A novel design for circularly-polarised (CP) arrays is proposed. The proposed array is composed of dual CP slot antenna elements arranged in a face-to-face configuration, and they are simultaneously excited through a slot line. By properly controlling the excitation amplitudes and phases of the two radiating elements, the cross polarisation and back radiation of the array can be effectively reduced. From the measured results, the proposed array can give an axial ratio of lower than 1 dB and a front-to-back ratio of higher than 12 dB within a 10 dB-return-loss bandwidth of >16%. Simulation results obtained with high frequency structure simulator (HFSS) also agree well with the experimental data.

A novel design of high-gain dual-band dual-circular-polarised antenna loaded with reflective metasurface is proposed. The proposed antenna consists of a linear-polarised coplanar waveguide antenna with a polarisation-switched metasurface structure. The metasurface is composed of 25 unit cells arranged in a 5 × 5 layout and each unit cell is a pair of symmetrical semi-circle loop with a diagonal microstrip, which is designed to adjust the axial ratio for dual-band circular polarisation. Simulated result shows that the impedance bandwidth of the proposed antenna is about 63% from 3.8 to 7.3 GHz, and the average boresight gain has a great enhancement of about 4 dBi. The proposed antenna possesses the characteristic of right-handed circular polarisation from 4.15 to 4.45 GHz, while the left-handed circular polarisation is from 5.75 to 6.4 GHz. The simulated and measured results are in good agreement so that the performance of the proposed antenna can be confirmed.

In this Letter, the authors report on the analysis, design, fabrication and measurement of a low windage resistance frequency selective rasorber (FSR) based on windmill-shape coupling slot line arrays. A lossy structure is added to the traditional frequency selective surface structure to make up a FSR with the characteristics of small insertion loss within the passband and weak reflection at frequencies below and above the passband meanwhile. In addition, the hollow structure can reduce windage resistance effectively. The measurements show that an insertion loss of 1.01 dB can be obtained at 7.71 GHz, while both lower and upper passband edges have at least 10 dB reflection reduction under the normal incidence. Both simulation and experiment results exhibit that the proposed FSR is insensitive to both transverse-electric (TE) and transverse-magnetic (TM) polarisation. The proposed FSR can be applied in the field of radome, cloaking and other microwave applications.

Diffractive antennas implement steerable beams based on reconfigurable diffractive devices such as Fresnel zone plates (FZPs). Compared with phased array systems, FZP antennas have concise architectures and simple beam-steering mechanisms. In this Letter, the authors demonstrate an FZP-based reconfigurable diffractive antenna with three degrees of freedom (3DoF), i.e. simultaneous reconfigurable steering angle, beamwidth and operating frequency. Such a highly reconfigurable antenna is implemented based on a single-layer metasurface mounted with microwave varactors. The principle behind this 3DoF reconfiguration is the electrically induced transparency and opacity switched by different sets of binary voltages applied across the varactors. Experimental measurements validated the proposed approach.

A novel approach to realise the wideband and polarisation-independent radar cross-section (RCS) reduction of a slot-coupled patch antenna is proposed. The RCS reduction is based on the unique reflection phase properties of artificial magnetic conductor (AMC) and perfect electric conductor (PEC) cell. By using the quasi-fractal structure in AMC design, 180° ± 37° reflection phase difference between AMC and PEC cells is obtained over 109% (6.4–21.7 GHz) frequency range. Then, the two different unit cells are applied properly to a slot-coupled patch antenna for wideband RCS reduction. The simulated and measured results indicate that the proposed antenna realises a remarkable RCS reduction between 6.0 and 30.0 GHz for both phi- and theta-polarised waves. Moreover, the radiation characteristics of the antenna are well kept simultaneously.

