The purpose of this paper is to present in fully quantum terms the fundamentals of quantum Fourier optics in a completely coherent and comprehensive method with several key applications. This paper’s physical and mathematical models for various elements of a typical spatial optical signal processing apparatus create a new paradigm and will be of utmost importance to advance many quantum optical science and engineering fields. The results in this paper will, significantly advance the state of the art of quantum Fourier optics.

In this communication, a wideband reconfigurable reflectarray based on reflector-backed active second-order bandpass frequency selective surface (FSS) is presented. The reflector is composed of periodic short-circuited parallel plate waveguide (PPW) and the FSS is composed of stacked non-resonant metallic elements separated by thin dielectric substrates. By integrating microwave varactors in the capacitive layers of FSS, more than 270 o continuous phase tunability is achieved within a factional bandwidth of 14%. A one-dimensional reflectarray prototype operating at C band is fabricated and measured. The experimental results show that it can achieve ±55 o beam scanning coverage. Symmetric beam steering is observed due to the center-fed configuration. With advantages of low cost and simple structure, the proposed reflectarray can be potentially used in wideband wireless communication and radar systems.

As a solution to the quest of compact antennas that occupy a wider bandwidth in the 5G NR FR2 bands while simultaneously providing dual-polarization and high port-to-port isolation, the paper presents an antenna that serves this purpose. The antenna is a multilayer structure with V slots provided in the bottom and top layer to accommodate the bandwidth of 5G NR FR2 from 26 GHz to 40 GHz. The bottom V slot is responsible for the lower resonance and a cavity supports the top V slot for the upper resonance. These two resonances can be controlled independently to provide multiple wide band operation. The 2x2 antenna has both vertical and horizontal polarization. The port-to-port isolation of the antenna is also better than 25 dB throughout the band. The antenna occupies a size of 13.8mm x 13.8mm making it relatively compact. It has been designed and simulated in CST Studio for performance and then has been fabricated. The simulated results agree well with the measurement results. The antenna can find potential application in the 5G NR FR2 28/38 GHz enabled user equipment such as 5G enabled cell phones and in applications of Terrestrial Networks (TN) and Non-Terrestrial (NTN) mobile networks.

Objective: In the last two decades, there has been a growing interest in exploring surgical procedures with statistical models to analyze operations at different semantic levels. This information is necessary for developing context-aware intelligent systems, which can assist the physicians during operations, evaluate procedures afterward or help the management team to effectively utilize the operating room. The objective is to extract reliable patterns from surgical data for the robust estimation of surgical activities performed during operations. The purpose of this paper is to review the state-of-the-art deep learning methods that have been published after 2018 for analyzing surgical workflows, with a focus on phase and step recognition. Methods: Three databases, IEEE Xplore, Scopus, and PubMed were searched, and additional studies are added through a manual search. After the database search, 343 studies were screened and a total of 44 studies are selected for this review. Conclusion: The use of temporal information is essential for identifying the next surgical action. Contemporary methods used mainly RNNs, hierarchical CNNs, and Transformers to preserve long-distance temporal relations. The lack of large publicly available datasets for various procedures is a great challenge for the development of new and robust models. As supervised learning strategies are used to show proof-of-concept, self-supervised, semi-supervised, or active learning methods are used to mitigate dependency on annotated data. Significance: The present study provides a comprehensive review of recent methods in surgical workflow analysis, summarizes commonly used architectures, datasets, and discusses challenges.

In this paper we present three new algorithms for sparse multichannel blind deconvolution. The first algorithm is based on a cascade of a forward and a backward prediction error filter. The second consists in an alternating minimization algorithm for estimating both the reflectivity series and the seismic wavelet. The last one is a regularized version of the so called Euclid deconvolution, solved by the well known fast iterative shrinkage-thresholding algorithm (FISTA). Simulation results illustrate that the three algorithms presents some quite interesting results in both synthetic and real data.

Objective: The need for distilling the hemodynamic complexity into clinically relevant quantities has imposed a reductionist approach to investigate aortic flows, resulting in a lossy compression of the information hidden in 4D aortic fluid structures. Aiming at reducing information loss, this study proposes a network-based approach to identify and characterize in vivo the large-scale coherent motion of blood in the healthy human aorta using 4D Flow MRI. Methods: Adopting the quantitative paradigm of the aortic flow as a “social network”, 4D flow MRI acquisitions were performed on forty-one healthy volunteers. Correlations between the aortic blood flow rate waveform at the proximal ascending aorta (AAo), assumed as a main determinant of the aortic flow, and the waveforms of the axial velocity (aligned with the aortic centerline) in the whole aorta were used to build “one-to-all” networks. The impact of the driving flow rate waveform and of the main aortic geometric attributes on the transport of large-scale coherent fluid structures was investigated. Results: The anatomical length of persistence of large-scale coherent motion was the 29.6% of the healthy thoracic aorta length (median value, IQR 23.1%-33.9%). Moreover, it was positively correlated with the average and peak-to-peak AAo blood flow rate values, suggesting a remarkable inertial effect of the AAo flow rate on aortic hemodynamic coherence. No association emerged between the anatomical length of persistence and aortic geometric attributes. Conclusion: The here proposed in vivo approach allowed to quantitatively characterize the large-scale aortic blood motion, strengthening the definition of coherent hemodynamic structures. Significance: The findings on healthy aortas may be used as reference values to investigate the impact of aortic disease or devices implantation in disrupting/restoring the physiological spatiotemporal coherence of large-scale aortic flow.

