In this Letter, a compact dual-band frequency selective surface (FSS) with the smallest size of 0.047λ ₀ × 0.047λ ₀ for global system for mobile communication system is proposed. This FSS element is composed of two 2.5D square loops, which creates two stop bands resonating at 940 and 1820 MHz with −18 dB bandwidth of 105 and 208 MHz, respectively. Besides, the proposed FSS performs a stable frequency response under different incident angles for the horizontal polarised electric field (transverse electric mode) and vertically polarised electric field (transverse magnetic mode). The detailed design process is illustrated in the Letter. To confirm the full-wave simulation results, a prototype is fabricated and the measured results show good agreements with the simulated ones.

This letter presents a high interport isolation, compact dual polarised slot coupled monostatic patch antenna with integrated tunable analogue/RF self-interference cancellation (SIC) circuitry for 2.4/2.5 GHz In Band Full Duplex wireless applications. The presented antenna deploys hybrid feeding for improved interport isolation through polarisation diversity and integrated single-tap RF SIC circuitry provides additional isolation on top of antenna isolation. Brief mathematical description for deployed single-tap RF SIC is also presented. The implemented prototype of the proposed antenna module provides around 80 dB interport isolation for 20 MHz bandwidth and better than 97 dB peak isolation when measured in a lab in the presence of environmental reflections. Moreover, 20 MHz SIC bandwidth with 80 dB interport isolation can be tuned within antenna's 10 dB return loss impedance bandwidth of 60 MHz.

Heart rate variability (HRV) can be used to assess autonomic control activities. Among the HRV analysis methods, the integral pulse frequency modulation (IPFM) model demonstrates the functional mechanism for the cardiac pacemaker that can explain the generation of heartbeats. In this study, the IPFM model was implemented into a low-cost very-large-scale integration (VLSI) chip, and the active spectrum of HRV was then estimated by solving for the nerve-modulation signal in the model based on the compressed sensing (CS) method. The proposed IPFM chip utilised a time-sharing scheme for hardware resource sharing, a look-up table to calculate the sine and cosine operations, and multiplexers to control the data flow. Furthermore, based on the hardware resource sharing scheme, the proposed chip achieved a low-cost and high-speed design. Implemented with TSMC 90 nm CMOS technology, the proposed IPFM core had an operation frequency of 62.5 MHz with only 9.8 K gate counts. The spectrum of HRV could be effectively estimated through the VLSI implementation of IPFM based on CS.

The optimisation of inductive telemetry link for increasing wireless range of a medical implant device is studied in detail. An active implant device using separate wireless power and data links at 2 MHz and 420 kHz has been fabricated and tested. A test setup consisting of RF power transmitter and data reader hardware has been implemented. Measurement results show that the fabricated device achieved a wireless range of 25 cm.

An alternative approach to the design of super class AB recycling folded cascode operational transconductance amplifiers is proposed. Instead of using local common-mode feedback for boosting dynamic currents, simple current mirrors with input transistors entering triode region for large currents are employed. Measurement results from a 0.5 μm CMOS test chip validate the proposal.

A graphical control system design approach is proposed in this Letter considering time delay margin and time-integral performance criterion. For two free design parameters, time delay margin boundaries are plotted into 2D parameter plane by using parameter space approach. The details of obtaining design equations are introduced. A time-integral error criterion is calculated for gridded design parameters and shown on the same plot to obtain contour curve representation of the error values. An application example is selected from automotive control literature to test the effectiveness of the proposed approach with simulations. Simulation results are obtained for different benchmark controllers and compared with each other in terms of time domain step response characteristics and error values.

The main concern of user-guided segmentation (UGS) is to achieve high segmentation accuracy with minimal user interaction. A novel convolutional neural network (CNN)-based UGS method is proposed, which employs a single click as the user interaction. In the proposed method, the input image in the Cartesian coordinate system is first converted into the polar transformed image with the user-guided point (UGP) as the origin of the polar coordinate system. The transformed image not only effectively delivers the UGP to the CNN, but also enables a single-scale convolution kernel to act as a multi-scale kernel, whose receptive field in the Cartesian coordinate system is altered based on the UGP without any extra parameters. In addition, a feature selection module (FSM) is introduced and utilised to additionally extract radial and angular features from the polar transformed image. Experimental results demonstrate that the proposed CNN using the polar transformed image improves the segmentation accuracy (mean intersection over union) by 3.69% on PASCAL VOC 2012 dataset compared with the CNN using the Cartesian coordinate image. The FSM achieves additional performance improvement of 1.32%. Moreover, the proposed method outperforms the conventional non-CNN-based UGS methods by 12.61% on average.

This Letter reveals the feasibility of decomposing a matrix into three component matrices for image recovery. Image recovery emerges in many areas, such as image processing, computer vision, and pattern recognition. Recently, the low rank assumption-based image recovery methods catch the researcher's attention. The authors assume the real data matrix has low rank and the error matrix is sparse. However, they are limited to the low-rank component being exactly low-rank, and the sparse component being exactly sparse. Either or both these assumptions are not exactly satisfied in practice and should be relaxed. This Letter presents a tri-decomposition method for dealing with the image data corrupted by both large sparse noise and small dense noise. The method parts the observed data into the clean data, sparse noise, and dense noise by different measure functions. Extensive experiments on face images and surveillance videos demonstrate the effectiveness of the proposed method.

