Ceramic (Ba_0.55Sr_0.45Ti_1.01O₃) – polypropylene polymer ER182 composites-based materials were applied for sub-THz range antenna lens application in telecommunications. Typical plano-convex -shaped lenses were simulated and measured with a standard rectangular waveguide at 220–330 GHz frequency band and applied on 150 mm on-air distance. The lens fabricated with ER182 polymer material increased the signal strength by 15 dB, ER182/7 vol% BST by 6 dB and ER182/30 vol% BST by −23 dB. Material loss tangent values were 0.008 for ER182, 0.034 for ER182/7%BST and 0.081 for ER182/30%BST. The directivity of ER182 material lens and WR3 waveguide combination was 26 dBi at 300 GHz.

Parabolic equation models discretised with the finite-difference method have been a research topic for a long time. Through the simulation of electromagnetic wave propagation characteristics in typical scenarios, it is pointed out that the fast algorithm needs to be developed for the solution of two-level Noye–Hayman (NH2) implicit method to meet the accuracy and computational requirements of electromagnetic wave propagation problems in three-dimensional environments. In this Letter, hierarchical skeletonisation is used for an efficient inversion of three-dimensional parabolic equation discretised by the NH2 implicit method which is the modification of conventional Crank–Nicolson (CN) method. Numerical results show that the NH2 implicit method is more accurate than the CN method at no additional cost for a given numerical example.

Providing an easy-to-use and non-invasive heart rate estimation system for home healthcare is imperative for detecting health deterioration in the early stages and for providing preventive treatments. To this end, we propose a ballistocardiographic method that estimates heart rate by analysing face videos and exploiting cyclic head motions caused by blood pumping through the carotid arteries during heartbeats. The proposed method addresses hand motion artefacts that occur while tracking head motions on a mobile platform. This is achieved by training an auto-encoder that retrieves a heart rate-related signal by learning consistent and statistically dominant relationships between facial landmarks. Moreover, unlike conventional methods, where a feature point tracker was used to track the facial landmarks, we propose using region-based tracking to increase tracking robustness in challenging lighting conditions. Experimental results demonstrate that the proposed method performs best in challenging light conditions and can robustly manage hand motion artefacts on a mobile platform.

A kind of impedance transformer with high impedance ratio based on tapped coupled-resonator filter theory is proposed. Differently with the published ones, the proposed filtering transformer incrementally takes impedance property of the filter into consideration. Thus, the high impedance ratio can be achieved. By proper design of impedance and type of resonator, two prototypes with impedance ratios as high as 400:50 and 800:50 and 3-dB fractional bandwidths of 20.2 and 22.8% are realised.

This communication describes a gate leakage compensation technique to ensure wide-temperature operation and wide-supply regulation of traditional self-cascode based voltage references in deep nanometre nodes (). Extensive simulations in 28 nm ultra-thin buried oxide fully depleted silicon-on-insulator CMOS process show that the proposed 4-transistor topology can operate down to 0.2 V, generating a reference voltage of 98.2 mV with mean and best temperature coefficients of 68.9 and , respectively, from to .

This Letter presents an electromagnetic modelling of an active integrated antenna array for 5G mobile communication. Here, the authors propose a functional method for active-integrated antenna array synthesis, based on the method of moments combined with the generalised equivalent circuit (MoM-GEC) in conjunction with a genetic algorithm for the synthesis of a 2D array with switched phase states. Their design strategy is applied to search for the optimal phase matrix with a directional beam and minimum side-lobe level.

This Letter presents a 29-/35 GHz dual-band high-gain hybrid π-network-based phase shifter. The hybrid π-network realised by combining electrical tuning through capacitors and magnetic tuning through transformers extends the phase shift range. Furthermore, the phase shift modules are inserted between the vertically stacking transistors to realise the embedded phase shifting with the minimum loss. The Gm-stages offer the additional signal gain to offset attenuation from the passive phase shifter. This prototype, fabricated in a 28 nm CMOS process, demonstrates a 360° continuous phase shift range with a maximum gain of 22 dB at 35 GHz, and 240° continuous phase shift range with a maximum gain of 20 dB at 29 GHz. It consumes 26 mW power and 1.25 mm × 0.75 mm chip area.