Human echolocation is the ability of an individual (which is often a blind person) to use his/her signal such as sound from tongue clicks to perceive the surrounding. Basically this requires the person to listen and analyse to the return echo of the tongue clicks. The main characteristics of the tongue click signal waveform have been reported, however the fundamental principle on a person's ability to identify his/her own signal is still vague. The possible detection mechanism of the tongue click signal waveform used in human echolocation technique is discussed and imitated it as artificial detection system. The proposed mechanism which is based on human hearing process in synthesising the signal illustrates that the detection performance is improved as compared to the detection performance by the traditional matched filtering technique. The findings of this Letter create new potential for the development of any artificial human echolocator system, sensor systems like radar and sonar as well as applications inspired by human echolocation miracles.

In this Letter, the micro biogeography-inspired multi-objective optimisation (μBiMO) method, an extension of the biogeography-inspired multi-objective optimisation (BiMO) algorithm, is proposed. It is suitable for solving multi-objective optimisations in the field of industrial design because it is based on a small number of islands (hence the name μBiMO), i.e. few objective function calls are required. An analytical test function is used for evaluating the performances of the standard BiMO method and to show the need for the proposed μBiMO algorithm. To test its performances in industrial applications, an optimal shape design problem in electromagnetics, i.e. an electromagnet optimal design, has been chosen as a case study. To the best knowledge of the author, the main element of novelty of the Letter is that the BiMO algorithm has been applied for the first time to a multi-objective design with a reduced number of islands (μBiMO). The algorithm so modified leads to a strong reduction of computational time, and, at the same time, improves the quality of results.

Introduction: A self-biased low-phase noise CMOS LC VCO (employing both nMOS and pMOS switching transistors) based on the trans-conductance linearisation of active devices is proposed. The self-biased push–pull configuration with resistors between the LC tank and device drain, not only reduces power consumption, but also is capable of removing the RF choking inductor needed in NMOS-only or PMOS-only topology, which results into a compact chip area. A capacitive-coupled feedback from the drain and LC tank improves the oscillation amplitude and reduces the tank loading. The proposed VCO is fabricated and measured with a 65 nm CMOS process. The core chip area is 0.12 mm². The measured oscillation frequency ranges from 3.96 to 6.1 GHz, the phase noise at 1 MHz offset is from −123 to −125 dBc/Hz, with a power dissipation from 13 to 23 mW under a 1.3 V supply voltage across the frequency tuning range. The achieved figure-of-merit with tuning range (FoM_T) from 196.5 to 199.5 dBc/Hz.

A building block for computing in memory systems is introduced. Based on the previously introduced racetrack memory proposed by IBM, a racetrack memory is used not only to store data but also to perform bitwise majority-based computations by coupling the memory with inputs provided by controllable magnets. This solution is defined as racetrack logic. Micromagnetic simulations are used to confirm that the proposed solution is technically viable.

The authors experimentally demonstrate successive torus doubling and the birth of strange non-chaotic attractors (SNAs) in a quasi-periodically forced Chua's oscillator. So far it has been believed that the torus doubling occurs only a finite number of times. In this Letter, the authors report experimental observation of a certain kind of torus doubling called swollen shape bifurcation occurs an infinite number of times and leads to the birth of SNA. In the Poincaré of the section, the torus undergoes a series of period doubling and after a particular value of control parameter, the infinitely period doubled torus in the localised regime termed as a chaotic band. While the Lyapunov exponent of the attractor has a negative value and confirms the attractor becomes strange and non-chaotic. Experimentally observed results are confirmed by a Poincaré map, singular-continuous spectrum analysis and the finite-time Lyapunov exponent.

An active contour model driven by novel local and global intensity fitting (LGIF) energies for image segmentation is presented. First, the GIF energy is defined by the squared-chord distance, which is more robust to noise. Second, the LIF energy is defined by the Lorentzian distance to calculate the local intensity information, which improves the generality of the presented model. Experiments are carried out on synthetic and real images and the results illustrate that the presented model can obtain higher segmentation accuracy and better segmentation efficiency; moreover, it is not sensitive to initial contour.