A kernel-based discriminative sparse representation method (KDSR) via l ₂-norm regularisation for robust face recognition is proposed. Sparse representation in the original sample space is usually a linear representation which does not consider the non-linear relationship of samples. To overcome this limitation, KDSR first maps the sample into high-dimensional feature space via kernel tricks and then performs the discriminative sparse representation scheme in high-dimensional feature space. KDSR can capture the non-linear relationship of samples and contains more discriminative information of samples, which makes it have good classification performance. In addition, KDSR has a closed-form solution, which makes it computationally efficient and easy to apply in practice. Experimental results on three benchmark databases demonstrate that KDSR can achieve better recognition performance than many state-of-the-art methods.

Facial expression recognition is one of the most important tasks in human–computer interaction, affective computing, computer vision, and related work. Feature additive pooling and progressive fine-tuning of the convolutional neural network (CNN) for facial expression recognition in a static image are introduced. Network is proposed that partially employs the visual geometry group (VGG)-face model pre-trained on a VGG-face dataset. The characteristics and distribution of the facial expression images in each database are biased according to the purpose of the publicly available facial database used. To alleviate this problem, a CNN model is developed that merges progressively fine-tuned CNNs into a single network. Experiments were carried out to validate the presented method using facial expression images from the Cohn–Kanade +, Karolinska directed emotional face, and Japanese female facial expression databases, and cross-database evaluation results show that the method is superior to state-of-the-art methods.

Geometric matching is a key step in computer vision tasks. Previous learning-based methods for geometric matching concentrate more on improving alignment quality, while the importance of naturalness issue is argued simultaneously. To this end, a novel homography geometric matching architecture with homography prior is proposed. Specifically, two choices for different purposes in geometric matching are provided. When compositing homography prior with affine transformation, the alignment accuracy improves and all lines are preserved, which results in a more natural transformed image. When compositing homography prior with thin-plate-spline transformation, the alignment accuracy further improves. Experimental results on Proposal Flow dataset show that the proposed method outperforms state-of-the-art methods, both in terms of alignment accuracy and naturalness.

Recently, the stereo matching task has been dramatically promoted by the deep learning methods. Specifically, the encoder–decoder framework with skip connection achieves outstanding performance over others. The skip connection scheme can bring detailed or in other words, residual information for the final prediction, thus improves the performance, which is successfully applied in many other pixel-wise prediction tasks, such as semantic segmentation, depth estimation and so on. In contrast to other tasks, the authors can explicitly obtain the residual information for stereo matching, which is achieved by back-warping the right image and calculating the reconstruction error. The reconstruction error is successfully used as unsupervised loss, but has not been explored for skip connection. In this Letter, the authors show that the reconstruction error in the feature space is very helpful to bring residual information for the final prediction. They validate the effectiveness of using reconstruction error for skip connection by conducting experiments on the KITTI 2015 and Scene Flow datasets. Experiments show that the proposed scheme can improve the performance by a notable margin and achieves the state-of-the-art performance with very fast processing time.

An efficient variable bit data hiding scheme based on combined chaotic system (CCS) and integer wavelet transform (IWT) is proposed. With the aim to maximise the security level, CCS is developed by combining two 1D chaotic maps. The developed combined tent-logistic (CTL), combined sine-logistic (CSL) and combined sine-tent (CST) maps with improved chaotic behaviour are utilised to generate the key sequences. These chaotic key sequences from CTL, CSL and CST are then quantised to embed the secret bits in high-high, high-low and low-high sub-bands of the IWT transformed cover image, respectively. The main advantage of this scheme is that the number of bits to be embedded in each and every sub-band coefficients is highly chaotic and extremely sensitive to the initial seeds.

A feature-based log parsing method is presented for extracting log events from unstructured free-text logs. It is a data-driven log analytic solution with no training data needed and suitable for various types of log parsing tasks. Experiments show that the proposed method can achieve higher accuracy and lower time complexity in large-scale log data than existing log parsing methods.

A frequency tracking loop based on triangle function frequency discriminator and adaptive loop bandwidth adjustment is proposed to track the Doppler frequency of weak direct-sequence spread spectrum (DSSS) signal. The proposed structure first provides a frequency discrimination with wide pull-in range to lock the large residual frequency offset after acquisition. Then, the adaptive loop bandwidth adjustment is employed to improve the accuracy of frequency estimation. Compared with the conventional tracking loop, the proposed structure has wider pull-in range and higher sensitivity. Simulation results verify the validity of the proposed method.

