The 50 most-cited articles in Electronics Letters, 2021

Abstract Grey wolf optimization (GWO) algorithm is a new emerging algorithm that is based on the social hierarchy of grey wolves as well as their hunting and cooperation strategies. Introduced in 2014, this algorithm has been used by a large number of researchers and designers, such that the number of citations to the original paper exceeded many other algorithms. In a recent study by Niu et al., one of the main drawbacks of this algorithm for optimizing real‐world problems was introduced. In summary, they showed that GWO's performance degrades as the optimal solution of the problem diverges from 0. In this paper, by introducing a straightforward modification to the original GWO algorithm, that is, neglecting its social hierarchy, the authors were able to largely eliminate this defect and open a new perspective for future use of this algorithm. The efficiency of the proposed method was validated by applying it to benchmark and real‐world engineering problems.

Abstract The authors report on a real‐time 48‐Gbit/s terahertz band wireless communication experiment employing a pair of resonant tunnelling diodes at the receiver and uni‐travelling carrier photodiodes at the transmitter. The wireless transmission of uncompressed full‐resolution 8K ultra high‐definition television video signal is successfully demonstrated in the 300‐GHz band for the first time.

Abstract High‐speed discrete multitone serial data transmission over multi‐mode fibre at gross data rates up to 224 Gbit/s achieved with directly modulated oxide‐confined vertical cavity surface emitting lasers at 850 and 910 nm are presented.

Abstract The 1300 nm range vertical‐cavity surface‐emitting lasers with the active region based on InGaAs/InGaAlAs superlattice are fabricated using molecular‐beam epitaxy and the double wafer‐fusion technique. Lasers with the buried tunnel junction diameter of 5 μm have shown single‐mode CW operation with the output optical power of ∼6 mW at 20°C. Opened eye diagrams are observed up to 10 Gbps.

Abstract A multi‐active‐region bipolar‐cascade edge‐emitting laser emitting at nearly 900 nm is presented. The three active regions and two tunnel junctions located in a single waveguide core share the same third‐order vertical mode. A slope efficiency of 3.6 W/A was measured with a threshold current density of 230 A/cm 2 . The epitaxial layer stack developed features with very low internal optical losses of 0.7 cm −1 . The voltage extrapolated to vanishing current is only 0.3 V larger than 3 times the voltage of 1.4 V originating from the photon energy.

Abstract Traditional network intrusion detection methods lack the ability of automatic feature extraction for encrypted network malicious traffic, and thus, the detection rates are low. Moreover, the means of this malicious traffic are concealed, and the key malicious features are usually hidden in many normal data packets, so fewer encrypted malicious traffic samples can be captured. This easily leads to insufficient system training, low detection rate, and high false alarm rate. This letter proposes an encrypted network malicious traffic detection model based on deep learning, in which automatic feature extraction is performed against encrypted network malicious traffic. The proposed model has self‐learning and self‐adaption abilities. Furthermore, a sample generation method of encrypted traffic based on deep Q ‐networks and deep convolution generative adversarial networks is proposed, in which new samples are learned from the training samples of encrypted traffic and solves problems, such as insufficient original training samples and unbalanced samples. In a validation…

Abstract Structural integrity of seawater pipelines in nuclear power plants is a very important issue. In accordance with the operating technical guidelines, the human operators directly enter the pipe and inspect it at every maintenance test. However, in this regard, safety issues such as narrow space and harmful gas are emerging every year. In response to these needs, a quadruped robot that can inspect underground pipes and assist workers has been developed. The robot has an articulated robotic arm that can receive an impact sound of hammering a pipe wall to test pipe integrity. The state of the pipe was examined using a Convolutional Neural Network algorithm. On the other hand, moving in a plumbing environment requires stable walking ability. To determine the gait sequence, a hierarchical gait controller is proposed. The hybrid controller, which consists of joint impedance and torque control, calculated from Model Predictive Control, can switch the gait modes comparing the reference and the current foot contact condition at each control cycle.

