Experimental results are employed to show that a thin metal backed resistively loaded frequency selective surface (FSS) exhibits significant radar backscatter suppression when the structure is illuminated by an antenna placed in close proximity to the periodic array. This is investigated in the range 9–11 GHz by measuring the shift in the resonant frequency of an impedance matched (|S ₁₁| ∼15 dB at 10.22 GHz) microstrip patch which is placed in five different positions ranging from λ/30 to λ/5 distance above the surface of the FSS. Detuning of the antenna by signals reflected from the absorber surface is shown to be much less significant than the upward frequency shift in the measured |S ₁₁| for a metal plate and tissue phantom, therefore this technology could be used to enhance the performance of on-body wireless sensors.

An antenna system composed of an activated patch backed by a ground plane and three square parasitic plates with inhomogeneous loops is analysed using the finite-difference time-domain method. Analysis of an antenna with an infinite ground plane reveals that a tilted beam of ∼60° from the broadside direction is obtained with a gain of ∼18 dBi. For confirmation, an antenna system with a finite-sized ground plane that approximates the antenna characteristics for the infinite ground plane is fabricated and measured. The measured results are in good agreement with the analysis results.

Troposcatter propagation mechanism is valuable for beyond-line-of-sight surveillance. However, channel fading caused by multipath scattering limits the sensing performance of the existing equipment. Based on the troposcatter multipath channel fading model on frequency and spatial domain, a modified covariance-based detector is proposed. The captured array signals are first maximal ratio combined in frequency domain and then weighted in spatial domain before accumulation. The performance of the proposed detector is derived theoretically and verified through simulations, showing significant improvement in the troposcatter scenario.

A novel FIR driver that can be applied to a low power mode in the (Low Power Double Data Rate 4) LPDDR4 standard has been developed. The proto-type transmitter architecture is implemented in 45 nm CMOS process and occupies 0.009 mm² area. The pre-emphasis tap weighting improves the eye opening by 27.9% vertically and 27% horizontally, respectively, and the transmitter consumes only 2.25 mW at 3.2 Gb/s.

Toeplitz matrix–vector product (TMVP) decomposition is an important approach for designing and implementing subquadratic multiplier. In this Letter, a symmetric matrix (SM), which is the sum of a symmetric TM and Hankel matrix, is proposed. Applying the symmetry property, 2-way, 3-way and n-way splitting methods of SMVP is presented. On the basis of 2-way splitting method, the recursive formula of SMVP is presented. Using the two cases n = 4 and 8, the SMVP decomposition approach has less space complexity than 2-way TMVP, TMVP block recombination and symmetric TMVP for even-type Gaussian normal basis multiplication.

The successive approximation register (SAR) ADCs suffer from capacitor mismatch when considered for high-precision applications. Recently, a mismatch error shaping (MES) technique suitable for SAR ADCs has been proposed, by which the DAC mismatch error is first-order highpass shaped. The MES technique to the second-order is generalised. A concrete implementation of the second-order MES is presented. Theoretical analysis and simulations verify the feasibility and validity of the proposed technique. Compared to the previous one, the proposed second-order MES shows significantly improved shaping effect.

An ultra-low power dynamic comparator is proposed with low dynamic offset variation for successive approximation register (SAR) analogue-to-digital converter (ADC). Dynamic offset can be cancelled with the cascode current source. Moreover, the power consumption can be reduced because it has no power consumption during the reset phase. With body-driven technology and cross-coupled inverter, the positive feedback during the regeneration is enhanced, which reduces remarkably delay time. Simulation results in a 0.18 μm CMOS technology confirm the performance of the proposed comparator. It is shown that the fluctuation of the total offset voltage (mean + 3std) is 0.15 and 0.39 mV with common-mode voltage from 0.5V _DD to V _DD at supply 1.2 and 0.6 V through Monte Carlo simulation, respectively. Furthermore, the delay of the proposed structure can be decreased to 1.837 and 118.2 ns at supply voltages of 1.2 and 0.6 V, while the power consumption is only 18.6 μW and 144 nW, respectively.

Previous methods for the construction of a continuous-curvature path have attempted to minimise the path length. A modified motion planning algorithm for a mobile robot that reduces not only the path length, but also the curvature change along the path is presented. Given two points in the plane, the proposed algorithm generates a closed-form solution for the position of a waypoint to connect two clothoid sets. As this method alleviates sudden curvature changes and side-slip by the robot, it improves the overall driving performance. Further, as the curvature change of the generated path becomes smaller, the centrifugal force and accumulated force on the robot also become smaller along the path. Thus, the proposed algorithm is well suited to service robots such as the intelligent wheelchairs used by elderly and disabled people. Simulation results demonstrate the superior performance of the proposed algorithm over the existing generic turn algorithm.

