Measurement of human body movement is an essential step in biomechanical analysis. The current standard for human motion capture systems uses infrared cameras to track reflective markers placed on the subject. While these systems can accurately track joint kinematics, the analyses are spatially limited to the lab environment. Though Inertial Measurement Unit (IMU) can eliminate the spatial limitations of the motion capture system, those systems are impractical for use in daily living due to the need for many sensors, typically one per body segment. Due to the need for practical and accurate estimation of joint kinematics, this study implements a reduced number of IMU sensors and employs machine learning algorithm to map sensor data to joint angles. Our developed algorithm estimates hip, knee, and ankle angles in the sagittal plane using two shoe-mounted IMU sensors in different practical walking conditions: treadmill, level overground, stair, and slope conditions. Specifically, we proposed five deep learning networks that use combinations of Convolutional Neural Networks (CNN) and Gated Recurrent Unit (GRU) based Recurrent Neural Networks (RNN) as base learners for our framework. Using those five baseline models, we proposed a novel framework, DeepBBWAE-Net, that implements ensemble techniques such as bagging, boosting, and weighted averaging to improve kinematic predictions. DeepBBWAE-Net predicts joint kinematics for the three joint angles under all the walking conditions with a Root Mean Square Error (RMSE) 6.93-29.0% lower than base models individually. This is the first study that uses a reduced number of IMU sensors to estimate kinematics in multiple walking environments.

Detection for high-dimensional multiple-input multiple-output (MIMO) and massive MIMO (MMIMO) systems is an active field of research in wireless communications. While most works consider spatially uncorrelated channels, practical MMIMO channels are correlated. This paper investigates the impact of correlation on Sphere Decoder (SD), for both single-user (SU) and multi-user (MU) scenarios. The complexity of SD is mainly determined by the initial radius (IR) method and the number of visited nodes during detection. This paper employs an efficient IR and proposes a new metric constraint in the tree searching algorithm, that significantly decrease the number of visited nodes and render SD feasible for large-scale systems. In addition, an introduced hardware implementation featured with a one-node-per-cycle architecture, minimizes the latency of the detection process. Trade-offs between bit error rate (BER) performance and computational complexity are presented. The trade-offs are achieved by either modifying the backtracking mechanism or limiting the number of radius updates. Simulation results prove that the proposed optimizations are effective for both correlated and uncorrelated channels, regardless of the level of noise. The decoding gain of SD compared to the low-complexity linear detectors (LD) is higher in the presence of correlation than in the uncorrelated case. However, as expected, spatial correlation adversely affects the performance and the complexity of SD. Simulation results reported here also confirm that correlation at the side equipped with more antennas is less detrimental. Hardware implementation aspects are examined for both a Virtex-7 FPGA device and a 28-nm ASIC technology.

Flying capacitor voltage balancing is critical for the performance of flying capacitor multilevel (FCML) converters. This paper investigates the intrinsic capacitor voltage balancing of multiphase FCML converters with coupled inductors. It is shown that the coupled inductor provides flying capacitor voltage balancing that minimizes steady-state imbalances due to periodic disturbances compared to converters with uncoupled inductors. A dynamic model of natural balancing of the converter is derived and used to estimate the time required for the flying capacitors to settle from an initial imbalance. The theoretical predictions are verified with analytical derivations, SPICE simulations, and experimental results.

Three-dimensional engineered heart tissues (EHTs) derived from human induced pluripotent stem cells (iPSCs) have become an important resource for both drug toxicity screening and research on heart disease. A key metric of EHT phenotype is the contractile force with which the tissue spontaneously beats. It is well-known that cardiac muscle contractility – its ability to do mechanical work – depends on tissue prestrain (preload) and external resistance (afterload). Objectives: Here, we demonstrate a technique to control both preload and afterload dynamically while monitoring contractile force exerted by EHTs. Methods: We developed an apparatus that uses real-time feedback control to monitor and regulate EHT forces. The system is comprised of a pair of high-speed piezoelectric actuators that can strain the EHT scaffold and a fast optical measurement tool to provide EHT contractile force feedback while monitoring tissue strain. Results: The system was used to regulate the effective stiffness of the scaffold. When controlled to have effectively isometric boundary conditions, EHTs exerted a contractile force that was almost twice as large as that observed under auxotonic conditions. Conclusion: These experimental results demonstrate that EHT contractility can be increased through feedback control to regulate boundary stiffness. Significance: The work advances our understanding of the role that mechanical environment plays in EHT contractility. This could be used to help study or alter EHT phenotype and potentially EHT maturation through controlled mechanical conditioning.

