g., computational- vs. memory-intensive tasks). The serverless computing paradigm forms a promising approach for mapping such types of applications, enabling fine-grained deployment and management in a per-function, and per-device manner. However, there is a tradeoff between QoS adherence and resource efficiency. Performance variability due to function co-location and prevalent resource heterogeneity make maintaining QoS challenging. At the same time, resource efficiency is essential to avoid waste, such as unnecessary power consumption and CPU reservation. In this paper, we present Darly, a QoS-, interference- and heterogeneity-aware Deep Reinforcement Learning-based Scheduler for serverless video analytics deployments on top of distributed Edge nodes. The proposed framework incorporates a DRL agent that exploits performance counters to identify the levels of interference and the degree of heterogeneity in the underlying Edge infrastructure. It combines this information along with user-defined QoS requirements to improve resource allocations by deciding the placement, migration, or horizontal scaling of serverless functions. We evaluate Darly on a typical Edge cluster with a real-world workflow composed of commonly used serverless video analytics functions and show that our approach achieves efficient scheduling of the deployed functions by satisfying multiple QoS requirements for up to 91.6% (Profile-based) of the total requests under dynamic conditions.

Human–Machine Interfaces (HMIs) have gained popularity as they allow for an effortless and natural interaction between the user and the machine by processing information gathered from a single or multiple sensing modalities and transcribing user intentions to the desired actions. Their operability depends on frequent periodic re-calibration using newly acquired data due to their adaptation needs in dynamic environments, where test–time data continuously change in unforeseen ways, a cause that significantly contributes to their abandonment and remains unexplored by the Ultrasound-based (US-based) HMI community. In this work, we conduct a thorough investigation of Unsupervised Domain Adaptation (UDA) algorithms for the re-calibration of US-based HMIs during within-day sessions, which utilize unlabeled data for re-calibration. Our experimentation led us to the proposal of a CNN-based architecture for simultaneous wrist rotation angle and finger gesture prediction that achieves comparable performance with the state-of-the-art while featuring 87.92% less trainable parameters. According to our findings, DANN (a Domain-Adversarial training algorithm), with proper initialization, offers an average 24.99% classification accuracy performance enhancement when compared to no re-calibration setting. However, our results suggest that in cases where the experimental setup and the UDA configuration may differ, observed enhancements would be rather small or even unnoticeable.

Abstract A review of the state‐of‐the‐art in short‐term Solar Power Forecasting (SPF) methodologies is presented in this paper. Over the last few years, developing and improving solar forecasting models has been the main focus of researchers, considering the need to efficiently increase their forecasting accuracy. Forecasting models aim to be used as an efficient tool to help with the stability and control of energy systems and electricity markets. Intending to further comprehend the factors affecting the quality of SPF models, this paper focuses on short‐term solar forecasting methodologies since they pose a crucial role in the daily operation and scheduling of power systems, since they focus on forecasting horizons typically ranging from 1 h to 1 day. The reviewed works are classified according to the climatic conditions, technical characteristics, and the forecasting errors of the different methodologies, providing readers with information over various different cases of SPF. Considering the need to improve the SPF efficiency, such classifications allow for important comparative conclusions to be drawn, depending on the location of each case and the meteorological data available. Future directions in the field of short‐term solar power forecasting are proposed considering the increasing development of SPF models’ architecture and their field of focus.

Non-Intrusive Load Monitoring (NILM), which provides sufficient load for the energy consumption of an entire building, has become crucial in improving the operation of energy systems. Although NILM can decompose overall energy consumption into individual electrical sub-loads, it struggles to estimate thermal-driven sub-loads such as occupants. Previous studies proposed Non-Intrusive Thermal Load Monitoring (NITLM), which disaggregates the overall thermal load into sub-loads; however, these studies evaluated only a single building. The results change for other buildings due to individual building factors, such as floor area, location, and occupancy patterns; thus, it is necessary to analyze how these factors affect the accuracy of disaggregation for accurate monitoring. In this paper, we conduct a fundamental evaluation of NITLM in various realistic office buildings to accurately disaggregate the overall thermal load into sub-loads, focusing on occupant thermal load. Through experiments, we introduce NITLM with deep learning models and evaluate these models using thermal load datasets. These thermal load datasets are generated by a building energy simulation, and its inputs for the simulation were derived from realistic data like HVAC on/off data. Such fundamental evaluation has not been done before, but insights obtained from the comparison of learning models are necessary and useful for improving learning models. Our experimental results shed light on the deep learning-based NITLM models for building-level efficient energy management systems.

