The use of 3D printing for the development of non-planar electromagnetic bandgap (EBG) structures for antenna applications is proposed. A coplanar waveguide (CPW) fed antenna is tested on a non-planar EBG substrate, fabricated using additive manufacturing techniques. Inexpensive fuse filament fabrication is used as the fabrication process. Silver-loaded conducting ink is employed for the metallic components of the EBG. The CPW antenna on the non-planar EBG structure has a satisfactory reflection coefficient at 2.45 GHz, which is suitable for Bluetooth/WLAN communications. The radiation patterns have reduced back lobes and improved gain compared with the antenna in free space.

A tissue-equivalent liquid has been developed that yields permittivity and conductivity values within ±10% of the IEEE 1528, IEC 62209 and FCC OET65-C target ‘head’ values between 850 MHz and 2.5 GHz. The formulation of the new tissue-equivalent liquid is presented along with measurements of the dielectric properties, thermal properties and density.

Implementation errors of pre-steering (PS) delays lead to severe performance degradation in the generalised sidelobe canceller-based adaptive wideband beamformer (G-AWBF). Therefore, a G-AWBF without PS delays is proposed. To eliminate PS delays, the G-AWBF is designed in the linear space spanned by the generalised steering vector (GSV) which is defined as the Kronecker product of spatial and temporal steering vectors. Based on GSVs, the quiescent weight and blocking matrix are constructed to substitute PS delays. Compared with the conventional G-AWBF, due to the removal of PS delays, an improved signal to interference and noise ratio can be achieved with reduced system complexity. With less adaptive weights and various versions of unconstrained adaptive algorithms being applicable, the proposed G-AWBF outperforms the existing Frost-based AWBF without PS delays in terms of computational load and robustness. Numerical experiments illustrate the effectiveness and superiority of the proposed G-AWBF.

A resonant feeding loop is used to obtain the desired wideband impedance characteristic. Two implementations are presented and their efficiency characteristics are compared: a low-admittance resonant feeding loop and a high-admittance resonant feeding loop. Experimentally it is demonstrated that the low-admittance feeding loop provides better radiation efficiency than the high-admittance feeding resonant loop.

A method using 180° hybrid coupler and vector network analyser to measure the radio frequency identification tag antenna impedance is presented. The impedance of a prototype antenna is measured by different experimental methods and compared with simulation results. The comparison results show that the proposed method has a good accuracy and it is more convenient.

As mobile and wearable health technologies have evolved, constant monitoring of bio-signals from patients has become important for accurate diagnosis using mobile and remote health services. As the need for constant monitoring of bio-signals increases, the amount of bio-signal data also increases, therefore an efficient compression method for specific bio-signals should be developed. In contrast to other types of signals, bio-signals contain important diagnostic information that should not be removed during compression. An effective compression algorithm for phonocardiogram (PCG) signals that contain important diagnostic information (murmur) is proposed. In this algorithm, the murmur is first estimated, then an adaptive thresholding scheme is applied in the wavelet domain to the normal portions and the murmur portions of PCGs depending on the murmur estimates during compression. Although other conventional compression methods result in substantial loss of murmur information, the proposed method is able to keep most of the murmur information in compressed PCGs.

An attempt has been made to analyse the progression of muscle fatigue using mean power ratios (MPRs) derived from B-distribution time–frequency spectrum of surface electromyography signals. For this purpose, signals are recorded from the biceps brachii muscle of 30 healthy adult volunteers. Fifteen subjects carried out isometric contractions and the rest performed dynamic muscle contractions until task failure. The signals are preprocessed and subjected to B-distribution-based time–frequency analysis. Subsequently, the MPR of medium-to-low-frequency band (M/LFB) and high-to-LFB (H/LFB) are extracted from the time–frequency spectrum. Furthermore, the slope is calculated using the linear regression technique. Results show that MPR of M/LFB and H/LFB exhibit a decreasing trend during the progression of muscle fatigue condition. The strength of the peaks present in MPR of M/LFB progressively reduces and approaches zero at task failure for all subjects. Additionally, higher slope values are found in MPR of M/LFB in comparison with H/LFB. It appears that MPR of M/LFB is more sensitive for muscle fatigue analysis and may be used as an index for real-time monitoring in work place.