Radio frequency interference (RFI) is a significant challenge for high-sensitivity phased array instruments. RFI can be suppressed using digital signal processing, but to improve dynamic range for wideband RFI, it can be desirable to remove interference in the analog domain before sampling. In previous work, it has been shown that analog true time delay (TTD) stages with a truncated Hadamard transform can place a wide-band spatial null on RFI from a given direction of arrival. We show that TTD and Hadamard projection is mathematically equivalent to the classical, narrow-band subspace projection beamformer, but with a structure that allows efficient implementation in analog circuitry. Simulation results show that TTD and Hadamard projection can place deep nulls on wideband RFI signals and still achieve SNR performance comparable to the optimal digital maximum signal to interference and noise ratio beamformer.

Surrogate-assisted multi-objective evolutionary algorithms have advanced the field of computationally expensive optimization, but their progress is often restricted to low-dimensional problems. This manuscript presents a multiple classifiers-assisted evolutionary algorithm based on decomposition, which is adapted for high-dimensional expensive problems in terms of the following two insights. Compared to approximation-based surrogates, the accuracy of classification-based surrogates is robust for few high-dimensional training samples. Further, multiple local classifiers can hedge the risk of over-fitting issues. Accordingly, the proposed algorithm builds multiple classifiers with support vector machines on a decomposition-based multi-objective algorithm, wherein each local classifier is trained for a corresponding scalarization function. Experimental results statistically confirm that the proposed algorithm is competitive to the state-of-the-art algorithms and computationally efficient as well.

Currently, the problem of uncovering complex network structure and dynamics from time series is prominent in many fields. Despite the recent progress in this area, reconstructing large-scale networks from limited data remains a tough problem. Existing works treat connections of nodes as continuous values, leaving a challenge of setting a proper cut-off value to distinguish whether the connections exist or not. Besides, their performances on large-scale networks are far from satisfactory. Considering the reconstruction error and sparsity as two objectives, this paper proposes a subspace learning based evolutionary multiobjective network reconstruction algorithm, termed as SLEMO-NR, to solve the aforementioned problems. In the evolutionary process, we assume that binary-coded individuals obey the Bernoulli distribution and can use the probability and natural parameter as the alternative representations. Moreover, our approach utilizes the logistic principal component analysis (LPCA) to learn a subspace containing the features of network structure. The offspring solutions are generated in the learned subspace and then can be mapped back to the original space via LPCA. Benefitting from the alternative representations, a preference-based local search operator is proposed to concentrate on finding solutions approximate to the true sparsity. The experimental results on synthetic networks and six real-world networks demonstrate that, due to the well-learned network structure subspace and the preference-based strategy, our approach is effective in reconstructing large-scale networks compared to six existing methods.

In this study, machine learning and statistical approaches are used to forecast water demand of a public company which is in charge of distributing potable water. Different methods and algorithms are briefly introduced, then assessed with the water demand case study and discussed in details.

By transferring knowledge from a source domain, the performance of deep clustering on an unlabeled target domain can be improved. When achieving this, traditional approaches make the assumption that adequate amount of labeled data is available in a source domain. However, this assumption is usually unrealistic in practice. The source domain should be carefully selected to share some characteristics with the target domain, and it can not be guaranteed that rich labeled samples are always available in the selected source domain. We propose a novel framework to improve deep clustering by transferring knowledge from a source domain without any labeled data. To select reliable instances in the source domain for transferring, we propose a novel adaptive threshold algorithm to select low entropy instances. To transfer important features of the selected instances, we propose a feature-level domain adaptation network (FeatureDA) which cancels unstable generation process. With extensive experiments, we validate that our method effectively improves deep clustering, without using any labeled data in the source domain. Besides, without using any labeled data in the source domain, our method achieves competitive results, compared to the state-of-the-art methods using labeled data in the source domain.