Wind power generation is characterized by high intermittency and volatility owing to the stochastic nature of wind. In addition to forecasting accuracy, forecasting uncertainty quantification can have a major impact on power system energy management and operations planning. In this paper, a new fully co-optimized hybrid short-term probabilistic Wind Power Forecasting (WPF) model is proposed for the construction of Prediction Intervals (PIs) in a spatiotemporal framework. A Multi-Objective Improved Adaptive Particle Swarm Optimization (MOIAPSO) algorithm is developed to optimize the model’s parameters. PIs are generated by nonlinear autoregressive networks with exogenous inputs (NARX) using the Lower Upper Bound Estimation (LUBE) method. Unlike previous related work, the components of the proposed hybrid NARX-LUBE-MOIAPSO model, as well as the initial settings and the configuration of the parameters, are determined based on a full co-optimization approach. The co-optimization is performed from a forecasting quality and training time trade-off perspective. Furthermore, a spatiotemporal framework is introduced to improve forecasting performance and comprehension of regional spatiotemporal uncertainty dynamics. The spatiotemporal framework comprises a novel conditional spatiotemporal forecasting methodology and the modeling of spatiotemporal dependencies based on a binary Probabilistic Forecasting Error (PFE) metric. The proposed model and spatiotemporal framework are tested on publicly available datasets consisting of turbine-specific measurements and generate accurate forecasts with efficient uncertainty quantification, while maintaining computational complexity at relatively low levels compared to other state-of-the-art hybrid probabilistic WPF models.

Smart computing on edge-devices has demonstrated huge potential for various application sectors such as personalized healthcare and smart robotics. These devices aim at bringing smart computing close to the source where the data is generated or stored, while coping with the stringent resource budget of the edge platforms. The conventional Von-Neumann architecture fails to meet these requirements due to various limitations e.g., the memory-processor data transfer bottleneck. Memristor-based Computation-In-Memory (CIM) has the potential to realize such smart edge computing for data-dominated Artificial Intelligence (AI) applications by exploiting both the inherent properties of the architecture and the physical characteristics of the memristors. This paper discusses different aspects of CIM, including classification, working principle, CIM potentials and CIM design-flow. The design-flow is illustrated through two case studies to demonstrate the huge potential of CIM in realizing orders of magnitude improvement in energy-efficiency as compared to the conventional architectures. Finally future challenges and research directions of CIM are covered.

Plasmonic metal-insulator-metal waveguide (MIM WG) is a promising building block of devices of a plasmonic computing system, and metal-semiconductor-metal (MSM) WG offers high-speed detection of plasmon signals. MIM and MSM WG couplers are an essential component of any integrated computing system that requires an interface between the plasmonic devices and the electronic circuits. However, the MIM and MSM WG couplers have not yet been explored and systematically studied. This work analyzes and benchmarks various coupling schemes of the plasmonic MIM and MSM WGs to single out the best coupling approach. From the detailed numerical analysis and considering the trade-offs among coupling loss, total capacitance, system noise, and bandwidth, a holistic metric is introduced to quantify and compare the system-level performance of the couplers, and a novel coupling scheme is identified as the most attractive choice for coupling the two WGs.

Under dynamic pricing, stable and accurate electricity price forecasting on the demand side is essential for efficient energy management. We have developed a new electricity price forecasting model that provides consistently accurate forecasts. The base prediction model decomposes the time series using wavelet transform and then predicts it by Long Short-Term Memory. Previous studies using this model have always decomposed time series in the same way without changing the mother wavelet. However, this makes it difficult to respond to changes in time series that vary daily or seasonally. Therefore, we periodically switch the mother wavelet, i.e., flexibly change the time series decomposition method, to achieve stable and highly accurate electricity price forecasting. In an experiment, the model improved prediction accuracy by up to 42.8% compared to prediction with a fixed mother wavelet. Experimental results show that the proposed flexible forecasting method can consistently provide highly accurate forecasts.

