Sixteen-chain sequentially rotated fed antenna array on low-temperature cofired ceramic substrate is presented. A 60 GHz broadband circular polarised (CP) stacked rectangular patch antenna is implemented as an element. All the feedlines have the same length which is differentiated from conventional T-junction divider feeder. By this way, several drawbacks of conventional sequential rotated array can be overcome such as phase error occurrence, amplitude mismatch, and axial ratio bandwidth degradation. Moreover, digitally controlled by phase shifters, beamforming performance can be evaluated. The simulated and measured S ₁₁ shows broad bandwidth from 57 to 66 GHz. In addition, the radiation patterns show a peak gain of 17.2 dBi and a wide coverage of ±63°.

A novel dual-band directional antenna with comb radiators using substrate integrated suspended line (SISL) technology is proposed for wireless local area network (WLAN) applications. The proposed SISL antenna is designed to be self-packaged using five substrate layers with embedded air cavities. H-shape strip and rectangle slot are used to achieve the band from 2.43 to 2.54 GHz with gain of 2.2 dBi. Via is utilised to connect the antenna patch with feeding line, then the second band from 5.33 to 6.22 GHz with gain of 10.1 dBi is obtained. Additionally, the cross-polarisation levels less than −15 dB and front-to-back ratio larger than 15 dB are achieved.

A compact printed antenna array for dedicated short-range communication (DSRC) at 5.8 GHz is presented. The proposed device is based on the sequential rotation arrangement (SRA) applied to a parasitic loaded patch antenna. Despite the compact dimension, the proposed array presents a left-hand gain of 10.5 dB at centre frequency, as well as an axial ratio equal to 0.5 dB and a return loss in excess of 16 dB. Therefore, the device is adequate to operate according to the European framework of DSRC for traffic monitoring in the C-band.

A study of printed cross-dipole antenna with slow-wave parasitic elements used to achieve high efficiency and directivity is presented. The proposed antenna is designed to operate at WLAN channels (5.15–5.85 GHz), and can produce dual-orthogonal polarisation through a crossed structure. The printed slow-wave elements play a role that concentrates on travelling wave and delays propagation of electromagnetic wave. Measurement results of this antenna show that it has high gain, low side-lobe level and wide bandwidth. This proposed design has advantages in terms of manufacturing the antenna easily and optimising the radiation characteristics.

An excitation technology of TE₃₀ mode in substrate integrated waveguide (SIW) and its application for designing a compact Yagi–Uda antenna array is proposed. Coplanar waveguide (CPW) structure is adopted as the input and two slots of CPW are further extended in the SIW along the third lines in the longitudinal direction to convert the electromagnetic field pattern between the slot and SIW. Therefore, TE₃₀ mode is successfully excited. A Yagi–Uda antenna array fed by the SIW operating at TE₃₀ mode is fabricated and measured. A broad impedance bandwidth of 46% (|S ₁₁| < −10 dB) ranging from 7.8 to 12.4 GHz is achieved with obtaining a maximum gain of 8.5 dBi. Compared with the traditional SIW antenna array based on TE₁₀ mode, the higher mode excitation technology not only simplifies the feeding network but also reduces the difficulty of processing.

A wideband vertically polarised omni-directional antenna, which is characterised with a height of 0.1 λL (λL: the wavelength at low frequency), a relative operating bandwidth of 48% (1.25–2.04 GHz), stable gains and improved cross-polarisation, is presented. A pair of λ/4 resonators is introduced to broaden the bandwidth; a symmetrical hook-shaped probe is employed to improve the cross-polarisation; the shorting pins between the patch radiator and ground plane are used to reduce the dimension of the antenna. To clarify the improved cross-polarisation, a comparison between the radiation patterns in asymmetrically and novel symmetrically feeding antennas is shown. Measured and simulated results agree well with each other, indicating good performance of wide bandwidth, low profile, stable radiation patterns and gains. Moreover, the polarisation purity of the proposed antenna has been significantly improved compared with that of the asymmetrically feeding antenna.

The use of a suitable propagation model for coverage prediction is key during the planning and dimensioning process of a digital terrestrial television (DTT) system. This study optimises one of the most detailed models for this purpose, i.e. Recommendation ITU-R.P.1812-4, with the aim of improving its accuracy in propagation path loss prediction for DTT systems in outdoor environments of Caracas city, Venezuela. For model adjustment, it is used the bioinspired technique particle swarm optimisation (PSO) and received signal level measurements performed in DTT systems in Caracas city under fixed reception conditions. The results, quantitatively analysed by RMS error and grey relation grade and mean absolute percentage error, shows a significant increase of the accuracy of optimised model in comparison with the original version and even with others propagation models.

