A novel uniplanar dual-band printed compound loop antenna for WLAN/WiMAX applications is proposed. The proposed antenna consists of a small inner rectangular ring, an outer rectangular loop with three slits and a parasitic strip on one side of the substrate. Fed by the inner ring and collaborated with double rings and the parasitic strip, three resonant frequencies can be achieved, where the parasitic metallic strip between the rings can not only induce a new middle resonant frequency but also improve the impedance matching at the low-frequency band greatly. By magnetic-electrical coupling, dual-band performance with the lower-wide band covering 2.37–3.78 GHz and the upper-frequency band covering 5.15–5.85 GHz can be obtained. It is experimentally demonstrated that the proposed antenna with the performance of broadband (covering 2.4/5 GHz WLAN including 2.4–2.484 GHz, 5.15–5.825 GHz and 3.5/5 GHz WiMAX including 3.4–3.69 GHz, 5.25–5.85 GHz, respectively), good dipole-like radiation characteristics, stable gain and high-radiation efficiency, is an excellent candidate for WLAN/WiMAX applications.

An active metamaterial absorber with electrical control of both polarisation and frequency is presented. The unit cell of the proposed absorber consists of an outer circular ring and an inner Jerusalem cross with two pin diodes mounted between. Tuning the diodes ON/OFF states, the absorption response of the metamaterial structure can be switched between two performances. As biasing the diodes OFF, a dual-band absorption response with polarisation-independent characteristics can be achieved. Although regarding to ON state, a polarisation-dependent performance is exhibited. Specifically, the same dual-band absorption response can be still obtained with respect to the incident wave with electrical field along the x-axis, whereas a different single band of absorption is obtained with electrical field along the y-axis. The simulated and measured verifications are presented and the results demonstrate the proposed controllable absorption performances.

A novel design of absorptive frequency selective surface (AFSS) with tunable absorptive/switchable transmissive property is presented. The AFSS is comprised of a resistive surface and a metallic bandpass FSS on which PIN diodes are mounted. By changing the working state of the PIN diode, a tunable absorption performance at 3.5–11.2 GHz is achieved with a switchable transmission at the frequency of 0.83 GHz. A prototype of the proposed structure is fabricated and measured under the normal incidence.

A wideband monostatic and bistatic radar cross-section (RCS) reduction of microstrip antenna array is proposed and investigated. By using coding metasurface on the proposed antenna, the reflected energy is diffused scattering to more directions. A remarkable monostatic RCS reduction is achieved from 6 to 7.6 GHz and 9.5 to 26 GHz covering the working frequency band, and the bistatic RCS of proposed antenna is also effectively suppressed. The proposed antenna still maintains in a compact structure, while the antenna radiation performance is preserved. Both the reference and proposed antennas are fabricated and measured in anechoic chamber, and the results show that there is a good agreement between simulations and measurements.

A novel capsule antenna with dual-band performance and small occupation of inner capsule space is proposed. The antenna is designed by combining rigid Rogers substrate with flexible polyimide, which takes full utilisation of capsule inner surface and the system ground for implanted electronics. The antenna structure is based on two identical orthogonal strips, each connected with one conformal L-strip. In addition, a shorting pin is introduced, bringing in a fundamental resonance at 1.4 GHz industrial, scientific and medical (ISM) band with linear polarisation. Besides, with the adoption of the shorting pin, two extra resonant modes are excited and split in higher band, generating two orthogonal electric fields with equal amplitude and π/2 phase difference; therefore, covering another ISM band at 2.45 GHz with circular polarisation property. Finally, the simulation results show a wide impedance bandwidth of 16.2% at 1.4 GHz ISM band (for |S ₁₁| <−10 dB), and a broad axial ratio (AR) bandwidth of 34.8% at 2.45 GHz ISM band (for |S ₁₁| < −10 dB and AR <3 dB), with dual-band operation further validated by measurement result.

A method to detect and track objects using an oscillatory neural model is presented that mimics the integrative component from the primary visual cortex to the vision-related parietal and temporal cortex. The locally excitatory globally inhibitory oscillator is incorporated into the proposed model to implement synchronisation and desynchronisation of neural oscillation, and the dual-layer architecture (composed of the form layer corresponding to the ventral pathway and the motion layer to the dorsal pathway) is also introduced to implement the integrated pathways of the human visual process. Objection detection corresponds to a function in the ventral pathway, and tracking of the detected object corresponds to a function in the dorsal pathway. Some experiments where skin regions were detected and tracked are carried out, and showed that the proposed model of the integrative pathways in the human visual process works successfully.