This article proposed a new protocol “untraceable and unclonable sensor movement in the distributed IoT environment”. The proposal has the following several advantages: (1) the new protocol using PUF achieved the main contributions of both untraceable and unclonable sensors. (2) the new protocol also achieved standard security features such as mutual authentication, perfect forward and backward secrecy. (3) The proposed authentication protocol can withstand from various kinds of attacks, such as to reply, DoS, impersonation, and cloning attack. (4) Analysis formally using BAN Logic has been conducted and denoted that the protocol achieved a secure mutual authentication. In addition, analysis formally using the RoR model and Scyther tool denoted that the proposed protocol withstands from various attacks. (5) A comparison of security features and computational complexity ensures that the proposal is secure and has low computational complexity

We present a general theory of payment systems that is capable of describing both traditional and electronic forms of payment. Starting from the three basic functions of money and general non-functional requirements, we derive the necessary and sufficient properties of technical implementations of money and payments. We describe possible scalable implementations of e-money schemes based on a general description of their data structures (money distributions) and payments. We define the notion of bill scheme, in which the value units are bills with invariant values, and show that only the bill scheme allows for scalable and practically efficient implementations through decomposition, where the components have to process a considerably smaller amount of data and a number of payment requests, compared to the whole system.

This work addresses the problem of reference tracking in autonomously learning agents with unknown, nonlinear dynamics. Existing solutions require model information or extensive parameter tuning, and have rarely been validated in real-world experiments. We propose a learning control scheme that learns to approximate the unknown dynamics by a Gaussian Process (GP), which is used to optimize and apply a feedforward control input on each trial. Unlike existing approaches, the proposed method neither requires knowledge of the system states and their dynamics nor knowledge of an effective feedback control structure. All algorithm parameters are chosen automatically, i.e. the learning method works plug and play. The proposed method is validated in extensive simulations and real-world experiments. In contrast to most existing work, we study learning dynamics for more than one motion task as well as the robustness of performance across a large range of learning parameters. The method’s plug and play applicability is demonstrated by experiments with a balancing robot, in which the proposed method rapidly learns to track the desired output. Due to its model-agnostic and plug and play properties, the proposed method is expected to have high potential for application to a large class of reference tracking problems in systems with unknown, nonlinear dynamics.

Heartbeat is a significant vital sign. The past decade has witnessed wireless heartbeat monitoring becoming more and more feasible. As a contactless approach, it can relieve users from wearing sensors. The unique advantages of milimeter-wave frequency-modulated continous-wave (FMCW) radar over other wireless devices draw our attention. However, there are plenty of limitations in related work based on it. Tiny vibration caused by heartbeat could be easily interrupted by body movements (BMs) and so on. Range bin selection methods in existing work are not robust enough. The inter-beat interval (IBI) estimation need to be further improved. We propose RF-Heartbeat, a robust heartbeat monitoring system based on milimeter-wave FMCW radar. RF-Heartbeat has novel signal process modules, including BMs detection method, and optimal range bin selection method, for obtaining a relatively high-quality heartbeat signal. Then, a global optimization model based on a template matching algorithm is implemented for improving IBI estimation accuracy. In the heartbeat monitoring experiment for a total of 72 hours and 56 minutes involving 10 participants, under a time coverage of 91.2%, the median error of IBI was 12 ms. RF-Heartbeat achieves a higher performance than state-of-the-art heartbeat monitoring systems. It has strong robustness to different individuals in different postures and positions. Our work provides effective solutions to some obstacles in practical applications of contactless heartbeat monitoring. It enables accurate and long-term heartbeat monitoring, which has wide applications in clinical space and daily life.

We propose a novel approach to realize unidirectional radiation in an asymmetric single resonator without using an external mirror, which is based on mutual coupling between two degenerate resonant modes through the background scattering of the asymmetric resonator.

In this paper, we analyze the connectivity in the underwater optical wireless networks in which similar nodes with a certain range and coverage angle, and random surface distribution are used. In our analysis, we first classify the nodes into two categories of nodes located in the inner area and nodes located in the border strip by defining the width of the border strip as the node range. Then, we conduct their connectivity analyses separately. Finally, by combining the mentioned analyses, we obtain closed form formulas to calculate the probability of connectivity (from orders 1 and 2). Moreover, by numerical evaluation, we will compare the results of the analyses with the results of computer simulation.