A new correctional low-dispersive finite-difference time-domain (FDTD) method based on hexagonal lattices (CH-FDTD) is introduced. The proposed method adopts scaling factor for the permittivity and permeability, which can generate nearly the exact phase velocity for a single frequency. Compared to the conventional FDTD method based on hexagonal lattices (H-FDTD) and rectangular lattices (Yee-FDTD), the numerical dispersion of the new method is significantly reduced.

The implementation of the feeding network in the arrays that composed of a large number of elements and using non-uniform excitations is a real challenging issue. The complexity of the feeding network can be greatly simplified by using uniformly excited arrays. However, such arrays suffer from sidelobe limitation. An optimised approach for sector sidelobe nulling in the equally spaced linear array pattern with uniform excitations except for two edge elements is presented. The genetic algorithm and particle swarm optimisation are independently used to find the optimal values of the amplitude and phase excitations for those edge elements.

A 3T-2R non-volatile ternary content-addressable-memory (nvTCAM) is proposed. Using a voltage limiter and self-controlled bias circuit, both faster match line development and more sensing margin were possible when compared to the conventional nvTCAM. The voltage limiter provides a clear distinction between mismatch cell and ‘don't care’ cell. In case of nvTCAM made of non-volatile memory (NVM) with a large resistance ratio, the sensing delay is reduced by 43.7% thanks to the combination of the self-controlled bias circuit and voltage limiter. The proposed circuit works properly with NVM with resistance ratio as low as 3. The proposed nvTCAM cell was evaluated using a 65 nm CMOS process.

WK-recursive based networks well conform to a modular design due to the properties of regularity and scalability. Here, the focus is on a triplet-based WK-recursive topology and its implementation using tiled shapes in multi-core processors design is explored. To further improve the network throughput and reduce the network latency, a pair of heterogeneous topologies is proposed to separate the core communication from the memory communication. It is demonstrated that the triangular topologies are feasible to be implemented with rectangular tile floor-planning.

A new ring-oscillator-based true random number generator (RNG) based on multiple sampling is proposed. The proposed generator uses the outputs of all gates in a ring oscillator, as positive- and negative-edge clock signals. The generator in two field-programmable gate array families is implemented, and it is verified that the generated bit sequences pass the RNG tests of the National Institute of Standards and Technology. The experimental results show that the proposed generator is faster and more compact than the existing ring-oscillator-based true RNGs.

In this Letter, a highly reliable automotive integrated protection circuit for human body model electrostatic discharge (ESD) of + 6 kV with an over voltage of 8.2–16 V and a reverse voltage of −16 to 0 V is presented. In the automotive application, the reliability of the electronic device is important. In order to increase the reliability, a protection circuit is proposed to preserve the chip from over or reverse voltage. In contrast to the conventional comparator-based protection circuits, only MOSFET and resistors are applied in this Letter. Especially, at the reverse voltage protection mode, the potential zeroing method is applied to block the latch up and break down in the MOSFET. In addition, to protect the chip from static electricity, the ESD is combined with the protection circuit. This circuit is implemented on chip in a 0.18 µm CMOS technology and the area of the fabricated circuit is 680 × 330 µm. The proposed circuit is operated to protect the chip from external voltage between 8.2 and 16 V or between −16 and 0 V.

Worst-case end-to-end delay analysis is important for real-time applications of switched Ethernet network. The Trajectory approach is one of the mature tools for worst-case delay analysis to calculate the upper bound of delay. The delay analysis based on the Trajectory approach as well as the packet serialisation of such networks is presented. An optimisation of packet serialisation is proposed in order to improve the Trajectory approach. A case study is illustrated to compare the improved Trajectory approach with the classical one, and the result shows that the improved Trajectory approach can get tighter upper bound of end-to-end delay.