A sampling and switching method for a binary weighted digital-to-analogue converter (DAC) in successive approximation register (SAR) analogue-to-digital converters (ADCs) is presented. This sampling method is proposed to reduce the switching energy. Thanks to the proposed switching method, only one reference voltage (V _cm = 1/2V _ref) is required which helps to improve the precision of the DAC along with energy reduction compared with those methods that use more than one reference voltage. The switching energy and area of the DAC reduce by 97.66% and 50% compared with the conventional binary weighted DAC.

A Maxwell model is presented to describe the hysteresis in a piezoelectric actuator. The effect of the number of elements is studied and both of the forward and inverse algorithms are presented. Using the inverse Maxwell model, hysteresis is compensated for and an almost linear performance is obtained. Experimental results validate the effectiveness of the proposed algorithm and show that hysteresis nonlinearity reduces from 13.8 to 0.4%.

The core power distribution in the advanced high temperature prismatic reactor (AHTPR 625) belonging to the high temperature gas cooled reactor family is controlled using the constrained receding-horizon control algorithm. The outputs from each region of the reactor core are used as the inputs of the system. The reactor model is based on the two-point-coupled xenon oscillation model. It utilises the nonlinear xenon and iodine balance equations along with the one-group, one-dimensional neutron diffusion equation including feedback with a power coefficient. Axially, the reactor core is divided into two regions. Each region has one input and one output. These regions are coupled with each other. The proposed control method demonstrates very fast response to input flux change in the form of a step and ramp change without any residual flux oscillations between the regions of the reactor core. The constrained receding-horizon control problem involves solving an optimisation problem for finite future samples and obtaining the control with few samples. Among the few constrained control samples only the first one is applied and the process to solve the optimisation problem is repeated again.

2D clustering aims at solving problems concerning bi-dimensional datasets in several application fields, such as medical imaging, image retrieval, computer vision and so on. A novel approach for 2D hierarchical fuzzy clustering is proposed, which relies on the use of kernel-based membership functions. This new metric allows to obtain unconstrained structures for data modelling. The performed tests show that the proposed approach can overcome well-known hierarchical clustering algorithms against different benchmarks, also having the chance to be deployed on parallel computing architectures.

A composite analysis-based model for image restoration by utilising the image edge's well preserving property of the total variation (TV) and the sparsely representation capability of the wavelet frame-based image restoration model is proposed. This model is an analysis-based approach that the wavelet frame coefficient is quite often able to be linked with the smoothness of the image and is more in accord with the model's essence of TV model. This model is solved using the alternating direction method of multipliers fast iterative shrinkage-thresholding algorithm and a faster alternating minimisation algorithm, and numerical experiments show that the proposed method is more suitable for image restoration.

Curves and lines are geometrical, abstract features of an image. Whereas interest points are more limited, curves and lines provide much more information of the image structure. However, the research done in curve and line detection is very fragmented. The concept of scale space is not yet fused very well into curve and line detection. Keypoint (e.g. SIFT, SURF, ORB) is a successful concept which represent features (e.g. blob, corner etc.) in scale space. Stimulated by the keypoint concept, a method which extracts distinctive curves (DICU) in scale space, including lines as a special form of curve features is proposed. A curve feature can be represented by three keypoints (two end points, and one middle point). A good way to test the quality of detected curves is to analyse the repeatability under various image transformations. DICU using the standard Oxford benchmark is evaluated. The overlap error is calculated by averaging the overlap error of three keypoints on the curve. Experiment results show that DICU achieves good repeatability comparing with other state-of-the-art methods. To match curve features, a relatively uncomplicated way is to combine local descriptors of three keypoints on each curve.

A wide dynamic range CMOS image sensor is proposed, exhibiting a linear response at a low illumination level and a logarithmic response at a high illumination level. To maintain high sensitivity and to improve the dynamic range of the proposed pixel, a cascode MOSFET is integrated into the structure of the proposed pixel, along with a photogate. The applied photogate voltage reduces the sensitivity of the proposed pixel under the linear response regime, at a moderate illumination level. Under high illumination levels, the biased cascode MOSFET serves as a voltage-light-dependent active load and is responsible for logarithmic operation. The dynamic range of the proposed pixel can be improved by adjusting the sensitivity and the basis of the logarithmic response.

A novel two-phase practical transmission scheme is proposed to exchange messages between the two users composing the two-way relay channel. These two phases are the listening phase and the dynamic decode-and-forward phase. Then, the achievable rate region is derived in the case that the two users transmit and receive over orthogonal frequency bands. Some numerical results are obtained to confirm the theoretical results.