Scenario generation has attracted wide attention in recent years owing to the high penetration of uncertainty sources in modern power systems and the introduction of stochastic optimization for handling decision-making problems. These include unit commitment, optimal bidding, online supply–demand management, and long-term planning of integrated renewable energy systems. Simultaneously, the installed capacity of solar power is increasing due to its availability and periodical characteristics, as well as the flexibility and cost reduction of photovoltaic (PV) technologies. This paper evaluates scenario generation methods in the context of solar power and highlights their advantages and limitations. Furthermore, it introduces taxonomies based on weather classification techniques and temporal horizons. Fine-grained weather classifications can significantly improve the overall quality of the generated scenario sets. The performance of different scenario generation methods is strongly related to the temporal horizon of the target domain. This paper also conducts a systematic review of the currently trending deep generative models to assess introduced improvements, as well as to identify their limitations. Finally, several research directions are proposed based on the findings and drawn conclusions to address current challenges and adapt to future advancements in modern power systems.

As radar sensors are being miniaturized, there is a growing interest for using them in indoor sensing applications such as indoor drone obstacle avoidance. In those novel scenarios, radars must perform well in dense scenes with a large number of neighboring scatterers. Central to radar performance is the detection algorithm used to separate targets from the background noise and clutter. Traditionally, most radar systems use conventional constant false alarm rate (CFAR) detectors but their performance degrades in indoor scenarios with many reflectors. Inspired by the advances in nonlinear target detection, In this article, we propose a novel high performance, yet low-complexity target detector and we experimentally validate our algorithm on a dataset acquired using a radar mounted on a drone. We experimentally show that our proposed algorithm drastically outperforms ordered statistics CFAR (OS-CFAR) (standard detector used in automotive systems) for our specific task of indoor drone navigation with more than 19% higher probability of detection for a given probability of false alarm. We also benchmark our proposed detector against a number of recently proposed multitarget CFAR detectors and show an improvement of 16% in probability of detection compared to censored harmonic averaging CFAR, with even larger improvements compared to both outlier-robust CFAR and truncated statistics log-normal CFAR in our particular indoor scenario. To the best of authors’ knowledge, this article improves the state-of-the-art for high-performance yet low-complexity radar detection in critical indoor sensing applications.

Bioimpedance is an unobtrusive noninvasive technique to measure respiration and has a linear relation with volume during normal breathing. The objective of this paper was to assess this linear relation during inspiratory loading protocol and determine the best electrode configuration for bioimpedance measurement. The inspiratory load is a way to estimate inspiratory muscle function and has been widely used in studies of respiratory mechanics. Therefore, this protocol permitted us to evaluate bioimpedance performance under breathing pattern changes. We measured four electrode configurations of bioimpedance and airflow simultaneously in ten healthy subjects using a wearable device and a standard wired laboratory acquisition system, respectively. The subjects were asked to perform an incremental inspiratory threshold loading protocol during the measurements. The load values were selected to increase progressively until the 60% of the subject's maximal inspiratory pressure. The linear relation of the signals was assessed by Pearson correlation (${r}$ ) and the waveform agreement by the mean absolute percentage error (MAPE), both computed cycle by cycle. The results showed a median greater than 0.965 in ${r}$ coefficients and lower than 11 % in the MAPE values for the entire population in all loads and configurations. Thus, a strong linear relation was found during all loaded breathing and configurations. However, one out of the four electrode configurations showed robust results in terms of agreement with volume during the highest load. In conclusion, bioimpedance measurement using a wearable device is a noninvasive and a comfortable alternative to classical methods for monitoring respiratory diseases in normal and restrictive breathing.

Scenario-aware dataflow (SADF) is a prominent tool for modeling and analysis of dynamic embedded dataflow applications. In SADF the application is represented as a finite collection of synchronous dataflow (SDF) graphs, each of which represents one possible application behaviour or scenario. A finite state machine (FSM) specifies the possible orders of scenario occurrences. The SADF model renders the tightest possible performance guarantees, but is limited by its finiteness. This means that from a practical point of view, it can only handle dynamic dataflow applications that are characterized by a reasonably sized set of possible behaviours or scenarios. In this paper we remove this limitation for a class of SADF graphs by means of SADF model parametrization in terms of graph port rates and actor execution times. First, we formally define the semantics of the model relevant for throughput analysis based on (max,+) linear system theory and (max,+) automata. Second, by generalizing some of the existing results, we give the algorithms for worst-case throughput analysis of parametric rate and parametric actor execution time acyclic SADF graphs with a fully connected, possibly infinite state transition system. Third, we demonstrate our approach on a few realistic applications from digital signal processing (DSP) domain mapped onto an embedded multi-processor architecture.