The fast and real-time extraction of precise auditory evoked potential (AEP) is important for the diagnosis of auditory diseases as well as for the brain–computer interface. For fast AEP extraction, the effect of various noises such as motion artefact, eye blink, or power line noise should be minimised in the AEP recording. The existing fast AEP extraction methods that use the Kalman filter or wavelet transform have limitations owing to parametric arbitrariness. In this study, the singular spectrum analysis (SSA) in the time domain was adopted and optimised for AEP extraction with various types of noises. Moreover, the hardware architecture for an optimised SSA was implemented in an FPGA to realise real-time operation. The results show that the optimised SSA method can reduce the stimulus repetition by 61.2% compared with the conventional ensemble averaging and obtain the maximum similarity to the original AEP signal of 83.2%. The designed hardware is favourable for wearable BCI applications in terms of hardware complexity and required clock frequency.

The exact blood vessel trees segmented from fundus images provide important information required for screening and following-up of diabetic retinopathy and age-related macular degeneration. The trained deep neural network presents an automated prediction of the blood vessels in retinal fundus camera images in the publicly DRIVE database with accuracy up to 0.9533 and area under the receiver operating characteristic curve up to 0.9752, which is better than manual recognition by expert human eyes. A resizing technique is introduced and applied to the multi-level network combining dropout and spatial-dropout layers to obtain more generalised training. The proposed model has the potential for the classification of other types of images.

A modulated training pattern for intra-panel interface is proposed and is applied to design power-efficient clock and data recovery (CDR) circuits for intra-panel interface. By modulating the position of the rising edge of the training pattern, the number of the delay cells to generate the multi-phase clock to capture display data safely is reduced. As a result, power, area, and electro-magnetic interference characteristics can be improved over the conventional training pattern. A phase-locked loop-based CDR circuit with the proposed scheme in a 65 nm CMOS technology is designed. The measured lock range was between 6 and 10 Gbit/s and the power efficiency was 0.38 mW/Gbit/s at 10 Gbit/s inputs.

By minimising a new cost function that contains robust set-membership error bound, a bias-compensated robust set-membership normalised least mean square (NLMS) algorithm is proposed, which is characterised by its robustness against impulsive noises and noisy inputs. To estimate the input noise variance in impulsive noise environments, a new estimation method is proposed in which there is no need to know the input–output noise variance ratio in advance. Simulations in a system identification context demonstrate that the proposed algorithm achieves improved robustness and better performance than the existing algorithms.

A novel 2 × time-difference amplifier (TDA) in 65 nm CMOS technology is presented. Unlike traditional analogue TDA, the proposed TDA, employing an all-digital synchronous architecture, achieves 2 × TDA in a time subtracting way based on time register. Multi-path negative skewed gated delay cells are utilised to minimise gating skew error in the time register. To make up for the delay mismatch in the two time registers, a foreground calibration scheme that calibrates the input offset of the phase detectors at the same time is applied. The TDA gain ranges from 1.99 to 2.02 over ± 150 ps input time-difference range and the standard variation of the gain is 0.02 under various process, voltage and temperature (PVT) conditions. The simulated average of power consumption is 62 μW under 100 MHz clock frequency.

This Letter reports the finding of a point-cycle chaotic bursting phenomenon in a simple third-order memristive diode bridge-coupled Sallen-Key lowpass filter (LPF). By parallel coupling a generalised memristor composed of a diode bridge and an inductor into a Sallen-Key LPF a simple third-order memristive chaotic oscillator is established. With the modelling of this oscillator, stability analyses, numerical simulations and hardware experiments are performed, upon which the striking point-cycle chaotic bursting phenomenon is well demonstrated and effectively validated.

Practical applications in digital communication need non-maximally decimated filter banks to provide oversampled baseband signals for other necessary operations. The non-maximally decimated filter banks allow the required prototype filter to have wider transitional band to significantly reduce the required computational complexity. This Letter presents a non-maximally decimated filter bank structure derived by using multi-rate signal flow graphs.