Abstract In this letter, we study the design of three‐dimensional (3D)‐printed ceramics exhibiting anisotropic dielectric permittivities at microwave frequencies for dielectric resonator antenna (DRA) applications. The anisotropy is engineered by using periodic structures made up of subwavelength asymmetric unit cells filled with zirconia and air. Ceramic samples with uniaxial anisotropy are designed, 3D‐printed, and measured. Birefringence up to 8 is achieved by controlling the volume fill rate of the unit cell. Besides, a single‐fed circularly polarised (CP) DRA that relies on a 3‐D‐printed uniaxial anisotropic ceramic is proposed in the 2.45 GHz ISM band. Its simulated and measured reflection coefficient, axial ratio, and realised gain patterns are in good agreement, thus demonstrating the possibility of exploiting 3‐D‐printed anisotropic ceramics for DRA applications.

Abstract In the field of image processing, tangling noise and artefacts elimination of objects are two essential tasks. Tangling noise and lack of intensity in certain applications also occur at the same time. In this paper, a new variational model is proposed based on total variation and l 0 the norm for simultaneously removing the tangling noise, estimating the location of missing pixels, and filling in them. To be specific, the total variation is used to regularize the estimated image and use the l 0 norm to make the missing pixel to be sparse. Moreover, the data fidelity term is given by a new forward description about the degraded process and the gamma noise assumption. Finally, an algorithm based on the alternating direction multiplier method is exploited to solve the model. By conducting simulated and real experiments, the damaged images can be effectively restored by the proposed method. In qualitative and quantitative terms, this approach works better.

Abstract Polarimetric high‐resolution range profile (HRRP) holds great potential for radar automatic target recognition (RATR) owing to its capability of providing both polarimetric and spatial scattering information. In recent years, deep learning (DL) has obtained state‐of‐the‐art results in many classification tasks and has drawn great attention in the RATR field. However, as one of the most challenging tasks in RATR, small training sample case will restrict the application of DL because its superior performance generally depends on a large number of training samples. A feature‐guided deep model based on Transformer framework is proposed for polarimetric HRRP recognition with limited training samples. In the proposed model, artificial features are introduced to the attention module to guide the model focus on the range cells of HRRP with more target scattering information so as to reduce the dependence of the model on the number of training samples. Several different approaches are also studied to measure the similarity between artificial features and HRRP data to further improve…

Abstract Checkerboard corners are routinely used during tasks such as camera calibration, and accurate detection of them is essential. Traditional methods such as those based on the Harris corner detector are usually affected by image artefacts such as noise and blur. The more recently proposed deep network approach also suffers from the lack of accurate training data. In this article, we make the following contributions: First, we propose a synthetic training data generation method that simulates the real imaging process, with the most notable being that the exact sub‐pixel corner positions in the image become readily available. Second, we design a relatively simple deep network and train it using the synthetic data generated by the proposed method. Finally, because the exact sub‐pixel corner positions can be obtained in the proposed method, this article paves a way for objectively comparing different checkerboard corner detectors in terms of metrics such as the mean Euclidean location error. Experimental results using both synthetic and real data show that the proposed detector…

Abstract Driver behaviour is an important factor in road safety. Computer vision techniques have been widely used to monitor the driver behaviour. The violation of privacy and the possibility of spoofing are two continuing challenges in camera‐based systems. To address these challenges, we propose an efficient approach to monitor and detect driver behaviour based on movement characteristics of the vehicle rather than the visual features of the driver. The main goal of this paper is to classify the driver behaviour into five classes: safe, distracted, aggressive, drunk, and drowsy driving. A lightweight 1D Convolutional Neural Network with high efficiency and low computational complexity is suggested to classify the driver behaviour. Experimental results confirm that our method could successfully classify behaviours of a driver with accuracy of 99.999%.

Abstract The recently invented laterally excited bulk wave resonators (XBARs) demonstrate outstanding parameters suitable for the design of low loss filters for the fifth generation of mobile phones. The operation of XBARs can be interpreted as an excitation of anti‐symmetric Lamb mode A1 by the interdigital transducer. However, it can also be seen as the generation of fundamental plate resonances in a set of parallelly connected quasi‐independent resonators, excited by a lateral electric field created by narrow electrodes with alternating polarity. By exploiting this latter interpretation, we propose a formula for the admittance of multi‐layered membranes, including the main piezoelectric layer of suitable orientation and different additional dielectric layers on both sides of the membranes. The structure can be asymmetric and the excitation of all shear wave thickness resonances, not only with odd numbers, is described. The formula can be used for the one‐dimensional modelling and optimisation of XBARs and the estimation of the resonance frequency dependence on the deposited…