A simple and effective technique to skip the computation of reliable portions of a frame (windows) for turbo code decoding is proposed. The proposed criterion relies on a very simple approximation of cross-entropy measure by means of thresholding. This criterion features negligible complexity and low-memory requirements. Simulation results show that, in the best case, up to 20% of windows can be skipped with no error-rate degradation. Such a significant computation reduction can be exploited to directly reduce the power consumption as well.

It is well established that the employment of interference alignment (IA) in the interference channel can enhance the achievable degrees of freedom (DOF) significantly. The feasibility of IA in a multi-carrier time-extended interference channel is considered. A new IA scheme is introduced that jointly performs the alignment over space, time and frequency. An asymmetric case is considered where the users are equipped with a different number of transmit/receive antennas. In the wake of this asymmetry, the feasible DoF of each user can be different. A closed-form necessary condition on the feasibility of asymmetric achievable DoF in the proposed scheme is derived.

How dynamic ranges (DRs) of single stages are combined together in a sensor acquisition chain are shown. This is a key feature in sensor design, since it shows a simple way to understand how and where the DR has been enhanced or deteriorated along the chain. The relationship devised is applicable to any kind of transducing processes such as the mechanical one as implemented by micro- and nano-electro-mechanical devices.

A temperature-stable heat pulse driver for low-voltage single supply soil moisture sensors based on junction transistors is presented. The temperature drift of the heat pulse, caused by the variations dV _BE/dT in the transistors placed in the feedback loop, is fully compensated for. A circuit measures the variations dV _BE/dT of the transistors in the feedback loop and generates a current which is proportional to the − (1/R0) dV _BE/dT that, when passed through another resistor inside the feedback loop (R ₁ = R ₀), provides full temperature compensation. The topology of the circuit also provides an output voltage (the voltage used to define the heat pulse) which does not need to be calibrated, since it depends only on a reference voltage external to the feedback loop.

A compact ultra-wideband (UWB) bandpass filter based on a novel multi-mode resonator (MMR) is proposed. The MMR consists of a pair of open-circuited stubs and three pairs of coupled-line sections, which can generate four main resonant modes in the desired band, i.e. two odd modes and two even modes. The measured results of a fabricated one show that its bandwidth can cover the UWB frequency range and it has a sharp skirt due to two transmission zeros near its passband. Return loss in the passband is better than −10 dB. Meanwhile, at least −40 dB suppression can be achieved from 11.86 to 15.13 GHz and −30 dB suppression from 12 to 16 GHz.

A new concept for a narrow-band filtering rat-race coupler is reported, which is constructed by a dual-mode cross-shaped dielectric resonator (DR). By arranging the two modes of the dual-mode DR, the functions for a rat-race coupler are realised, and the bandpass filtering characteristic is integrated. Meanwhile, the single-cavity configuration results in a miniaturised size. An experiment on a fabricated prototype has validated this novel concept.

A novel compact quad-band bandpass filter (BPF) with multi-transmission zeros is proposed. This filter utilises a sextuple-mode impedance resonator to achieve three passbands, and the fourth passband composes of single-end short-circuited resonators. Adopting an embedded and folded structure is employed to realise miniaturisation. Stub–stub coupling is introduced to produce transmission zeros, which can improve the passband selectivity. Finally, a quad-band BPF centring at 0.9/1.45/2.45/3.5 GHz is designed, fabricated and measured. The measured results have good agreement with the simulated ones.

A lattice-form tuneable chromatic dispersion compensator whose number of operable channels is doubled by adding an interleave filter is reported. In this scheme, interleaved signals are introduced into two different inputs of the lattice-form compensator, and are provided with the same dispersion compensation value every half free spectral range. Thus, this configuration can activate inoperable channels that account for nearly half of the total bandwidth in existing lattice-form compensators. The proposed compensator was fabricated by monolithically integrating all components with silica-waveguide technology. The dispersion compensation value of −321 to 299 ps/nm was obtained over 40 GHz bandwidth with a channel period of 50 GHz.