In the current scenario of increasing demand for solar Photo-voltaic (PV) systems, the need to predict their feasibility and monitor performance is more than ever. Although PV systems are known for their reliability, they are not above the damaging effects of their surroundings. Various lossy phenomena affect overall plant performance. In this paper, several of such losses, namely thermal, soiling, module degradation and inverter clipping, are discussed. Algorithms to evaluate these losses are proposed which are data-driven and empirical in nature. This is done as an effort to leverage the analytical capabilities provided by the plant data. The paper also compares the estimated losses with those obtained using the PVsyst simulation. As the latter is an independent industrial standard, it helps in understanding the ground reality of PV performance and insights for better operational monitoring. These insights are of immense business value and are aimed at optimizing performance and thereby revenue. As part of our asset management, all the solar PV plant components have sensors whose measurements are sent to the servers on a real-time basis. This is incorporated into our analytics portal which is used for operations and monitoring. The data used for this study is time-series in nature with a temporal least count of 5 minutes (instantaneous values spaced every 5min throughout the period of data capture). The actual data and its list of parameters is dependent on solar plant capacity and design site. For the reference dataset, a grid-connected solar rooftop PV plant in India was studied and its loss parameters were estimated. The plant components are discussed in the prologue of the results section. Solar PV is such a technology which has been enjoying increasing demand and this market scenario is quite favourable for innovation in energy research. This paper hopes to not only introduce the context of PV losses but also tries to engage the motivation to adopt data-driven and empirical methodologies to understand modern systems. This approach is better in the sense that it only gets better at prediction as time goes by and there is more data. Industrial research such as the above work in critical analysis of PV systems not only helps identify possible limitations but also suggest room for improvement. Since energy generation and project cost are key towards maximizing revenue, these estimation models aimed at predicting PV losses are to be deemed indispensable. As with any estimation, there is no one unique way of hitting the bull’s eye that is to know the exact value. The algorithms proposed above are very much dependent on the quality and quantity of data. However, the comparison between losses estimated using plant data and standard simulation using energy modelling can act as feedback towards improving the design and maintenance of such PV systems.

With the advent of new loads and generation on the low voltage grid, voltage fluctuation has increased, especially in active distribution grids with a high penetration of distributed resources and a large deployment of electric vehicles. The coordination of different technologies has emerged as the best way for voltage regulation, among others, smart inverters, open soft points or transformers with on-load regulation capability. This paper proposes a novel way to model the control strategies for the automatic voltage controller of On-Load Tap-Changer transformers. The purpose is to standardize and simplify the way these strategies are represented, in order to facilitate (i) their  election by Distribution System Operators, (ii) their future integration with other systems, and (iii) to increase the ability to anticipate On-Load Tap-Changer behavior. The proposal has been validated using real data, obtaining an accuracy of 99.15% in the tap changer positions. A unified framework is also introduced, which allows the proposed functional representation to be combined with the On-Load Tap-Changer controller behavior estimation. A experimental validation has been carried out where more than 150 000 strategies have been simulated, finally determining the one that best fits the objectives.

Despite several beneficial applications, unfortunately, drones are also being used for illicit activities such as drug trafficking, firearm smuggling or to impose threats to security-sensitive places like airports and nuclear power plants. The existing drone localization and neutralization technologies work on the assumption that the drone has already been detected and classified. Although we have observed a tremendous advancement in the sensor industry in this decade, there is no robust drone detection and classification method proposed in the literature yet. This paper focuses on radio frequency (RF) based drone detection and classification using the frequency signature of the transmitted signal. We have created a novel drone RF dataset using commercial drones and presented a detailed comparison between a two-stage and combined detection and classification framework. The detection and classification performance of both frameworks are presented for a single-signal and simultaneous multi-signal scenario. With detailed analysis, we show that You Only Look Once (YOLO) framework provides better detection performance compared to the Goodness-of-Fit (GoF) spectrum sensing for a simultaneous multi-signal scenario and good classification performance comparable to Deep Residual Neural Network (DRNN) framework.