A simplified model predictive current control (MPCC) for four-switch three-phase inverter (FSTPI)-fed permanent magnet synchronous motor (PMSM) considering the periodic fluctuation of the capacitor voltages is proposed. After analysing the inherent features of the FSTPI-fed PMSM system, an offset voltage vector and an ideal reference voltage vector are defined, which allows to deal with the reference voltage in the ideal voltage vector plane. Thus, the sector division is simplified. Furthermore, a concept of virtual duty cycle is introduced to employ an equivalent null vector in the MPCC in order to reduce the current ripples. Combining with a simple sector division, the active voltage vector and its virtual duty cycle are determined with only one prediction. Comparative experimental studies between the conventional and the proposed MPCC are executed to validate the feasibility of the proposed simplified MPCC.

The limitations of sensorless control of permanent magnet synchronous machines (PMSMs) are discussed and a viable solution is proposed. The main concept of sensorless control of drives relies on additional information given by the machine during its normal operation. This information provided by the machine is essentially the back-electro motive force and the variance of the stator inductivity, which are dependent on the rotor position. Several approaches and methods have discussed these problems, and in most cases they are not avoidable and that some methods work better on certain speeds of the drives. The direct flux control (DFC) method to combat the above problems at all speeds is presented. The flux linkage signal which contains the necessary information about the rotor position can be measured between the neutral point of a PMSM and an artificial one. The mathematical derivation and the observations from the experiments show that this signal contains a second and a fourth harmonic, which can be used to calculate the rotor position. Furthermore, the limitations of implementing DFC are also addressed.

Augmented colour difference gradients in the extended support region contribute to the reduction of high-frequency loss in image reconstruction via a demosaicing process. However, distant gradients must only be used for highly inter-channel correlated regions. Otherwise, the pattern grid remains, even after demosaicing. Herein, a demosaicing scheme is proposed, in which the geometric distances as well as the edge directions (as in a bilateral filter) are incorporated in order to calculate colour difference gradients weights. The inter-channel correlation is closely related to the colour saturation. Thus, the weights are adaptively adjusted according to the inter-channel correlation considering colour saturation. Experimental results show that the proposed scheme improves the results of a state-of-the-art demosaicing process.

The low-complexity minimum mean square error detector based on Richardson method for massive MIMO systems was proposed, where the very important relaxation parameter was obtained by intensive simulations for each given massive MIMO channel matrix. Thus, a simple approach is developed to determine the optimum relaxation parameter of Richardson method for any massive MIMO system configuration. Furthermore, by exploiting the massive MIMO channel property of asymptotic orthogonality, a simple way to obtain the quasi-optimum relaxation parameter is proposed, which is given as a function of the number of users and the received base station antennas of a massive MIMO system.

Monte Carlo (MC) methods have become very popular in signal processing during the past decades. The adaptive rejection sampling (ARS) algorithms are well-known MC techniques which draw efficiently independent samples from univariate target densities. The ARS schemes yield a sequence of proposal functions that converge towards the target, so that the probability of accepting a sample approaches one. However, sampling from the proposal pdf becomes more computationally demanding each time it is updated. The parsimonious ARS method, where an efficient trade-off between acceptance rate and proposal complexity is obtained, is proposed. Thus, the resulting algorithm is faster than the standard ARS approach.

A compact two-way unequal Wilkinson power divider with a power division ratio of 2 is proposed. The proposed design can reduce more than 75% of the occupied area, compared with the conventional unequal power divider using composite transmission line. In the proposed design, the total transmission-line length is much less than conventional one. A resistor and a compensative capacitor are used to isolate output ports. The proposed divider is designed, fabricated and tested. The results show that at the designed frequency of 1.5 GHz, the input return loss (s ₁₁) and output return losses (s ₂₂, s ₃₃) are better than 20 dB and the output ports isolation (s ₂₃) is better than 25 dB.