A broadband, compact on-chip contact pad to microstrip transition for use in silicon (Si) based technologies operating up to 330 GHz has been investigated and fabricated. The pad is designed as a short coplanar waveguide structure to concentrate the electrical field in the narrow gap between signal and ground pads, thus eliminating the need for a substrate shield while preserving low interaction with the lossy silicon substrate. The working principle is confirmed by electromagnetic field simulations using a detailed model of a high-frequency measurement probe, which allows to investigate the field distribution with a realistic excitation source. The fabricated test structure with two transitions in a back-to-back configuration shows excellent broadband properties and insertion loss below 4.2 dB up to 330 GHz, making the proposed structure well suited for millimetre-wave applications.

An electro-optic filter based on guided-mode resonance for optical communication has been designed and fabricated. The rigorous coupled-wave analysis is used to calculate the reflection spectra of the designed filter. The grating structure of the filter is patterned by interference lithography with two ultraviolet laser beams. The polymer-dispersed liquid crystal is injected into the gap between the glass substrate and the bottom of the grating to fabricate the electro-optic filter. The measured spectra of the fabricated filter exhibit a 38.7 nm shift of the resonant reflected wavelength by controlling the applied voltage.

Owing to economical and environmental concerns, the photovoltaic (PV) technology has been significantly developed over the past few years. To ensure the safe operation of PV systems, it is necessary to develop effective fault detection techniques. Low-fault current of high-impedance, low-mismatch, and low-irradiance faults may lead to they remain undetected, resulting in potential fire hazard and energy loss. The use of blocking diodes and operation of maximum power point tracking algorithm increase the difficulty of PV fault detection. Using the Teager–Kaiser energy operator, a novel index to detect the challenging PV fault conditions without any prior information about the PV array and the training data set is proposed. The proposed index can differentiate the string-to-ground, string-to-string, and open-circuit faults from partial shadings in both grid-connected and islanded modes of operations. Simulation results in MATLAB/Simulink environment validate the performance of the proposed fault detection index.

Nested array (NA) has received much attention for providing more degrees of freedom compared with the traditional uniform linear array. An off-grid compressed sensing (CS) method is proposed using atomic norm minimisation for NA to identify the unknown direction-of-arrival of multiple sources, which overcomes the drawback of grid-mismatch effect of the conventional CS algorithms such as orthogonal matching pursuit, and achieves better performance than the subspace methods. Simulation results verify the effectiveness of the proposed method.

A reweighted -norm minimisation algorithm for high-resolution direction-of-arrival (DOA) estimation under unknown mutual coupling is proposed in this Letter. First, the proposed method forms a new block representation model by parameterising the steering vector. Then, a reweighted -norm constraint based on the new data model is proposed, in which the principle of a novel multiple signals classification (MUSIC)-like algorithm is used to construct a weighted matrix. Finally, the DOAs can be achieved by the recovered sparse matrix. Owing to the use of the whole received data and reweighted procedure, the performance of the proposed method is better than the state-of-the-art methods. Extensive simulation experiments confirm that the above inference is correct.

In this Letter, a compact equation for calculating energy sub-bands inside III–V gate all around nanowire MOSFET is developed taking into consideration the penetration of the wave function into the gate oxide and the effective mass discontinuity at the semiconductor–oxide interface. The values of the sub-band energies result from solving Schrodinger's equation in cylindrical coordinates is expressed in Bessel functions. The authors use an approximation for Bessel functions with the introduction of one fitting parameter. The results show very good agreement with self-consistent Schrodinger–Poisson solver data.

Flexible protonic/electronic hybrid rectifying devices gating by chitosan electrolyte were fabricated on polyethylene terephthalate (PET) substrate. indium zinc oxide (IZO) source/drain electrodes with a self-assembled IZO channel were prepared by RF sputtering with a nickel shadow mask. The IZO channel was modified by protonic/electronic coupling and drain current (I _D) exhibited rectifying characteristics by short connecting source electrode with bottom indium tin oxide (ITO) film. Such devices were used to mimic inhibitory synapses. Synaptic behaviours, like paired-pulse depression and long-term depression were successfully mimicked in flexible artificial inhibitory synapses. Such inhibitory synaptic devices have potential application for serving as a building brick for future brain-inspired computation systems.

The first report is presented on the RF operation of AlGaN channel transistor with Al-composition above 80%. The reported device is a polarisation-graded field effect transistor (PolFET) featuring an ultra-wide bandgap (UWBG) AlGaN channel with Al-composition grading from 65 to 82%. Transistors with a gate length of 0.8 μm were fabricated, with a current density of 265 mA/mm, the highest observed for AlGaN channel transistor with Al-composition above 80%, and f _T/f _MAX of 5.4/14.2 GHz. This demonstration of RF operation of polarisation-graded transistor provides a direction for achieving ultra-wide bandgap AlGaN based devices for high performance RF and power switching applications.

This Letter deals with the problem of seeking an electromagnetic signal source using a group of autonomous robots by designing an event-triggered communication scheme. First, the finite difference method based on radial basis functions is used to estimate the gradient of the scalar field generated by the signal source. Second, a consensus controller is proposed to control multiple robots to seek the signal source along the gradient ascend direction. At the same time, an event-triggered communication scheme is designed such that the burden of communication can be lightened. Finally, simulation and experimental results show the effectiveness of the proposed event-triggered communication scheme for the problem of seeking an electromagnetic signal source.