Abstract Here, a new 4:2 approximate compressor is presented by the gate diffusion input (GDI) technique. Although GDI cells suffer from threshold voltage drop, the dynamic threshold approach and carbon nanotube field‐effect transistors are merged to overcome the mentioned problem. The proposed cell has full‐swing outputs, while its error and power delay product are at low rates. Therefore, low voltage multipliers that are used in image processing can benefit from the proposed compressor.

Abstract Identification of bird species from audio recordings has been a major area of interest within the field of ecological surveillance and biodiversity conservation. Previous studies have successfully identified bird species from given recordings. However, most of these studies are only adaptive to low‐noise acoustic environments and the cases where each recording contains only one bird's sound simultaneously. In reality, bird audios recorded in the wild often contain overlapping signals, such as bird dawn chorus, which makes audio feature extraction and accurate classification extremely difficult. This study is the first to focus on applying a blind source separation method to identify all foreground bird species contained in overlapping vocalization recordings. The proposed IVA‐Xception model is based on independent vector analysis and convolutional neural network. Experiments on 2020 Bird Sound Recognition in Complex Acoustic Environments competition (BirdCLEF2020) dataset show that this model could achieve a higher macro F1‐score and average accuracy compared with…

Abstract In this paper, we consider the reconfigurable intelligent surface (RIS)‐aided MIMO systems to realize a high quality link between the base station (BS) and the users via a RIS. To achieve this goal, the active beamforming at the BS and the passive beamforming at the RIS should be jointly optimized. However, most of the existing joint optimization schemes that maximize the sum‐rate have high computational complexity due to the complex derivative calculations and matrix inversion. To this end, a biologically inspired particle swarm optimization (PSO) algorithm is exploited to solve the non‐convex sum‐rate optimization problem with low‐complexity. Simulation results show that the proposed scheme provides a better trade‐off between performance and complexity than existing solutions for RIS‐aided MIMO systems.

Abstract Fatigue driving is the main reason of traffic accidents. The electroencephalograph (EEG) signal can accurately reflect the changes of human physiological state, so EEG is considered as a reliable method to detect fatigue driving. In this letter, the Euclidean space data alignment approach is adopted to reduce individual differences. Then, to consider the spatial correlation between multiple channels EEG signals and further improve the accuracy, an efficient long short‐term memory (LSTM) network structure is designed to detect the fatigue state of drivers. The experimental result shows that the method improves the generalisation capability and achieves the performance with an average accuracy of 95.70% for three categories. Evaluation of the algorithm performance in comparison with other traditional methods indicates that the method has higher accuracy and lower standard deviation. Generally speaking, the proposed method is promising for the application of fatigue driving monitoring.

Abstract In this paper, the first demonstration of simultaneous UV written Bragg gratings and waveguides in germanium and boron‐doped planar silica is presented using a 5th harmonic solid‐state nanosecond laser operating at 213 nm wavelength. The fabrication of high‐quality gratings by using a high peak power density, yielding sufficient uniformity and without any surface damage is demonstrated. The photosensitivity of the doped silica layer is investigated by measuring the local effective refractive index of the optical modes. The written gratings are used to measure grating refractive index modulation, grating detuning bandwidth and the waveguide propagation loss with a minimum value of 0.28 0.07 dB cm. This paper shows that this new generation of pulsed UV lasers is a promising alternative for conventional longer wavelength CW laser sources used in small spot direct grating writing in doped silica.

Abstract We present a multi‐active‐region bipolar‐cascade distributed‐Bragg reflector (DBR) laser emitting around 910 nm. The three active regions and two tunnel junctions are located in a single waveguide core sharing the same third‐order vertical mode. A slope efficiency of 2.25 W/A was measured with a threshold current density of 400 A/cm 2 . Due to the implemented 7th Bragg order surface grating, the laser exhibits a narrow optical spectrum with a width of 0.3 nm and a thermal detuning of 60 pm/K.