A hepta-band bandpass filter (HB-BPF) based on folded cross-loaded stepped impedance resonator (SIR) was investigated. A cross-loaded SIR microstrip structure was arranged to produce several transmission zeros. In order to reduce the filter size, a folded cross-loaded SIR was proposed. The HB-BPF was designed on FR4 microstrip substrate with ε_r = 4.4, thickness h = 0.8 mm, and tan δ = 0.0265. This HB-BPF achieves transmission coefficients of 0.15, 0.61, 0.45, 1.14, 1.57, 1.77, and 2.08 dB at 0.74, 1.49, 2.25, 3.19, 3.93, 4.68, and 5.50 GHz, respectively. The HB-BPF only occupies 0.1 × 0.22 λg, where λg is the guided wavelength at a first passband centre frequency. A good agreement between simulated and measured results validates the design method.

The first substrate integrated waveguide dual-band bandpass filter with wide-stopband performance is presented. This filter is synthesised and designed with fourth-order Chebyshev responses and identical fractional bandwidths for the two passbands centred at 12 and 16 GHz. Multiple spurious resonances have been intrinsically suppressed by introducing the harmonics-staggered technique and the centred coupling windows. Good agreements can be observed between the measured and simulated results, and better than 25-dB rejection level up to 2.05f ₁ has been successfully achieved.

A real-time processing method for demultiplexing a high-speed terahertz (THz)-wave OFDM signal is proposed and demonstrated. An electrical signal, which is down-converted from the THz-wave OFDM signal, is up-converted into an optical OFDM signal through an optical modulator. Then, the generated optical OFDM signal is demultiplexed into sub-carrier channels in the optical domain by utilising an integrated-optic OFDM demultiplexer with discrete Fourier transform function. The processing principle of the method and some preliminary experimental results including BER measurement to demonstrate the effectiveness of the method are reported. A 2 × 10 Gbit/s THz-wave OFDM signal was demultiplexed into sub-carrier channels with the method as preliminary investigation.

For the first time, the model of a physical nano-scale memristor is integrated analytically. A closed-form expression for the time evolution of the device memristance during the turn-on process is mathematically derived. The complexity of the inverse imaginary error function-based analytical formula clearly reflects the high degree of nonlinearity in the nano-device switching kinetics, which may typically span several orders of magnitude and is critically dependent on input and initial condition. The excellent agreement between the analytical solution and numerical simulation results clearly demonstrates the correctness of the theoretical derivation. The introduction of this formula represents the first step towards a systematic approach to circuit design with memristors.

A passively mode-locked femtosecond erbium-doped bidirectional fibre laser with low pump threshold using chemical vapour deposition grown graphene saturable absorbers is demonstrated. Two synchronised and counter-propagating soliton output pulse trains with a repetition rate of ∼10.38 MHz and pulse widths of 660 and 860 fs are obtained simultaneously from the fibre laser. Simultaneous mode locking of the fibre laser has been realised with a threshold pump power of 56 mW from a single-pump laser diode, indicating an ∼62% reduction in pump threshold compared to previous results.

Two-colour terahertz (THz) quantum-well photodetectors are designed, fabricated, and characterised. To realise two-colour detection, a stacked structure is adopted by sequentially growing two one-colour detectors. It is found that the spectra of the stacked device are in accordance with those of the two individual detectors connected in series. The relatively broad spectral bandwidth may enable its use in a THz spectrometer.

A transient control for a dual-active-bridge (DAB) dc/dc converter to realise power adjustment in a fast and smooth way is proposed. The converter is able to transfer power between two dc sources with two phase-shifts as control variables. To eliminate the possible high-current overshoot and non-zero dc-bias current, the adjustments of the two phase-shifts are distributed to the gating signals of different switch arms according to a ratio x, which has an universal expression for any transitions among different operation modes in a DAB converter. Experimental results have proved the validity of the proposed transient control in elimination of current overshoot and dc-bias current.

In this Letter, a new dual/multiple-input non-isolated boost converter is proposed. It provides a positive boosted output voltage without any additional transformer. It is suitable to interface sources of different current–voltage characteristics to a common load, with a simple topology and a low part count. The topology and the behaviour are similar to the single-input boost converter; hence, the control strategy can be accomplished easily. It can be used in low-voltage energy harvesting applications to collect energy from multiple sources. Simulations and experimental results are presented to demonstrate the effectiveness of the proposed topology.