In this Letter, a compact and high-performance ultra-wideband (UWB) bandpass filter based on half-mode substrate integrated waveguide (HMSIW) is proposed. A novel slow-wave structure with lowpass characteristics is developed and loaded on the HMSIW to obtain the UWB frequency response and realise size-reduction. Moreover, two rectangular slots are etched on the transition structures of the microstrip-to-HMSIW to obtain a good UWB impedance matching. As a verification, a prototype is designed, simulated, fabricated and measured. Good agreements can be observed between simulated and measured performances.

A new compact multilayer dual-band balanced-to-balanced filtering power divider (FPD) based on substrate integrated waveguide (SIW) resonator is proposed in this Letter. Originating from the field distributions of resonant modes in the square SIW cavity, TE₂₀₁-mode and TE₂₀₃-mode are utilised to achieve desired differential-mode (DM) transmission and high common-mode (CM) suppression. Meanwhile, dual DM passbands are generated by reasonably coupling two SIW resonators in the back-to-back form with the help of four coupling apertures etched on the middle-layer ground. Besides, a pair of isolated resistors is loaded at back-diagonal regions of the upper-layer SIW resonator, achieving high port-to-port isolation. For demonstration, a prototype dual-band balanced-to-balanced SIW FPD operating at 11.0 and 18.4 GHz is ultimately designed, fabricated and tested. Good agreement between simulated and measured results is obtained, verifying the feasibility and validity of the design concept.

Low profile broadband substrate-integrated waveguide to rectangular waveguide (RWG) transitions have been proposed for RF verification of W-band automotive radar. The transition is designed on a PCB, which is assembled between the radar board and instrument waveguide port. In the design of the transition, a magnetoelectric (ME) dipole is introduced to excite the basic mode of RWG, and a bowtie slot is designed to excite the ME dipole with a broadband performance. The measured frequency band of back-to-back transitions is 68.4–85.3 GHz for a 5 mil RO3003 covered multilayer radar board and 68.6–89.3 GHz for a 10 mil RO3003 covered board, respectively. The calibrated single transition loss is 0.58 and 0.33 dB, respectively. The proposed low profile transitions are robust to assembly errors and convenient for fast RF verification of automotive front-end in the design stage.

A balanced substrate-integrated waveguide (SIW) filter based on a virtual electric wall (VEW) is presented and studied in this Letter. By introducing VEW on the symmetry plane for replacing the metallic-via-array, conductor losses from metal sidewalls can be avoided. As proof-of-concept example, a six-pole balanced filter using asymmetric inductive windows is designed and fabricated with a measured centre frequency of 29.45 GHz, as well as fractional bandwidth of 7.2% to cover satellite ground terminal application. The measured response shows a good concordance with the simulated one, indicating that the proposed balanced filter has the advantages of high common-mode suppression and low insertion loss.

In this Letter, an active area optimisation technique to improve the performance parameters of the film bulk acoustic resonator (FBAR) is proposed. The active area of the back trench membrane-based FBAR is optimised to remove the spurious modes, higher harmonic modes, and to confine the acoustic signal at its central part during the resonance. The effect of thickness variation of SiO₂ layer on the performance parameters was studied using finite-element analysis (FEA) simulation. The SiO₂ film, on silicon substrate, was used as the support layer and zinc oxide was used as the piezoelectric film for the resonator. The authors have successfully demonstrated the FBAR through FEA for an optimised active area of 320 × 320 µm², series resonance frequency (f _s) 1.249 GHz, and parallel resonance frequency (f _p) 1.273 GHz with an effective electromechanical coupling coefficient of 4.65%.