This paper details the design of a three-phase transmitter-receiver topology for use in a dynamic wireless power transfer (DWPT) system for heavy-duty electric vehicles. The proposed approach and topology eliminate the power oscillation and voltage/current sharing problems that plague planar three-phase DWPT systems. Multi-objective optimization is employed to maximize magnetic coupling and minimize undesired positive–negative sequence coupling, yielding a Pareto-optimal front that delineates the tradeoff between the two objectives for the proposed topology. Selected designs are simulated for a 200-kW/m DWPT system, verifying the elimination of power oscillation and phase imbalance with minimal impact on magnetic performance.

In many emerging Internet of Things (IoT) applications, the freshness of the data collected by the sensor nodes is an important design criterion. Age of Information (AoI) quantifies the freshness of the received information or status update. This work considers a setup of deployed IoT devices in an IoT network; multiple unmanned aerial vehicles (UAVs) serve as mobile relay nodes between the sensors and the base station. We formulate an optimization problem to jointly plan the UAVs’ trajectory, while minimizing the AoI of the received messages and the devices’ energy consumption. The solution accounts for the UAVs’ battery lifetime and flight time to recharging depots to ensure the UAVs’ green operation. The complex optimization problem is efficiently solved using a deep reinforcement learning algorithm. In particular, we propose a deep Q-network, which works as a function approximation to estimate the state-action value function. The proposed scheme is quick to converge and results in a lower ergodic age and ergodic energy consumption when compared with benchmark algorithms such as greedy algorithm (GA), nearest neighbour (NN), and random-walk (RW).

This paper presents two deep unfolding neural networks for the simultaneous tasks of background subtraction and foreground detection in video. Unlike conventional neural networks based on deep feature extraction, we incorporate domain-knowledge models by considering a masked variation of the robust principal component analysis problem (RPCA). With this approach, we separate video clips into low-rank and sparse components, respectively corresponding to the backgrounds and foreground masks indicating the presence of moving objects. Our models, coined ROMAN-S and ROMAN-R, map the iterations of two alternating direction of multipliers methods (ADMM) to trainable convolutional layers, and the proximal operators are mapped to non-linear activation functions with trainable thresholds. This approach leads to lightweight networks with enhanced interpretability that can be trained on few data. In ROMAN-S, the correlation in time of successive binary masks is controlled with a side-information scheme based on L1-L1 minimization. ROMAN-R enhances the foreground detection by learning a dictionary of atoms to represent the moving foreground in a high-dimensional feature space and by using reweighted-L1-L1 minimization. Experiments are conducted on both synthetic and real video datasets, for which we also include an analysis of the generalization to unseen clips. Comparisons are made with existing deep unfolding RPCA neural networks, which do not use a mask formulation for the foreground. The models are also compared to a 3D U-Net baseline. Results show that our proposed models outperform other deep unfolding models, as well as the untrained optimization algorithms. ROMAN-R, in particular, is competitive with the U-Net baseline for foreground detection, with the additional advantage of providing video backgrounds and requiring substantially fewer training parameters and smaller training sets.

Quantum Processed SAR and Optical Image Data for Deep Learning Applications

This paper presents the comprehensive numerical analysis of three-sliding manifold control algorithm to estimate rotor angle/position, θe , sensorlessly in brushless DC synchronous machine. The Lyapunov stability and convergence in the discrete time is also analyzed for each of these algorithms for both continuous and discontinuous controls. The algorithm has also been examined for computation cost and complexity given by the upper bound, Big O. The control estimation algorithm is irrespective of the machine construction but this paper also elaborates 7.5 Hp (5.5 kW) brushless machine construction with winding table in a non-salient eight pole rotor, double layer wye-winding with 48 slots.

In the last decade, there has been a great interest in finding new priors for Blind Source Separation (BSS). Some of the most interesting of them are sparsity, non-negativity and boundedness of the sources, since they consist in weaker assumptions when compared to the classical statistical independence hypothesis. In this paper, we propose a new criterion for Bounded Component Analysis (BCA), using the $\ell_\infty$ norm, with an associated algorithm based on Givens Rotations. We analyze our proposal and evaluate it with numerical simulations for three kinds of bounded signals. From both theoretical and experimental results we show that the $\ell_\infty$ norm is a suitable contrast function for BCA, presenting very good results when compared to the state-of-the-art algorithm.

This paper presents a novel CMOS tunable positive and negative active inductor simulator (AIs), positive capacitance and resistance multiplier and negative capacitance and resistance simulator. The proposed design use only one second-generation voltage-mode conveyor (VCII) and three passive elements. Applications to the proposed design in the design of different types of tunable filters are also presented. The functionality of the designs is confirmed using Tanner Tspice in 0.18µm TSMC CMOS technology.