Different from traditional methods depending on the procedure of intermediate ‘cyclopean’ view construction, a novel framework based on saliency-guided multi-scale feature consolidation for stereoscopic image quality assessment is proposed. For quality representation, the underlying features are extracted from three aspects: (i) global natural statistics features, (ii) local spatial and spectral entropy features and (iii) the kurtosis and skew of disparity distribution. Then the binocular features are consolidated by a saliency-guided weighted process. Finally, a machine learning technique of support vector regression is used for objective quality mapping. Experimental results demonstrate the promising performance of the proposed method.

A method for aggregating binary features to represent an image in a single vector is proposed. Binary features are decomposed into several sub-features to permit the individual aggregation of each sub-feature into a histogram. The method is completely quantisation-free, enabling exceptionally fast image representation. Experimental results demonstrate that the method is a competitive alternative for image representation when runtime efficiency is crucial.

Image caption generation has attracted considerable interest in computer vision and natural language processes. However, existing methods usually use convolution neural network (CNN) for extracting image feature and recurrent NN (RNN) to predict next produced word, which may make the obtained features unadaptable to the word generated at current time. A new time-varying parallel RNN (TVPRNN) to deal with this task is proposed. TVPRNN uses two classical CNNs (i.e. VggNet and inception v3) for extracting global image feature, jointly with RNN to obtain time-varying feature at each time step, which are used for representing current word. Visual and textual representation in a multimodal space is fused. Moreover, visual attention mechanism is introduced to guild the proposed network. The approach is evaluated using the benchmark Flickr8k, Flickr30k and MSCOCO datasets. The experimental results show that TVPRNN achieves better or on par with the state-of-the-art methods.

In sparse representation (SR), a test image is encoded by a sparse linear combination of training samples. The L ₁-regulariser used in SR is beneficial to produce a good reconstruction of the test face image with sparse error, but it is incapable to guarantee the robustness against local structural noise. To enhance noise tolerance of SR-based classifier, an improved L ₁-regulariser based on trimmed sparse coding (TSC) by using an extra penalty on correlation among all coding coefficients is proposed. Different from traditional single-coding scheme in SR, multiple coding coefficients are used to represent patches of a test image by its corresponding training patches. The consistency penalty imposed into the new SR model improves the confidence for accuracy classification. Experimental results show the superiority of TSC on two benchmark databases, and it outperforms other state-of-the-art methods.

645-GHz signal generation with a harmonic oscillator based on a 250-nm InP heterojunction bipolar transistor technology is demonstrated. The oscillator is based on the common-base cross-coupled topology, generating a second harmonic signal through the push–push operation. The fabricated oscillator exhibits oscillation frequencies ranging from 561.5 to 645.1 GHz with bias variation. The measured peak output power is –17.4 dBm with a dc power dissipation of 49.3 mW (dc-to-RF efficiency of 0.04%). Additionally, terahertz imaging was successfully demonstrated with the developed oscillator employed as a signal source.

A monopulse sum-difference comparator based on double-layer substrate integrated waveguide (SIW) for X-band applications is investigated and experimentally validated. A 90° 3-dB directional coupler with a double-stepped coupling slot in the common broad side between a pair of folded parallel SIWs and an equal-length unequal-width 90° phase shifter are adopted to form a magic tee. The four magic tees are elegantly interconnected to design a highly compact comparator. Experimental results show the bandwidth (reflection coefficient < −10 dB) of the comparator can achieve 10% with an operating frequency range of 9.5–10.5 GHz, in which the output amplitude and phase imbalance are within 0.75 dB and 14° (RMS), respectively. The insertion losses are about 1.5–2.5 dB and the isolations between the three channels are all better than 20 dB. In particular, the double-layer configuration of the comparator network reduces the occupied area with the potential for high integration and makes it to be a good candidate for miniaturised and compact monopulse antenna systems.

The authors demonstrate a mode-locked Nd:fibre laser working at 910 nm incorporating a phase biased non-linear loop mirror. The broadest spectral width of the pulse is 11.9 nm and the output power is 20 mW with 450 mW pump power. The dechirped pulse width is 206 fs. To the best of authors' knowledge, this is the first demonstration of a 900 nm femtosecond neodymium fibre laser based on a non-linear loop mirror.