This Letter describes a method to quantify online and in real-time, the level of degrading of the series parasitic resistance R _s of a photovoltaic (PV) panel. This method aims to obtain the herein proposed M-factor, which results from the calculation of the slopes of the I–V curve in the boundaries of its maximum power-point. For this purpose, the slight modification of the maximum power-point tracker controller’ perturb and observe algorithm is required. This procedure can be easily integrated into its firmware with no additional costs. The M-factor calculation is performed instantaneously and since no disconnection is required, no significant power losses are produced.

This study has been made to improve the detection capacity of radar using phase coded waveforms especially the Golay set sequence while keeping the sidelobe ambiguity function to the possible minimum. This Letter begins with a description of Golay complementary sets construction from a graph and a determination of the parameters on which they depend to be optimised after. Then, the simulated annealing (SA) algorithm has been proposed to deal with low ambiguity problem of the Golay set, due to its simplicity and also to its effectiveness to solve this kind of problem. Simulation results show that the optimisation with SA has impressively improved the sidelobe rejection of Golay set sequences with a gain of up to 20 dB.

A GaN-based gate driver circuits have been successfully designed and fabricated using a commercially available 6-inch GaN-on-Si platform. The driver circuits consist of three-stage direct-coupled FET logic (DCFL) inverters featuring monolithically integrated depletion-mode (D-mode) and enhancement-mode (E-mode) high electron mobility transistors (HEMTs). At room temperature (RT), at a supply voltage of 4 V the single-stage DCFL fabricated on the same die exhibits a large gate swing (3.84 V) and large noise margins (logic-low noise margin (NM_L) of 1.55 V and logic-high noise margin (NM_H) of 2.18 V), both of which are desirable for all-GaN IC applications. Meanwhile, the gate driver circuits were characterised up to 200°C on a PCB with a capacitance load of 1 nF in series with 4.7 Ω resistance to resemble the load condition. At 200°C, the gate driver circuits function well even at 5 MHz operation frequency. The turn on/off propagation delay and rise/fall time of the gate driver circuits are 5/40 and 22/18 ns, respectively. As far as the authors’ knowledge, this is the highest reported operation frequency for GaN-based gate driver circuits under such high temperature, making it very promising for high-temperature high-frequency all-GaN integrated circuit (IC) applications.

AsTeGeSiN threshold switching devices, which have recently gained attention for their reliable operation, are investigated for the degradation mechanism under constant voltage stress (CVS). Threshold switching devices are essential for 3D stacking of high-density memories. The degradation mechanism can be divided into two conduction regions, i.e. before and after the threshold voltage (V _th) by applying CVS. It is found that charge trapping, which can easily be recovered, is the main degradation mechanism in the low-conduction region. This is because Poole–Frankel conduction by deep traps occurred in the low-conduction region. However, trap generation may occur when going from deep traps to shallow traps, which is an important factor in the high-conduction region stress voltage. Switching stress for different duty cycles in two conduction regions is also performed. Therefore, when operating an AsTeGeSiN threshold-switching device as a selector, the read sequence is not a problem despite frequent usage. However, for the write sequence, reducing the number and duration of pulses helps extend the device's lifetime.

A new coupling scheme of a 4.28-THz quantum-cascade laser is constructed by employing two off-axis parabolic reflectors. Two parallel terahertz beams are coupled out from the symmetrical facets of the wire laser. The total effective laser power of the terahertz beams is measured by a thermal power-meter under the modulation state. The highest value of the laser effective output power is about 1.4 W at a driving current of 19.3 A and a duty cycle of 1%. This dual-beam coupling scheme makes full use of the output power of the wire laser and has important applications in terahertz imaging and biological-effect research studies.

In this work, monolithic mode-locked GaSb-based lasers at 1.7 µm emission wavelength with heterodyne measurement of the frequency comb are presented. Mode-locking operation was confirmed by measuring the optical beatnote at the round-trip frequency of 9.56 GHz and analysing the frequency comb. Mode-locking operation occurred from 0.1 to 3.3 V reverse bias voltages at the absorber section and driving currents from 160 to 400 mA at the laser section with a minimum beatnote width of 6.6 kHz. The frequency comb spans over 13 nm with a mode spacing of 0.10 nm, measured by heterodyne-detection with a distributed feedback laser.