Abstract An improved cluster validity index for fuzzy clustering that is able to overcome three intrinsic drawbacks in conventional cluster validity indexes is proposed. The effectiveness of the proposed index is demonstrated by the analysis of its limiting behaviour and comparison with the performance of the Xie‐Beni index, Kwon index, and Tang index on twelve data sets.

Abstract This letter presents a wide dynamic range (WDR) feature extraction (FE) readout scheme for machine vision applications using CMOS image sensors (CISs). The proposed scheme with the proposed pixel structure has two operating modes, the normal and WDR modes. In the normal operating mode, the proposed CIS captures a normal image with high sensitivity. In addition, as a unique function, a bi‐level image is obtained for real‐time FE even if a pixel is saturated in strong illumination conditions. Thus, compared to typical CISs for machine vison, the proposed CIS can reveal object features that are blocked by light in real time. In the WDR operating mode, the proposed CIS produces a WDR image with its corresponding bi‐level image. A prototype CIS was fabricated using a standard 0.35‐μm 2P4M CMOS process with a 320 × 240 format (QVGA) with 10‐μm pitch pixels. At 60 fps, the measured power consumption was 5.98 mW at 3.3 V for pixel readout and 2.8 V for readout circuitry. The dynamic range of 73.1 dB was achieved in the WDR operating mode.

Abstract Electroencephalogram (EEG) signals by virtue of its good time resolution can provide a more comprehensive way for emotion recognition. To further mine the emotion‐relevant information of EEG signals, a kind of innovative and effective feature named mean differential entropy (MDE) is proposed for EEG emotion recognition (EER). The MDE feature extracted from EEG signals supplements the additional potential information of each EEG frame on the basis of time‐frequency information. To verify its effectiveness, extensive experiments are conducted on DEAP (database for emotion analysis using physiological signals) dataset. The experiment results of the proposed methods on public EEG emotion dataset DEAP reached accuracies of 94.56% and 91.84% in valence dimension and four emotion labels with gradient boosting decision tree (GBDT), and 95.15% in arousal dimension with Random Forest (RF), which proved that the MDE feature is helpful to improve the performance of EER.

Abstract For the scenario where the overall layout is known and the obstacle distribution information is unknown, a dynamic path planning algorithm combining the A* algorithm and the proximal policy optimization (PPO) algorithm is proposed. Simulation experiments show that in all six test environments, the proposed algorithm finds paths that are on average about 2.04% to 5.86% shorter compared to the state‐of‐the‐art algorithms in the literature, and reduces the number of training epochs before stabilization from tens of thousands to about 4000.

Abstract In this study, a polarisation‐analysing CMOS image sensor is fabricated for sensitive polarisation modulation detection. Although the image sensor with on‐pixel polarisers can image the incident polarisation collectively, its sensitivity to a weak polarisation change is not high. With the proposed method, an external polariser is used to enhance the polarisation modulation sensitivity with a polarisation image sensor. The performance of this highly sensitive polarisation image sensor and imaging experiments are evaluated using a flow channel.

Abstract We present a record 20‐W continuous‐wave output power membrane‐external‐cavity surface‐emitting laser (MECSEL) double‐bonded to silicon carbide (SiC) heat spreaders emitting around 1178 nm. We achieved a maximum slope efficiency of 27.7% and an optical‐to‐optical efficiency of 26.7%. By utilising an intracavity birefringent filtre, the wavelength was tuned over 75 nm and maximum output power of 12.7 W was recorded at 1178 nm. This study shows the potential of MECSELs to achieve high‐performance and cost‐effective optically pumped semiconductor disk lasers targeting applications in sodium guide stars.

Abstract This letter proposes a novel wearable metasurface antenna (MSA) based on electric ring resonators (ERR) for 5.8 GHz wireless body area network (WBAN) applications. All substrates of the proposed design are made from the semi‐flexible Rogers RT5880 material. To achieve size miniaturization, 6 × 8 ERR elements are placed on a flat surface connected through a corporate‐feed network. The total dimensions of the proposed MSA are 120 × 90 × 1.574 mm 3 (2.2 λ 0 × 1.69 λ 0 × 0.029 λ 0 ). Furthermore, an average efficiency of 79.9%, and a peak gain of 12.82 dBi with a stable radiation pattern have been realized in the operating band. The performance of the proposed MSA is analyzed in close proximity to the human body tissue. The experimental results show good agreement with the numerical data. With all these features, the proposed design is a good candidate for wearable WBAN applications.