In this Letter, a new criterion based on frozen BER (FBER) is proposed to early stop belief propagation (BP) decoding of polar codes. The authors first verify that the behaviour of FBER along with the iteration is almost the same as the information BER. Based on this phenomenon, the authors propose a new criterion by detecting the status of FBER to determine whether the BP decoding is successful or not. To reduce the complexity, only a part of frozen bits transmitted in most reliable frozen sub-channels are considered. Simulation results show that the proposed criterion is more hardware friendly than the existing criteria.

Aimed to eliminate the lift-off effect on the accuracy of the result of Barkhausen noise detection, this Letter simulated three conditions of the yoke by Ansoft. After studying the simulation results, it proposed a new sensor structure, utilising the mechanical structure of the spring to eliminate the lift-off of the pick-up coil and the BP neural network was adopted to optimise the detected magnetic Barkhausen noise characteristic value. At last, it verifies that the detection error caused by the lift-off of the sensor is reduced effectively.

A triple-mode narrow-band bandpass filter is presented by exploiting three fundamental resonant modes, TE₁₀₁, TE₀₁₁ and TM₁₁₀ modes, in a single air-filled rectangular cavity. Compared with the conventional multimode cavity filter, the proposed filter is to simultaneously excite these three resonant modes without installing any cuts or screws inside the cavity, so as to keep the integrity of the cavity and decrease the fabrication cost. In particular, an inclined U-shaped slot is utilised at each of the microstrip feed line to properly excite these three modes such that not only three poles are achieved within the desired passband but also a pair of transmission zeros are generated in the lower and upper stop-bands. A single metal rectangular filter is finally fabricated and tested. The measured results have well validated the simulated results.

A frequency-tuning crossover based on half-mode substrate-integrated waveguide (HMSIW) is presented for the first time. Two HMSIW transmission lines are parallelly arranged to form the four-port crossover. Three tunable capacitors placed between the HMSIW transmission lines are used to obtain the tunable passband. For verification, both simulated and measured results are given and good agreement between the simulation and measurement is observed. The measured passband can be tuned in a frequency range from 2.2 to 5.0 GHz (tuning range of 78%) with better than 19 dB return loss and 15 dB isolation, while the insertion loss varies from 0.93 to 1.97 dB.

An ultra-thin compact frequency selective surface with dual passband, which can be independently tuned, has been designed. It produces two passbands at 8.2 and 11.4 GHz with transmission zeros on both sides of the passband to improve frequency selectivity. The electromagnetic coupling structure makes the overall structure ultra-thin, and the single-ended short resonator reduces the unit size. Simulation was performed to demonstrate the effectiveness of the design.

Efficient feeding geometries for rectenna design at 2.45 GHz are described. The proposed rectenna consists of two layers low cost FR-4 substrates in which circular polarisation and harmonic rejection properties are embedded. A voltage doubler rectifier with open stub matching network and the radial stub is designed at the bottom substrate to transform the ac signals to a dc voltage. A right triangular slot is designed in the ground plane to suppress second and third harmonics which improves the rectenna conversion efficiency to 75.043% at an input power of 20 dBm. The proposed rectenna design has the advantages of providing circular polarisation and harmonic rejection with a compact size and low cost which makes it a good candidate for microwave power transmission applications.

A radial power splitter is used to distribute signal power to the following stages locating radially around all directions. A well-known disadvantage of a radial power splitter implemented on a planar circuit board is the poor performance in output matching and output port isolation. An eight-way radial power splitter including the proposed ring-shape isolation network is presented. The ring-shape isolation network, comprising several series high-impedance lines and capacitors, plays the crucial role of significantly improving output matching and isolation of the eight-way power splitter. The background principle and design of the ring-shape isolation network are described. An eight-way radial power splitter is designed, simulated and fabricated on a printed circuit board. The measurement results, decently matching the simulations, show the input and output reflection coefficients as low as –20 dB at 1 GHz. The measured output isolation coefficients are below –20 dB at 0.7–1.14 GHz between any two output ports, exhibiting the widest isolation bandwidth among the reported radial power splitters.