An iterative adaptive approach (IAA) based algorithm is presented to realise coherent integration for the random pulse repetition interval radar. The proposed method firstly removes the effect of Doppler ambiguity via phase compensation, and then accomplishes target focusing via conducting the azimuth IAA and range inverse fast Fourier transform. Theoretical analysis and numerical experiments have verified the effectiveness of the proposed method.

As a typical signal processing problem, direction-of-arrival (DOA) estimation has been adapted to a wide range of applications in radar-based systems. A high DOA resolution requires a large number of antenna elements which increases the overall cost. To minimise the cost, it is desirable to choose an optimum sub-array from a full array. To enable cognition, the subarrays are selected based on the present target scenario. By using deep learning (DL) based techniques, the authors show a cognitive sparse array selection technique. By using hardware simulations, they demonstrate the applicability of the deep learning (DL)-based sparse antenna selection network in direction-of-arrival (DOA) estimation problems. They show that the DL-based sub-arrays lead to a higher direction-of-arrival (DOA) estimation accuracy by 6 dB over random array selection.

Coprime arrays can highly increase degree-of-freedom by operating the second-order virtual signal. However, the corresponding virtual array is discontiguous, and thus most existing direction-of-arrival estimation approaches fail to utilise all received information. To avoid performance loss, the authors intend to convert the discontiguous array into a contiguous one in this Letter and propose two virtual array interpolation algorithms for the coprime array from the perspective of signal reconstruction. By promoting the sparsity of the signal via simplified atomic norm minimisation and enhanced matrix completion, respectively, the two algorithms can effectively reconstruct the signal of the interpolated virtual array in a gridless manner. Simulation results are presented to demonstrate the performance advantage of the two proposed approaches.

In this Letter, a new scheduling scheme for belief propagation (BP) is proposed to reduce the decoding latency. All the stages update messages at the same time, it improves the reliability and accelerates the decoding convergence. The authors also remove the last stage of factor graph to reduce computational complexity. Simulations show that the average decoding latency of a (1024,512) polar code is reduced to 29.4 clocks.

The mobile network operators should expand their network capacity as service traffic increases in mobile networks. It is important to balance network capacity and traffic volume in terms of service quality and operating costs, and this network balance can be quantified by the capacity margin, which is defined as the difference between network capacity and traffic volume. There are two approaches to assess the network capacity: single-cell and national scale. However, the single-cell approach has too narrow to estimate the whole operator network, and the national scale approach should consider regional traffic density distribution. This Letter proposes a methodology for assessing the capacity margin from the perspective of traffic-intensive regions where the actual network balance is determined. The authors introduce the concept of a virtual sample region that consists of traffic-intensive cells in the metropolitan area. Then they show the methodology to estimate and predict the network capacity and traffic trends in this region. Finally, they derive the capacity margin of the mobile network. They used the real measured data in the top 10% traffic-intensive cells about three years in Seoul, South Korea, and evaluated 4G downlink capacity margin in this region until 2024.

Owing to the rapid and facile deployment of unmanned aerial vehicles (UAVs), the third-generation partnership programme has recently shown considerable interest in UAVs for assisting current terrestrial base stations. Hence, the feasibility of using UAVs has been vigorously investigated for enhancing the performance of traditional ground-based communications, e.g. coverage and quality-of-service enhancement under natural disaster or overwhelming data requirement conditions. Here, however, the energy-efficient operation of the UAV is necessary to obtain the essential benefits. To strengthen the base of operation and management of UAVs in cellular systems, the authors consider in this Letter a single mission in which a small-size UAV transmits a given data volume to a cell boundary user. They design the transmit power and altitude of the UAV to complete the mission with high-energy efficiency. From the numerical results, it is verified that the UAV transmitting data with the designed transmit power at the designed altitude can complete the mission with near-optimal, i.e. maximum, energy efficiency.