Abstract In this paper, the Minkowski fractal is used to design the geometry of a frequency selective surface (FSS) unit cell, which resembles the shapes found in nature. The proposed FSS unit cell has both angular and polarization stabilities of incident wave. For the miniaturization of FSS and decrease of resonance frequency, interdigital capacitances (IDCs) are devised in the FSS structure, which do not eliminate angular and polarization stabilities. The dimensions of the unit cell are 5 mm × 5 mm and miniaturized 0.095λ 0 ×0.095λ 0 . The resonance frequency is specified as 5.75 GHz with 155 MHz bandwidth. An equivalent circuit is derived for the unit cell to evaluate its frequency responses. The effects of variations of the geometrical dimensions of the FSS unit cell on its performance are studied. A prototype model of proposed FSS is fabricated and measured. The performance of its full‐wave computer simulation results and measured data are compared and are shown to be in good agreement.

Abstract This letter presents a new intelligent control scheme for the accurate trajectory tracking of flexible link manipulators. The proposed approach is mainly based on a sliding mode controller for underactuated systems with an embedded artificial neural network to deal with modelling inaccuracies. The adopted neural network only needs a single input and one hidden layer, which drastically reduces the computational complexity of the control law and allows its implementation in low‐power microcontrollers. Online learning, rather than supervised offline training, is chosen to allow the weights of the neural network to be adjusted in real time during the tracking. Therefore, the resulting controller is able to cope with the underactuating issues and to adapt itself by learning from experience, which grants the capacity to deal with plant dynamics properly. The boundedness and convergence properties of the tracking error are proved by evoking Barbalat's lemma in a Lyapunov‐like stability analysis. Experimental results obtained with a small single‐link flexible manipulator show the…

Abstract Nowadays, processing of data via neuromorphic non‐linear dynamics is a key element for brain‐inspired computers. In particular, an appropriate neuromorphic device is essential to emulate neuromorphic functions. In this brief, it is found that the Chua Corsage memristor has rich neuromorphic behaviours, which has potential values to realise the hardware construction for neuromorphic computing. The theory of local activity and edge of chaos are applied to analyse the neuromorphic dynamics of the second‐order Chua Corsage memristor based neuron model, and find that the periodic spikes (action potentials) appear either on or near the edge of chaos domain, which are excited by the subcritical Hopf bifurcation or the unstable state of the neuron model, respectively. Furthermore, the third‐order Chua Corsage memristor based neuron model can generate self‐sustained oscillations, periodic spikes, and chaos near the edge of chaos domain.

Abstract Global optical flow estimation is the foundation stone for obtaining odometry which is used to enable aerial robot navigation. However, such a method has to be of low latency and high robustness whilst also respecting the size, weight, area and power (SWAP) constraints of the robot. A combination of cameras coupled with inertial measurement units (IMUs) has proven to be the best combination in order to obtain such low latency odometry on resource‐constrained aerial robots. Recently, deep learning approaches for visual inertial fusion have gained momentum due to their high accuracy and robustness. However, an equally noteworthy benefit for robotics of these techniques are their inherent scalability (adaptation to different sized aerial robots) and unification (same method works on different sized aerial robots). To this end, we present a deep learning approach called PRGFlow for obtaining global optical flow and then loosely fuse it with an IMU for full 6‐DoF (Degrees of Freedom) relative pose estimation (which is then integrated to obtain odometry). The network is evaluated…

Abstract This letter presents a novel locally active memristor with four‐stable pinched hysteresis loops which has never been reported before. The non‐volatile memory and local activity of the memristor are verified by the power‐off plot and DC V – I plot. A memristive Hindmarsh–Rose (HR) neuron model is constructed by employing the proposed memristor to characterize the electromagnetic induction effect on the membrane potential. Co‐existing multiple firing patterns are explored under different memristor initial states. Interestingly, the firing rhythms of the neuron model can be regulated by adjusting the memristor initial values in different stable regions. Circuit simulations are included to verify the theoretical analysis.