A 1.3 μm Q-switched fibre laser using a praseodymium doped fluoride fibre and a polyvinyl alcohol (PVA)/graphene based saturable absorber (SA) is proposed and demonstrated. Stable Q-switched outputs are obtained over a pumping range between 472.4 and 567.2 mW at 1.02 μm. A pulse repetition rate and pulse width of 34.86 kHz and 3.50 μs, respectively, is obtained at a maximum pump power of 567.2 mW along with an average output power of 25 µW and output energy of 0.73 nJ. The Q-switched pulses have a 1.8 nm full width at half maximum and 32 dB signal-to-noise ratio.

A bidirectional DC/DC converter is presented with the advantages of high conversion ratio in boost and buck mode, low conduction loss, recycling the energy of leakage inductor, and zero-voltage switching (ZVS) in some switches. This new topology can realise a higher step-up voltage gain by the using of a coupled inductor. Then an active clamp circuit was applied to reduce conduction loss and recycle the energy of the leakage inductor. The soft-switching technique is applied to achieve ZVS turn-on. The main circuit of the DC–DC converter is analysed in detail and the performance of the proposed converter is verified by the experimental verifications with a 200 W prototype.

A novel non-circulating-current magnetic-integrated three-bridge dual-buck inverter (DBI) is proposed. By adding the third bridge, the proposed inverter can remove two input capacitances and enhance the utilisation of input voltage in traditional dual buck half-bridge inverter (DBHBI). In addition, the partly coupled inductors with non-dotted ends connected are applied to elevate the power density. The dc link diode is utilised to eliminate circulating current caused by magnetic induction absolutely. The generating principle of circulating current and operating principle of the proposed inverter have been analysed in detail. Finally, experimental results are obtained on a 1 kW prototype to verify the practical application of the proposed inverter.

The enormous potential benefits of gallium nitride based power switching devices, only commercially available very recently, in terms of power switching device loss are highlighted. This is first demonstrated through a simulated prediction of loss in multilevel converters, followed by experimental validation. While the simulations focus on losses in multilevel converters, the observations made are relevant in a broad range of applications.

Direct current (DC) flux in power transformer initiates its abnormal operation, and can eventually lead to the transformer damage. A new fluxgate sensor structure to detect the transformer's DC flux is presented. A magnetic shunt path is designed to extract fractional flux from the transformer's core limb. The DC flux can be obtained by measuring the demodulation outcomes of the output signal from the sensor's detecting coil, for there is a monotonic relation between them. The DC flux can be measured by the demodulation of output from the sensor's detecting coil. As verification, the experiment has been implemented with a 10 kVA power transformer. Measurement results show that when the external DC current, used to produce the DC flux in the transformer, varies from 0.2 to 2.5 A, the magnetic modulated output signal changes significantly and the RMS value is monotonously changing with the excited DC current.

The displacement measurement with frequency-modulated continuous-wave radar based on radar interferometry technique has attracted growing interest in recent years. One of the main challenging issues is the static clutter interference in the same range bin, which results in amplitude shrinkage. A novel approach from signal processing side is proposed, i.e. implementing a post-processing procedure based on phasor offset compensation, which can be formulated as a circle fitting problem regarding phasor representation. Both simulated and experimental results are provided to validate the effectiveness and superiority of the proposed technique.

Uniformity, good Hamming-correlation and long period are critical requirements for frequency-hopping (FH) sequence set. A new class of no-hit zone FH (NHZ-FH) sequence set is designed based on chaotic nonlinear map to meet above three demands. Some properties of the proposed sequence set and its performance for frequency-hopping-multiple-access (FHMA) system are investigated. The results show that the proposed NHZ-FH sequence set can offer more robustness to eavesdropper and lower BER for quasi-synchronous FHMA system in comparison with other traditional FH sequence sets.