For the first time, under the actual three-phase 765 kV power transmission lines, electric field energy is harvested and Zigbee-based sensor node powered by the harvested energy is operated autonomously. A conductor tube is placed between power lines and the earth surface for electric field energy harvesting. The voltage difference between conductor tube and the earth surface is rectified and the harvested electric energy is stored in a storage capacitor. A self-sustaining power management circuit adopting low-power comparator with hysteresis is devised to monitor the stored energy and to operate the sensor node autonomously according to the stored energy. Experimental results under the actual three-phase 765 kV power lines show that the data transfer period is almost inversely proportional to the square of the height of the sensor module from the earth surface. The temperature data is transmitted in every 25 s when the conductor tube is located at 4 m from the earth surface. Since the presented non-contact energy harvesting apparatus is very simple, robust and low-cost, it is believed that the presented technology can be widely applied to power transmission, distribution and smart grid system.

A non-intrusive mechanism of gathering real-time performance of inverter data is explained. Network architecture was designed to get this data from various inverters in an area of 200 acres with synchronised time interval for comparing the performance of inverters of different topology and solar panel technologies at same climatological conditions. Ecosystem of things as termed it the Internet of photovoltaics was created at the National Institute of Solar Energy in India, where all the inverter nodes were in the same network, communicating with a centralised server for receiving commands and sending required information.

A variable frequency drive for an induction motor is proposed. The drive uses a power factor (PF) correction bridgeless single-ended primary inductor converter-controlled rectifier operating in discontinuous inductor current mode as a front-end in order to improve the input power quality and a variation of the constant volts per hertz controller, with feedback to regulate the velocity of the motor shaft. The frequency slip is measured and compensated, since the input stage. Experiments with and without load are carried out and presented. Input power quality measurements are also presented. The proposed system is effective to regulate the velocity and achieving a close to unity PF.

A novel CMOS maximum power point (MPP) tracking circuit with high-power throughput, high-tracking efficiency and low complexity for solar converters is presented. The power acquisition circuit guarantees the tracking efficiency and the control logic module which implements the multiplexing of power conversion, power storage and power comparison, simplifies MPPT circuit significantly. Simulation and measurement results show that the MPPT circuit operates properly and the tracking efficiency stays larger than 99.9% with the output current of the power acquisition circuit varying from 5 to 100 μA

The symmetric autocorrelation function (SAF) has a huge computational complexity, which may limit the practical application of SAF-based scaled inverse Fourier transform (SCIFT). Taking the relationship between SAF and summation into consideration, a frequency-domain method is proposed for the fast implementation of the SCIFT, and the energy integration performance remains unchanged. Mathematic analysis and numerical simulations are provided to illustrate the efficiency of this method.

To detect threat signals in electronic warfare support systems, a detector that uses a plurality of windows with various sizes should be designed such that the length of all the signal sources can be considered. Since a large number of these windows cause excessive computational complexity, the number of windows of the detector is reduced by using a small number of representative windows. In this case, since a window is dedicated to the unknown signal of a certain interval, deterioration of the detection performance is inevitable owing to the inconsistency between the lengths of the received signal and the window size. Hence, the deterioration of the detection performance should be minimised by analysing the relation between the lengths of a window and a signal. However, the conventional analysis methods of detection performance are not suitable because they are based on the premise that the lengths of the signal and window are consistent with each other. The authors propose a novel analysis method using processing gain to overcome this limitation, which can be applied irrespective of the inconsistency between the lengths of a window and a signal. Based on this analysis, they present a method to obtain an optimal window length that minimises degradation of the detection performance and subsequently verify the result using simulation.

MgAl₂O₄ (MAO)-based magnetic tunnel junctions (MTJs) with an MAO thickness of ∼1.25 nm are fabricated and their cycling characteristics under dynamic voltage stress are evaluated. The speed of breakdown strongly depended on the pulse polarities used, bipolar, positive (+) unipolar, and negative (−) unipolar. The bipolar condition yielded more rapid breakdown under cycling. Between the two unipolar conditions, positive bias yielded more rapid breakdown than negative bias; the difference between these is understood to arise from the conditions of the interface between the MAO and ferromagnetic layers. Among apparently normal MTJ cells showing little resistance drift, 20% were degraded during a long cycling test in the bipolar stress condition. Thus, the use of bipolar voltage stress is essential to screen for potentially defective MTJs, and the asymmetric condition at the interface is minimised by process control for application of the simple unipolar bias condition.