Abstract A high‐speed, room‐temperature quantum cascade laser packaged module emitting at 8.14 μm is presented. The cavity length of the laser is as small as 0.5 mm, and the threshold current at 298 K is 110 mA. A matching printed circuit board is inserted into the packaged module to reduce the RC constant. As a result, the rectification response of packaged module is relatively flat upto 2 GHz and the –3 dB cut‐off frequency is increased from 0.9 to 2.2 GHz.

Abstract An implicit Crank–Nicolson procedure can be replaced with an explicit iteration process. An explicit finite‐difference time‐domain method based on the iterated Crank–Nicolson scheme that has been widely used for solving Einstein's equations is newly developed. The formulation is presented with two iterations and its stability condition is also derived. Numerical results are found to agree well with those obtained from the traditional explicit finite‐difference time‐domain method, showing the validity of the present iterated Crank–Nicolson–finite‐difference time‐domain method.

Abstract Here, a gated‐anode diode (GAD) is proposed where an anode electrode is formed by connecting a gate electrode and an ohmic electrode of a normally‐off GaN HEMT for a 5.8 GHz rectenna. A wide recessed gate GaN GADs were prepared and the recess length dependence of their electrical characteristics was investigated. Typical DC characteristics of the HEMTs are a threshold voltage ( V th ) of +0.3 V and a maximum drain current ( I max ) of 300 mA/mm. The GADs showed the characteristics of maximum forward current ( I f ) of 350 mA/mm, reverse breakdown voltage ( BV r ) of 40 V, and off‐state capacitance ( C off ) of 0.28 pF/mm by using optimized recess length. We constructed SPICE model of the GADs. The SPICE simulation predicted a rectifier efficiency of 81% and a DC output power of 10 W for bridge type 5.8 GHz rectifier using four GADs with each gate width of 0.8 mm.

Abstract When operating in dark and confined spaces, the capacity of the robots to localise in an absolute reference frame is of utmost importance. This letter presents an absolute localisation system, using deep 3D landmarks, for known confined spaces. The system estimates the robot's relative localisation using visual inertial odometry. Local deep 3D landmarks are extracted from the robot's view. Similar 3D landmarks are, also, extracted from the prior map which are then registered with the local landmarks to provide absolute localisation via an extended Kalman filter. To the best of knowledge, deep 3D feature registration has not been used before for absolute localisation. The proposed localisation system is tested within a representative application area—i.e. a structured, confined space—and the results indicate greater accuracy and lower processing time when compared to mainstream 3D registration approaches.

Abstract In this paper, a novel design of high roll‐off frequency selective surface (FSS) is proposed based on a single‐layer substrate integrated waveguide (SIW) cavity. Fully canonical filtering responses with three transmission poles (TPs) and three transmission zeros (TZs) are obtained by employing transversal signal‐interference techniques (TSITs). The FSS unit cell can be regarded as a combined triple‐mode cavity consisting of one cross‐slot resonator and two SIW cavities. By using the filtering coupling matrix, the FSS can be well designed and analysed. Accordingly, to verify the design concept, a prototype is fabricated and experimented. Measured and simulated results are in good agreement, showing high performances in terms of high roll‐off and low profile.

Abstract This work proposes a wearable system for the non‐invasive measurement of the blood glucose level. In particular, the blood glucose variations are related with the skin resistance measured with a suitable sensor. A microcontroller is aimed to measure, store and transmits on a wireless channel, the blood glucose level in a simple, non‐invasive and unsupervised way. The system is particularly useful to monitoring the glucose level in diabetic patients, and to perform the oral glucose tolerance test (OGTT) useful to identify metabolic diseases and to diagnose diabetes mellitus in the early stage. The preliminary obtained results are quite promising and demonstrated the potentialities of the device as diagnostic and home care tool.

Abstract The fabrication and characterisation details of novel distributed Bragg reflector (DBR) diode lasers emitting around 1064 nm are presented here. The AlGaAs epitaxial layer stack used here allows the removal of the active quantum wells in passive sections where no current is injected. The DBR lasers fabricated with a two‐step epitaxial approach without an active layer in the grating sections show improved performance in terms of power, efficiency, and linewidth in comparison to its all‐active DBR counterparts.