Nanostructured materials have attracted considerable interest over the last few decades to enhance sensing capabilities thanks to their unique properties and large surface area. In particular, noble metal nanostructures offer several advantages including high stability, non-toxicity and excellent electrochemical behaviour. However, in recent years the great expansion of point-of-care (POC) and wearable systems and the attempt to perform measurements in tiny spaces have also risen the need of increasing sensors miniaturization. Fast constant potential electrodeposition techniques have been proven to be an efficient way to obtain conformal platinum and gold nanostructured layers on macro-electrodes. However, this technique is not effective on micro-electrodes. In this paper, we investigate an alternative one-step deposition technique of platinum nanoflowers on micro-electrodes by linear sweep voltammetry (LSV). The effective deposition of platinum nanoflowers with similar properties to the ones deposited on macro-electrodes is confirmed by morphological analysis and by the similar roughness factor (~200) and capacitance (~18 μ F/mm 2 ). The electrochemical behaviour of the nanostructured layer is then tested in an solid-contact (SC) L i + -selective micro-electrode and compared to the case of macro-electrodes. The sensor offers Nernstian calibration with same response time (~15 s) and a one-order of magnitude smaller limit of detection (LOD) ( 2.6 × 10 − 6 ) with respect to the macro-ion-selective sensors (ISE). Finally, sensor reversibility and stability in both wet and dry conditions is proven.

Instruction memory organisations are pointed out as one of the major sources of energy consumption in embedded systems. As these systems are characterised by restrictive resources and a low-energy budget, any enhancement in this component allows not only to decrease the energy consumption but also to have a better distribution of the energy budget throughout the system. Loop buffering is an effective scheme to reduce energy consumption in instruction memory organisations. In this paper, the loop buffer concept is applied in real-life embedded applications that are widely used in biomedical Wireless Sensor Nodes, to show which scheme of loop buffer is more suitable for applications with certain behaviour. Post-layout simulations demonstrate that a trade-off exists between the complexity of the loop buffer architecture and the energy savings of utilising it. Therefore, the use of loop buffer architectures in order to optimise the instructionmemory organisation from the energy efficiency point of view should be evaluated carefully, taking into account two factors: (1) the percentage of the execution time of the application that is related to the execution of the loops, and (2) the distribution of the execution time percentage over each one of the loops that form the application.

We consider the problem of packet scheduling for the transmission of multiple video streams over a wireless local area network (WLAN). A cross-layer optimization framework is proposed to minimize the wireless transceiver energy consumption while meeting the user required visual quality constraints. The framework relies on the IEEE 802.11 standard and on the embedded bitstream structure of the scalable video coding scheme. It integrates an application-level video quality metric as QoS constraint (instead of a communication layer quality metric) with energy consumption optimization through link layer scaling and sleeping. Both energy minimization and min-max energy optimization strategies are discussed. Simulation results demonstrate significant energy gains compared to the state-of-the-art approaches.

Multiantenna systems and more particularly those operating on multiple input and multiple output (MIMO) channels are currently a must to improve wireless links spectrum efficiency and/or robustness. There exists a fundamental tradeoff between potential spectrum efficiency and robustness increase. However, multiantenna techniques also come with an overhead in silicon implementation area and power consumption due, at least, to the duplication of part of the transmitter and receiver radio front-ends. Although the area overhead may be acceptable in view of the performance improvement, low power consumption must be preserved for integration in nomadic devices. In this case, it is the tradeoff between performance (e.g., the net throughput on top of the medium access control layer) and average power consumption that really matters. It has been shown that adaptive schemes were mandatory to avoid that multiantenna techniques hamper this system tradeoff. In this paper, we derive smartMIMO: an adaptive multiantenna approach which, next to simply adapting the modulation and code rate as traditionally considered, decides packet-per-packet, depending on the MIMO channel state, to use either space-division multiplexing (increasing spectrum efficiency), space-time coding (increasing robustness), or to stick to single-antenna transmission. Contrarily to many of such adaptive schemes, the focus is set on using multiantenna transmission to improve the link energy efficiency in real operation conditions. Based on a model calibrated on an existing reconfigurable multiantenna transceiver setup, the link energy efficiency with the proposed scheme is shown to be improved by up to 30% when compared to nonadaptive schemes. The average throughput is, on the other hand, improved by up to 50% when compared to single-antenna transmission.