A ring resonator loaded with five open stubs is proposed for the design of dual-band bandpass filters. To shrink the filter layout, a pair of loaded quarter-wavelength stubs is utilised to produce the first passband. On the other hand, the introduced coupling between the quarter-wavelength stubs and the ring resonator produced transmission zeros for enhancing the selectivity. Both bandwidth and central frequency ratio of the two passbands can be adjusted freely by the circuit parameters. Two coupling lines are added beside the resonator as feed-lines to construct a filter with high selectivity and compact size. A filter at 1.57/2.45 GHz is designed, fabricated and measured for verification.

An active three-way circulator is proposed to have the tunable property by the tunable transistor feedback network (TTFN). This tunable design shows its tunable frequency range from 1.15 to 1.85 GHz (bandwidth of 47%) with the isolation >20 dB, the insertion losses of around 3.5 dB and return losses better than 10 dB at each port at the tuned frequency. This tunable property is controlled by the varactor diodes added at the input and output of the FET in TTFNs.

A novel self-aligned (SA) graphene FET (GFET) with small access resistance is fabricated. Only one photolithography is needed to define the gate and high gate capacitive efficiency is obtained using a metal gate-stack. In addition, damages to graphene resulting from the plasma are avoided. The cap metal layer is used as an etch stop layer and the etched stem metal layer is used as a support layer, which leads to the simplification of the fabrication process and the formation of the SA structure. Based on the same gate length (3 µm), the normalized G _m of GSA-GFET is 8 times larger than the reported SA-GFET. Compared with the non-SA-GFET, the contact resistance of the SA-GFET is reduced by 50% and G _m is 3.2 times improved.

Recently, the authors proposed a physical-layer key generation scheme for optical orthogonal frequency division multiplexing (OFDM) techniques in an indoor scenario. The key generation protocol exploits bipolar and real-valued optical OFDM samples to create confidential keys. It determines a chunk of cyclic prefix samples positioned in the space of small channel impact to generate the confidential keys during the session duration. This key-generation approach offers strong computational security to counter passive eavesdroppers and key creation on the fly, with zero bit-mismatch rates. In this Letter, the authors evaluate the randomness of the confidential keys produced from the recently proposed key-generation protocol to validate if the keys are truly random. The authors apply a statistical test suite of the National Institute of Standards and Technology to test randomness. The results demonstrate that the keys generated from the proposed key generation system pass all the tests. The results also demonstrate that the proposed key generation system provides uncorrelated keys with an approximately equal probability of bits (discrete uniform distribution) on an average, which yields a high security system against statistical attacks.

Low-temperature-grown GaAs (LT-GaAs)-based Fabry–Pérot cavity photoconductors, designed for RF and THz optoelectronics applications using1550 nm lasers, are studied. The sub-sampling of continuous waves at frequencies up to 300 GHz is presented. The duty-cycle-limited conversion losses measured up to 67 GHz show that this GaAs-based photoconductor behaves as a nearly perfect photoswitch controlled by a 1550 nm pulsed laser.

Index confinement is studied experimentally and through modelling for lithographic vertical-cavity surface-emitting lasers (VCSELs) and contrasted with other types of VCSELs. Modelling shows that the index confinement is set by the height of a shallow internal mesa that produces the optical mode confinement. Overgrowth with a semiconductor mirror enables a range of index confinement from zero (no index guiding) similar to proton-implanted VCSELs to much higher confinement than viable for oxide VCSELs. Lasing spectra are studied for side-mode-suppression ratio and beam patterns are compared for high and low indexes confined VCSELs of different sizes.

A two-switch equaliser for series-connected battery stack using zeta type converter and symmetrical capacitor–diode circuit is proposed. The proposed equaliser consists of a zeta type front-end dc–dc converter and symmetrical capacitor–diode circuit, uses only two active switches and two magnetic components, thus it benefits from small circuit volume and flexible for extension. Meanwhile, it can work under open loop, which makes its control simple. Furthermore, high efficiency is achieved by zero voltage switching of all switches. Finally, experimental results of a prototype of the proposed equaliser are presented to verify the analysis results.

A novel target recognition method is proposed for high-range resolution profiles (HRRPs) of radar targets under low signal-to-noise ratio (SNR) conditions. This method achieves good recognition performance for noisy HRRPs with discriminative sparse-low-rank representation. The framework of this method is constructed based on sparse representation and low-rank representation, which are applied to extract the local and global characteristics of target HRRPs. To guarantee the noise-robust and highly discriminative features of the HRRPs, dictionary learning is adopted. In the training stage, a discriminative dictionary is produced based on hinge loss theory to improve the recognition performance. Denoising dictionary optimisation is implemented for noise suppression during the testing stage. Experimental results on measured HRRP data demonstrate that the proposed method can recover the original HRRPs and significantly improve the recognition performance for HRRP test samples under relatively low SNR conditions.