Abstract In order to improve the recognition accuracy of graphene pressure sensors, a graphene/silicon pressure sensor array is studied based on backpropagation (BP) neural network. The principle of the pressure sensor array and the workflow of BP neural network are introduced. The 100 groups of training samples ranging from 0 to 1000 kPa are studied based on levenberg‐marquardt (L‐M) optimisation algorithm, and a multiple BP (M‐BP) neural network is designed to improve the recognition accuracy of the pressure sensor array. The training recognition accuracy and test recognition accuracy of M‐BP neural network are about 99.9%. This work plays an important role in the application of graphene pressure sensors in more fields, especially in the solutions for weak pressure detection in artificial intelligence and human‐computer interaction.

Abstract Complex vector fitting (CVF) is a robust method to efficiently and accurately model general linear and passive baseband systems. Such models are important for the analysis and design of modern electronic and optical communication systems. This letter introduces an efficient toolbox implementation of the CVF algorithm. The effectiveness of the code is illustrated on a suitable application example.

Abstract The Internet of things (IoT) paradigm is paving the way to ensure connectivity and monitoring within different market fields. Thus, it is critical to understand the packet transmission performance in Low Power Wide Area Network (LPWAN), especially LoRaWAN. Packet delivery rate (PDR) is considered as a dominant factor of the network performance and its dependency on the signal power. However, previous works do not model the PDR as a function of the received signal power. An in‐depth investigation of the PDR is done by performing an outdoor measurement campaign in the area of the Campus Beaulieu in Rennes. For each received LoRa packet, the effective signal power (ESP) values are obtained as well as their high influence on PDR is proven. With the given results, modelling the PDR using an ESP‐parameterized beta distribution function is proposed. Feasibility of the proposed model is assured by simulating PDR against ESP, hence, the simulated PDR values follow the distribution of the measured ones well. This PDR model gives important guidelines for future LoRaWAN network…

Abstract To enable accurate and reliable indoor localization in a multi‐building environment, a novel constrained convolutional neural network (CNN)‐based indoor localization system (C‐CNNLoc) is proposed using WiFi fingerprinting approach. The proposed network has a sequential structure that firstly classifies a building, followed by estimating the user's location coordinate within the pre‐detected building. Furthermore, the location accuracy is improved by introducing a new loss function that incorporates a penalty term associated with the building boundary. Experimental results illustrate that the proposed method outperforms the existing solutions on the average distance error. The gain comes from that the approach tailored to a multi‐building indoor localization with the sequential structure is prone to successfully correct outliers, that is, predicted location coordinates that lie outside the buildings.

Abstract This letter presents the design of a novel dual‐mode dual‐band bandpass filter that utilises the TM210 and TM020 modes of cylindrical cavity resonators for millimetre‐wave operation when fed with standard WR‐10 waveguide ports. In this manner, the two selected modes of the cylindrical cavity resonators are demonstrated as a single propagation path for fixed dual‐passband responses without the need of cavity perturbations or tuning screws. The passbands are designated for centre frequencies at approximately 102.8 and 110.9 GHz and exhibit a four‐pole Chebyshev characteristic in each of the passbands, which are separated by a transmission zero location. Simulated and measured results of the prototype are presented to verify the design.

Abstract Brillouin optical time‐domain reflectometry (BOTDR) has the advantage of requiring access to only one end of the sensing fibre, which is an important property for some applications. The coherent detection of spontaneous Brillouin scattering allows high sensitivity of temperature/strain measurements, close to one of the well‐known Brillouin optical time‐domain analyser‐based solutions. In‐line amplification based on bidirectional erbium‐doped fibre amplification (EDFA) modules can improve sensing distance and temperature accuracy. This study demonstrates a distributed sensing over 150 km with 5‐m spatial resolution and a temperature error of less than 2°C in 10 min without requiring specific BOTDR adaptation. A record performance of 0.7°C in 60 min has been also obtained, which corresponds to the best performance obtained in reflectometer configuration at this distance.