An ultra-low power frequency-shift keying (FSK) demodulator based on injection-locking ring oscillator for wireless body area networks is presented. The proposed system uses the power-efficient injection-locking ring oscillator (ILRO) to replace the LC oscillator which occupies much more area on chip with higher power consumption. In addition, through the ILRO, the frequency modulated input signal is converted to a full swing rectangular signal, which can be directly demodulated by a chain of down-conversion passive mixers, lowpass filters and a comparator. Power efficiency and simplicity of the proposed structure eliminate conventional FSK demodulator based on power-hungry phase locked loops, which needs to be calibrated by auto-frequency calibrator to periodically compensate the free running frequency deviations due to temperature variations and charge pump leakage current. The proposed FSK demodulator is designed and simulated in 180 nm CMOS technology while consuming only 23.37 µW from a 0.7 V supply.

The multiband detection problem in relatively small sample scenarios where the number of subbands is comparable to the number of samples in magnitude is described. Combined with the sequential hypothesis testing, an eigenvalue ratio based method is proposed for the multiband spectrum sensing (MSS). As the distribution of the new statistic when only noise is present can be precisely obtained by using the random matrix theory, the proposed method is able to reliably set the theoretical threshold, and outperforms the traditional MSS methods based on information theoretic criteria and principal component analysis, especially in the cases with small samples. Meanwhile, the proposed method belongs to a blind detection scheme and can be used in cases without prior knowledge of the primary signal, the channel and the noise. Simulation results show the superiority of the proposed method.

A simple and effective technique is presented to reduce the limit-cycle oscillation (LCO) in the reconfigurable rectifier for wireless power receiver. It is theoretically analysed that only one switching edge of the switching signal is needed to be synchronised and a phase-shift circuit is proposed for synchronisation. Simulation results in commercial 0.18 µm CMOS technology show that the LCO is greatly suppressed with the proposed synchronisation circuit and the output voltage ripples are much smaller compared to the conventional circuit.

Low-complexity minimum-mean-square error (MMSE)-based integrated equalisation (IEQ) for multiple-input multiple-output (MIMO) systems with hybrid automatic repeat request (HARQ) is proposed. In the proposed MMSE-based IEQ, the calculations of the MMSE filtering vector for each symbol in the conventional MMSE-based IEQ are newly derived in a simplified form. This not only allows the proposed IEQ to have the same MMSE filtering vector to the conventional IEQ but it also allows the matrix size to be inverted and the number of matrix inversions are also significantly decreased. Therefore, without sacrificing the error performance, the proposed IEQ requires a significantly lower computational complexity than the conventional IEQ as the number of inputs increases. Furthermore, the proposed IEQ can outperform the MMSE-based iterative equalisation with the log-likelihood ratio-level combining in terms of both error performance and computational complexity.

The performance of maximum-ratio combining for free-space optical systems using heterodyne detection is analytically studied. The analysis includes the joint effect of turbulence- and misalignment-induced channel fading. Particularly, novel closed-form expressions regarding the scintillation index, outage probability and average bit-error rate are derived. Based on these results, some useful outcomes are manifested, such as the system diversity order, the crucial role of diversity branches, and the impact of composite channel fading onto the system performance.

Large-scale antenna array (LSAA) is one of the key factors as well as a fundamental technology in fifth generation wireless communication systems. Beampattern sidelobe (SL) suppressing is a complex nonlinear optimisation problem, especially for LSAAs, because of the high-dimensional solution space. A SL reduction method for LSAA based on a hierarchical cuckoo search (HCS) algorithm is proposed. HCS divides the candidate solutions into different levels to improve population utilisation, so that it is suitable for dealing with large-scale elements. The effectiveness of the proposed method for synthesis of the beampattern of LSAA is verified by theoretical and EM simulations.

In this Letter, the use of the effective radiation pattern (ERP) concept for wireless systems analysis is proposed, in order to account for multipath angular spread effects upon ideal antenna radiation patterns. The method for producing the ERP that optimally models any real antenna radiation pattern is presented and the proof of concept is demonstrated via large-scale simulations for a multicellular beamforming system. Results show that the accuracy in SIR calculations is better than 1 dB.