Optical pulse generation based on polarisation modulation and fibre dispersion is proposed that has the ability of generating multiform optical pulse trains by applying a single electro-optic modulator and a low frequency driving signal. In the experiment, using a 10 GHz RF source, the generation of a 20 GHz time and polarisation interleaved optical pulse train, a 10 GHz single-polarisation pulse train and a 20 GHz single-polarisation pulse train is demonstrated. By using differential group delay in a polarisation maintaining fibre to realise time-division-multiplexing, the repetition rate of the pulse train can be further increased. Based on this principle, the generation of a 40 GHz time and polarisation interleaved optical pulse train, and a 40 GHz single-polarisation pulse train is successfully implemented.

A C-band GaN high power amplifier (HPA) microwave monolithic integrated circuit (MMIC) with second harmonic tuned circuits is presented. The MMIC is designed with three stages to ensure high gain, and the final stage is matched with optimised second harmonic impedance to improve the power added efficiency (PAE). Experimental results show that the class AB GaN HPA MMIC, with drain voltage of 28 V, can be realised with more than 40% PAE and 60 W output power, with associated gain of 25 dB, in 5–6 GHz at 100 μs pulse width and 10% duty cycle. The chip area is 3.2 × 5.3 mm (16.96 mm).

The frequency-domain analysis of the performance of a wind turbine that is placed in a mountainous complex terrain is presented. The variables of interest were the wind speed and the active power of the wind turbine, and the experimental data for the analysis were collected for three months with a sampling time of 10 minutes. The power spectral density (PSD) of the wind speed and of the active power of the wind turbine were obtained for every day of the three months under analysis, and this information was used to estimate the magnitude squared of the frequency response of the system and to estimate the average value of the magnitude squared of the frequency response. Then, the energy distribution of the wind turbine over a range of frequencies was analysed.

A W-band 2.5 GHz bandwidth pulse generator comprising a voltage-controlled oscillator (VCO) and a novel pulse former in a 65 nm CMOS technology is presented. The proposed sub-harmonic pumped pulse former, which plays roles as both a frequency doubler and a pulse former, is designed to obtain high efficiency using a transformer-based current reuse circuit. The VCO is implemented with a parallel combining transformer, which makes it insensitive to load variations due to pulse formation. Measurements show that the pulse generator achieves −1 dBm peak output power at 79 GHz. It can generate 800 ps pulse signals with an RF carrier of 76.1–79.6 GHz, achieving the energy cost rate of 50 pJ/pulse with a 1.25 GHz pulse repetition period.

Compared with sequential single-input–multiple-output array, a novel design method of high-efficiency sparse array is proposed to achieve the similar azimuth resolution and lower grating-lobe level. Theoretical analysis and numerical experiments demonstrate that the designed array is irredundant and the efficiency is increased. The work can benefit the development of high-resolution imaging and low-cost ultra-wideband radar.

This Letter presents preliminary results of the analysis of amplitude statistics of wind turbine clutter as extracted from multistatic radar data. It is shown that the T-location-scale distribution provides good fitting of the experimental data, and that there are combinations of bistatic angle and polarisations where the bistatic clutter has more favourable statistics for target detection than the simultaneous monostatic clutter.

Here a novel mainlobe interference suppression technique based on eigenprojection algorithm and beam pattern similarity constraints is proposed. This method can effectively suppress the multiple mainlobe interferences and sidelobe interferences. The output of adaptive beamforming is calculated via adaptive weight vector obtained by the convex objective function and processed echo data based on the eigenprojection matrix. The simulations have been presented to evaluate the theoretical analysis.

Ferroelectric (Pb,La)(Zr,Ti)O₃ capacitors were fabricated using chemical solution deposition with Al:ZnO (AZO) top electrodes that were deposited using pulsed laser deposition (PLD), where the oxygen pressure was varied systematically. The oxygen pressure during deposition of the AZO layer affected the surface morphology of the AZO top electrodes, as well as the ferroelectric properties of the capacitors. As the oxygen pressure increased, the AZO grains gradually appeared clearer in the SEM images, indicating less dense stacking, and the polarisation–voltage hysteresis loops expanded horizontally. The largest values of remnant polarisation and coercive voltage were obtained at 10 Pa. Appropriate ferroelectric properties were obtained for oxygen pressures in the range of 0.5–2.0 Pa. The hydrogen degradation resistance during annealing in 3% H₂ 200°C and 1 Torr was independent of oxygen pressure during PLD.