Abstract Approximate computing can accept errors in computation systems to achieve better utilisation of hardware resources. Approximate computing has been successfully developed in integrated circuits, such as approximate arithemtic circuits and approximate accelerators. Unfortunately, recent works show that approximate circuits also have security vulnerabilities. Hardware Trojan is one of the biggest threats to hardware security. However, very limited research has been conducted on hardware Trojans in approximate circuits. In this letter, two types of hardware Trojans, namely Function‐destructive Trojans and Information‐leaking Trojans, are presented. The Trojans are designed based on the characteristics of approximate circuits. The effectiveness of the hardware Trojans are analysed using the evaluation metrics for approximate circuits.

Abstract Cyclist trajectory prediction is an essential task in autonomous driving and surveillance systems. This task is challenging due to that the bicycles go much faster than the pedestrians and a minor prediction error could lead to a severe deviation in the actual path. Existing cyclist trajectory prediction models usually employ the social pooling mechanism to depict the mutual interactions between targets. They ignore that the pooling operation is leaky in information. Moreover, they prefer to use the recurrent architecture to capture the time‐varying features, which is not efficient in computation and parameter learning. To address these issues, a spatial‐temporal multi‐graph module which employs the topology of graphs to represent social interactions and design multi‐kernel functions to depict the social attributes from various aspects is proposed. Instead of the recurrent architecture, a temporal convolution to forecast the future paths is introduced. Experimental results on real‐world datasets demonstrate its superior performance against state‐of‐the‐art baselines. It…

Abstract SiO 2 film deposition and subsequent high‐temperature annealing resulted in the generation of a two‐dimensional electron gas (2DEG) at Al(Ga)N/GaN hetero‐interfaces, of which the 2DEG was originally fully depleted. The obtained mobilities and sheet carrier concentrations were over 1100 cm 2 /Vs and 3.0 × 10 12 cm –2 , respectively. Surface energy lowering, which is proof of the generated 2DEG, was observed by electron state analysis using hard X‐ray photoelectron spectroscopy. This damage‐less method that selectively generates a 2DEG can contribute not only toward improving some characteristics in existing devices but also toward creating entirely novel devices.

Abstract This study investigates an outer rotor bearingless flywheel motor for flywheel batteries in vehicle that exhibits the advantages of high integration and low loss due to outer rotor direct‐drive and bearingless technologies. A hybrid excitation outer rotor bearingless flywheel motor with four‐degrees‐of‐freedom radial suspension and permanent magnet bias is also proposed, further improving the advantages of high integration and low loss. In addition, suspension force density is ameliorated and suspension–torque coupling is reduced through structure design. To verify the aforementioned advantages of hybrid excitation outer rotor bearingless flywheel motor, its working principles are introduced and its electromagnetic characteristics are analyzed using the finite element method. Comparisons with 12/8 ORBFM are also conducted by focusing on suspension loss and suspension force density.

Abstract InAs quantum dots (QDs) are receiving attention as next‐generation E‐band light source that offers high‐temperature operation and temperature insensitive operation. However, high‐density crystal defects occur at the interface between the InGaAs buffer layer and GaAs, resulting in reduced device performance and shortened lifetime. Here, E‐band QD lasers are demonstrated on InGaAs buffer layer, which suppressed the spread of dislocation by introducing a high‐temperature annealing and a strained layer superlattice filter. In the device, the peak wavelength at room temperature is measured to be 1427 nm and the threshold current density was 440 A/cm 2 . This result indicates that E‐band QD laser structures on low threading dislocation density are promising for the realisation of high‐performance E‐band lasers.

Abstract In this letter, an investigation of the use of a covariance matrix adaptation evolution strategy (CMA‐ES) algorithm is conducted as the phase‐locking method for multi‐channel coherent beam combining (CBC) for the first time. A comprehensive numerical analysis is carried out on the operational performances of the CMA‐ES based phase‐locking algorithm implemented into 7‐ and 19‐channel CBC systems in a filled‐aperture format. Through numerical simulations it can be verified that the CMA‐ES algorithm can readily lead to over 0.90 of normalised beam combining efficiency with appropriate algorithm parameter sets, which can also be optimised by a combinatorial study. The proposed CMA‐ES based phase‐locking algorithm is a feasible option for novel phase‐locking technique for a CBC system particularly when the local extremum issue becomes severe such as in CBC under turbulent atmospheric conditions.