A 2.5 GHz Class-C voltage-controlled oscillator (VCO) with dynamic back-gate-biased MOSFET is proposed. A dynamic gate biasing circuit is used to reduce power consumption by switching N-type MOS from initial Class-AB to Class-C operation in steady state. Moreover, the VCO uses dynamic back-gate bias to reduce threshold voltage of switching MOSFET during the start-up oscillation. The Class-C differential VCO is implemented in TSMC 0.18 μm bipolar-complementary-metal-oxide-semiconductor (BiCMOS) process. The measured phase noise is −124.8 dBc/Hz at 1 MHz offset frequency from 2.48 GHz carrier while consuming 2.64 mW power from a 0.8 V supply. Tuning range of VCO is 0.72 GHz, from 2.48 to 3.3 GHz, whereas the control voltage was tuned from 0 to 2 V. The VCO occupies a chip area of 446 × 840 μm² and provides a figure of merit of −197.55 dBc/Hz.

For protecting an intellectual property (IP) core, it must be harder to reverse engineer. Structural obfuscation can play an important role in achieving this goal. A novel structural obfuscation methodology during architectural synthesis using multiple compiler-based high-level transformations (HLTs) that yield functionally equivalent designs (data flow graphs) which are camouflaged in identity is proposed. The proposed obfuscation methodology is driven through a number of HLT techniques such as redundant operation elimination, logic transformation and tree height transformation. In addition to performing obfuscation, performing area–delay tradeoff during exploring low-cost obfuscated design is also possible using these HLT techniques in the proposed methodology. Owing to multiple stages of HLT incorporated in the proposed approach during obfuscation, it yields a highly robust design which on integration with particle swarm optimisation-based exploration framework produced low-cost obfuscated IP designs. Results of the proposed approach yielded an enhancement in strength of obfuscation of 20.19% and reduction in obfuscated design cost of 59.66% compared with a similar approach.

Owing to their simple structure and considerable improvement to the transient response, phase-lead controllers are broadly used in industrial applications. They are designed using root locus or Bode diagrams to introduce extra phase shift at the crossover frequency. Traditional phase-lead design techniques are either graphical or based on trial-and-error. In design techniques using root locus diagrams, there is no direct control on the values of a given time-domain performance index. To make this criterion satisfactory, the designer has to try different locations of the controller's zero and pole. Based on minimisation of the controller's pole-zero ratio subject to pole placement requirements, a simple analytical design technique is developed in this Letter. The necessary and sufficient conditions to determine optimal controller parameters are provided. To evaluate the performance of the proposed method, it is applied to a single integrating plant with time delay.

A dynamic key-frame selection technique is proposed for enhanced odometry estimation. The odometry estimation problem has a drift problem in which errors are accumulated over time. The proposed technique finds the key-frame which reduces the odometry drift for every time step based on scan similarity comparisons. The enhancement of the proposed method is verified through the performance comparison with the frame-to-frame and the conventional frame to key-frame methods.

Active contour model driven by novel fitting term is proposed for image segmentation in this letter. The novel fitting term contains two parts: one is L ¹ fitting term, which describes the change of energy between inside and outside the curve; another is the reciprocal cross-entropy fitting term, which can compute the local energy of every point on the curve. The reciprocal cross-entropy fitting term is more robust than L ¹ fitting term when the regional grey scale fluctuates greatly, and the detail direction of the curve can be controlled better by it. Experiments on several synthetic and real images have shown that the proposed model achieves better segmentation results and efficiency than the other models; in addition, it is insensitive to different initial contours.

The development, training and evaluation of computer vision algorithms rely on the availability of a large number of images. The acquisition of these images can be time-consuming if they are recorded using real sensors. An alternative is to rely on synthetic images which can be rapidly generated. This Letter describes a novel method for the simulation of Kinect v1 depth images. The method is based on an existing empirical noise model from the literature. The authors show that their relatively simple method is able to provide depth images which have a high similarity with real depth images.

The conventional beamformers based on the projection methods are known to degrade due to the subspace swap at low signal-to-noise ratio values and the high or inaccurate dimension of subspace. A novel beamforming technique that combines the constructed covariance matrix and the estimated steering vector is proposed. The steering vector of each source signals is estimated using its cyclostationarity. Unlike the projection methods, the signal-plus-interference covariance matrix is constructed by further estimating the power of each source signals but avoiding forming the subspace. Numerical examples illustrate the notable improvement in comparison with existing beamforming techniques.