The IEC 61853 standard series aims to provide a standardized measure for PV module energy rating, namely the Climate Specific Energy Rating (CSER). For this purpose, it defines procedures for the experimental determination of input data and algorithms for calculating the CSER. However, some steps leave room for interpretation regarding the specific implementation. To analyze the impact of these ambiguities, the comparability of results and the clarity of the algorithm for calculating the CSER in part 3 of the standard, an intercomparison is performed among research organizations with 10 different implementations of the algorithm. We share the same input data, obtained by measurement of a commercial crystalline silicon PV module, among the participating organizations. Each participant then uses their individual implementations of the algorithm to calculate the resulting CSER values. The initial blind comparison reveals differences of 0.133 (14.7%) in CSER. After several comparison phases, a best practice approach is defined, which reduces the difference by a factor of 210 to below 0.001 (0.1%) in CSER for two independent PV modules. The best practice presented in this paper establishes clear guidelines for the numerical treatment of the spectral correction and power matrix extrapolation, where the methods in the standard are not clearly defined. Additionally, we provide input data and results for the PV community to test their implementations of the standard’s algorithm. To identify the source of the deviations, we introduce a climate data diagnostic set. Based on our experiences, we give recommendations for the future development of the standard.

LIGO interferometer is considered the most sensitive and complicated gravitational experimental equipment ever built. Its main objective is to detect the gravitational wave from the strongest events in the universe by observing if the length of its 4-kilometer arms change by a distance 10,000 times smaller than the diameter of a proton. Due to its sensitivity, interferometer is prone to the disturbance of external noises which affects the data being collected to detect the gravitational wave. These noises are commonly called by the gravitational-wave community as glitches. This study focuses on identifying those glitches using different deep transfer learning algorithms. The extensive experiment shows that algorithm with architecture VGG19 recorded the highest AUC-ROC among other experimented algorithm with 0.9898. While all of the experimented algorithm achieved a considerably high AUC-ROC, some of the algorithm suffered from class imbalance of the dataset which has a detrimental effect when identifying other classes.

In this paper, thirteen experiments are performed to measure the conducted electromagnetic emissions namely common mode and differential mode noises along with time-varying spectra. These emissions are generated by switching IGBTs of three phase three level inverter whose switching pulse width modulated frequency is varied at twelve frequencies from 8 kHz to 120 kHZ. Two equipments under test (EUT) are simultaneously measured for emissions using CE102 of MIL-STD-461G standard with line impedance stabilization network (LISN) and spectrum analyzer. The EUTs are inverter and asynchronous machine, where the inverter is nonlinear space vector modulated three-phase three-level diode neutral point clamped inverter, and the asynchronous machine is non-linear rotating machine, 20HP (15 kW), 400V, 50Hz, Y-winding, squirrel-cage induction motor. The harmonic distortion measurements namely total harmonic distortion(THD), THD in percentage, signal-to-noise ratio(SNR), signal-to-noise and distortion ratio(SINAD) and spurious-free dynamic range(SFDR) are calculated, tabulated and plotted at 15 harmonics resulting from the emissions. The time varying spectra is captured using Welch’s method and Short-time Fourier transform and is also submitted as dataset for each phase of the asynchronous machine and the inverter.

An adaptive switching controller based on dynamic zero moment point for versatile hip exoskeleton is proposed in this work. The linear finite hysteretic state machine is designed to recognize hybrid motion phases. The torque planning strategy based on dynamic zero moment point is deployed to obtain assistant torque adaptively under different locomotion. Experiments are carried out to verify the performance of the controller, confirming the stability and accuracy of the motion phase recognition, which also demonstrates excellent kinematic performance. The net metabolic rate can be reduced by 5.75% while wearing the versatile hip exoskeleton. Compared with existing research, the performance of the proposed controller has significant advantages. The proposed controller is capable of multiple types of locomotion including flat walking, stair climbing, and lifting heavy objects with low complexity and energy consumption. The Biological and Medical Ethics Committee of Beihang University has approved the experiments.

In this paper, brushless DC motors are evaluated for nonlinear flux-linkages and thermal losses in the windings and rotor, which causes additional voltage drop and can lead to very dangerous symmetrical three phase line-line-line (LLL) short-circuit. In the paper, it is shown that if the motor does not have appropriate cooling mechanism or other protectionary switchgear, the temperatures in 8-pole (or higher poles) motor with continuous load reaches 1000◦ C, which is nearer to the melting point of copper winding. Thirteen BLDC motors with pole pairs varying from 1 to 13 are studied for comparison for the simulation of thermal losses. The resulting eleven outputs namely rotor angle matrices, flux linkages matrices, flux linkage partial derivatives with respect to current vector, rotor angle, per phase current, each phase winding temperatures, rotor temperature, have been collected as datasets. On knowing the estimated increase in temperature, manufacturers can decide appropriate insulation, cooling mechanism and other motor protectionary circuits.