An artificial neural network (ANN)-based noise model of MOSFETs for breakdown operation is presented for the first time. This ANN architecture for avalanche noise model determination is developed using a feed-forward back propagation neural network program with optimisation algorithm of Levenberg Marquardt. Good agreement between measured and simulated noise parameters is achieved in the breakdown region. This simple and accurate breakdown characterisation method based on the ANN can be applicable to noise modelling for devices operating in the breakdown region.

This Letter presents an antenna selection criterion to reduce Rician distributed self-interference (SI) effects in full duplex radio. The presented criterion can minimise the effects of residual SI, which is caused by channel estimation errors, by maximising the desired signal power to the residual SI and the noise ratio. The presented approach using numerical examples is verified.

A method of enhancing performance is described on the Cramer–Rao lower bound (CRLB) of time difference of arrival (TDOA) and frequency difference of arrival (FDOA) measurements under multiple emitters. The TDOA and FDOA measurements between two receivers can be simultaneously estimated by using the cross ambiguity function (CAF). However, in the case of multiple emitters, these measurements are contaminated by the inter-symbol interferences (ISIs) which deteriorate the estimation quality. Therefore, it is required to reduce the effect of ISIs for enhancing the performance prior to calculating CAF. For separating the interference signals from the desired signal, we select the angle of arrival estimation algorithm with multiple array antennas. The performance of the estimator can be expressed as CRLB and the separation probability is calculated by employing the locally most powerful test. From the numerical evaluation results, it was observed that the performance could be significantly improved especially for the situation under heavy ISIs.

This Letter reports on new experimental high dynamic range sub-sampling performances obtained from nanoscale microwave photoconductive switches operating under a 0.8 µm wavelength pulsed illumination. Experimental results have been performed on different samples benefiting of a low temperature grown Gallium Arsenide absorbing layer on GaAs substrate, coplanar microwave electrode profile optimisation, and the use of an ultra-low noise jitter optical source delivering optical pulses of 1 to10 ps pulse width at a 2 GHz repetition frequency at 0.8 µm optical wavelength. Thanks to carrier lifetime tunability achievement from 1 to 20 ps, in association with a carrier mobility of 3900 cm²/V.s, experimental demonstrations of 54 and 45 dB signal-to-noise ratio microwave signal sampling for signals at 5,76 and 20 GHz frequency bandwidth respectively confirms submicron devices functionality as a new class of single channel photonic analogue to digital converters.

A frequency-domain implementation of the soft constraint satisfaction multimodulus blind algorithm is proposed. This leads to a significant reduction in the computational complexity, thus making it suitable for broadband wireless systems. The convergence of the developed algorithm is then improved by normalising each of the frequency bins in the weight update. Simulation results support the superior performance of the proposed algorithm over its time domain counterpart in terms of faster convergence rate.

To meet the extra traffic demand of hot-spot (HS) users not expected in the original network planning, it is desirable to deploy new small cells on top of the existing heterogeneous network (HetNet) without replanning the overall network. It is proposed to maximise the minimum user throughput in a HetNet with unexpected recurring HSs by jointly optimising the number and locations of new small cells and user associations of all cells. A reduced-complexity iterative algorithm is devised to solve the joint optimisation problem. The simulation results show that the proposed iterative algorithm significantly outperforms the random deployment of new small cells and achieves performance very close to numerically solving the joint optimisation in terms of minimum user throughput and required number of new small cells, especially for a large number of unexpected HS users.

Mobile phone usage has changed significantly over the past few years and smartphone data usage is still not well understood on a statistically significant scale. This Letter analyses 2.1 million smartphone usage data values and explore the current wireless downlink–uplink demand asymmetry for different time periods and across different radio access networks. The current data demand over 2G networks remains largely symmetric with strong temporal variations, whereas the demand over 3G networks is asymmetric with surprisingly weak temporal variations is shown here.

Dynamic load modulation is a technique for increasing RF power amplifier efficiency by manipulating a tuneable matching network (TMN) between the RF transistor and the antenna in harmony with the envelope of the transmitted signal. Generally, TMNs are based on varactor diodes with control voltage swings approaching 100 V, but future devices are likely to need far greater. The varactor driver amplifier proposed here uses a split frequency architecture to amplify a 3 MHz bandwidth 3GPP 4G long-term evolution envelope signal. Under these conditions, it can produce a peak voltage of 166 V at a power consumption of 0.7 W.