A motion-compensation method that applies sparse reconstruction (SR) to reconstruct the Doppler spectrum of targets based on a random transmission scheme is proposed for time-division multiplexing (TDM) multiple-input multiple-output (MIMO) radar. Since the random transmission can eliminate the characteristic of periodic time-delay in conventional TDM scheme between transmit cycles, the angle information of a target is not affected by its motion. Therefore, the angle and velocity are no longer coupled with each other and can be estimated separately. This method not only overcome the space–frequency coupling problem but also enhances the unambiguous Doppler interval. Another advantage is that the method is valid even when the estimated target velocity is ambiguous. The results reported here offer the possibility of utilising SR to solve conventional TDM MIMO problems. The effectiveness of the proposed method is demonstrated by experimental results.

A time-variant total electron content (TEC) estimation method with fully polarimetric geosynchronous synthetic aperture radar (GEO-SAR) data is proposed based on inner aperture Faraday rotation angle estimation and an accurate TEC inversion model. With a long integration time and sensitivity to ionosphere effects, the fully polarimetric GEO-SAR data are utilised for estimation with high accuracy for both the time-variant TEC and the time interval. Superiority of the proposed method over conventional ionospheric sounding methods is verified by simulation results.

A novel Trench Insulated Gate Bipolar Transistor is proposed, where an integrated Zener diode is introduced. When the anode voltage becomes relatively high, the Zener diode could automatically clamp the potential of the carrier stored layer which locates beneath the base region of the active trench nMOS. Then in the blocking-state, the dose of the carrier stored layer could be as high as possible to reduce the on-state voltage without affecting the breakdown voltage. In the on-state, the drain-to-source voltage of the trench nMOS is also clamped, which helps to decrease the saturation current. Simulation results based on a 1.2 kV device show that, in comparison with the conventional one, the saturation current and the on-state voltage are decreased by 42.8 and 43.1%, respectively.

The realisation of a fully vertical gallium nitride (GaN)-on-silicon (Si) Schottky barrier diode (SBD) without using wafer bonding and Si substrate removal process is reported. The SBD presented a turn-on voltage of 0.69 V at 1 A/cm², breakdown voltage of 148 V with a specific on-resistance of 13.9 mΩ cm². The ideality factor and Schottky barrier height were 1.35 and 0.71 eV, respectively. An effective critical electric field was estimated to be 1.48 MV/cm. The Baliga's figure of merit for the SBD was calculated to be 1.57 MW/cm². These results indicate the great potential of GaN-on-Si in achieving cost-effective fully vertical power device.

This Letter presents a single-channel speech enhancement technique based on modulation domain signal processing. In the proposed approach, noise-dominant low-frequency (4–16 Hz) temporal modulation envelopes are discarded based on the modulation domain a priori signal-to-noise ratio criterion. Moreover, the authors have employed coherent harmonic demodulation method which ensures distortion-free filtering of modulators. The experiment results in various acoustic non-stationary noise scenario show that the proposed approach can significantly improve the speech quality and intelligibility while compared to the other existing state-of-the-art baseline techniques.

A generalised transmit–receive joint spatial modulation (GTRSM) scheme is proposed, which amalgamates both transmitter- and receiver-side spatial modulation (SM) schemes. The proposed GTRSM jointly employs the combinations of multiple instead of a single activated transmit and receive antennas and baseband symbols to transmit information. The optimal maximum-likelihood detector for GTRSM is formulated, and the channel capacity of GTRSM is also analysed. Simulated results demonstrate that the proposed GTRSM behaves good BER performance and can significantly improve the channel capacity compared with conventional SMs.

A hybrid self-interference (SI) cancellation scheme, composed of a two-tap analogue canceller and a digital multi-path SI canceller, is proposed for single-antenna full-duplex radios. The analogue canceller mitigates the two strongest components of the SI to prevent saturating the receiver. The digital canceller deploys an auxiliary receiver chain to capture the RF-transmitted signal to mitigate the residual multi-path SI signal. As the proposed scheme uses the RF-transmitted signal as reference, the transceiver impairments such as power amplifier non-linearities and phase noise are suppressed along with the SI. Experiments have validated the effectiveness of the proposed scheme on a 20 MHz long-term evolution signal of +20 dBm power.