A new cyclic redundancy check (CRC) concatenation with polar codes based on partial protection is proposed, where only the crucial bits chosen from the total information bits are protected with the CRC. The index set of crucial bits is determined by the row weight of generator matrix of polar codes and the indexes with minimum row weight make up this crucial set. Using this new concatenation, the number of polar codewords with minimum Hamming weight can be efficiently decreased. Simulation results show that the performance of the proposed partial protection scheme owns a lower error floor than that of the classical blind protection.

A simple method is proposed for the derivation of bit error floor (BEF) expressions of multi-level phase modulations with noisy phase reference based on the geometric features of constellations. To validate the result, average BEF expressions of 8 + 8 and 4 + 12 amplitude and phase-shift keyings adopted in digital video broadcasting-satellite 2 extensions are derived and consultative committee for space data systems, and analyse the results.

New direction-of-arrival (DOA) and non-circular phase estimation algorithms for a mixture of circular and non-circular signals, which are applicable to arbitrary planar arrays are presented. For simplicity, the non-circular signals referred to are assumed to be strictly non-circular. First, the DOAs of circular and non-circular signals are estimated based on two polynomial formulae. Then the estimated DOAs of circular signals can be used to estimate the DOAs of non-circular signals more precisely. Furthermore, the non-circular phases can be calculated by the estimated DOAs of non-circular signals. Simulation results show that the proposed algorithms outperform the compared algorithm.

A compact dual-band five-way Wilkinson power divider with good RF performance is presented. Two-section impedance transformers are utilised to obtain the dual-band function. To facilitate assembling the isolation resistors among outputs, a second layer is introduced. The proposed power divider works at 0.915 and 2.45 GHz with small phase and magnitude imbalances.

A terahertz (THz)-wave interferometer was devised for measurement of the THz-wave phase shift. In this measurement system, the relative phase shift is deduced from the interference between the THz waves generated by two photomixers. Experimental results with this interferometer revealed that the semiconductor phase modulator could be used as a THz-wave phase shifter by utilising its chromatic dispersion.

An energy efficient 1060 nm GaAs-based oxide-confined vertical-cavity surface-emitting laser designed for high-speed modulation and single-mode emission has been developed. A record data rate of 50 Gbit/s at an energy dissipation of 100 fJ/bit is demonstrated for a device with >50 dB side-mode-suppression and 0.2 mA threshold current, making this laser a promising light source for long-reach optical interconnects.

The analytical characterisation of the magnetic field generated by a wireless charger based on the inductively coupled power transfer technique is presented. With this tool, the magnetic field could be computed in any spatial position and no dipolar approximation is necessary to be taken. The model is contrasted with measurements performed in a 3.7 kW prototype for electric vehicle. The comparison shows that there is a good agreement between the measurements and the model.

Power factor correction (PFC) converter with a twice-line-frequency ripple cancellation converter (RCC) has attracted much attention due to its ability of twice-line-frequency output voltage ripple cancellation and high power conversion efficiency. The audio susceptibility of buck-type RCC operating in discontinuous conduction mode is derived and the effect of loop gain of RCC on the cancellation of twice-line-frequency output voltage ripple is studied. Experimental results of a 50 W prototype are presented to verify the analysis results.

The Clarke and Park transforms have been standard analysis techniques in three-phase power system analysis for over a century. However, these transforms were originally designed for static electric grids and therefore yield non-ideal oscillatory outputs in dynamic smart grids. Using recent advances in widely linear adaptive filtering algorithms, the authors propose a self-stabilising three-phase transformation which is capable of generating synchronous reference frames regardless of the type of voltage imbalance or frequency deviation.

We describe a novel approach to extract Huynen–Euler parameters from a polarimetric scattering matrix. In contrast to existing methods, this technique is able to determine the full set of parameters while requiring the handling of only four special cases via case distinction. Results from a Monte-Carlo simulation show that the maximum error caused by the decomposition is close to floating point accuracy.

The information-theoretic waveform for target classification based on spectral variance has been studied, and its advantages have been shown. The waveform design algorithm has a low computational load and can be applied to real-time applications. In looking for more sophisticated methods while keeping the advantages of an information-based approach, a new design algorithm by deriving a strict lower bound of the information measure is developed. The proposed design algorithm requires a small amount of calculation and shows better classification performance in terms of percentage of correct detection. The proposed method is compared with other methods, and the improved performance is shown in numerically generated graphs.

Three GaAs Schottky diodes with different junction configurations are modelled. The main contribution of this Letter is the simplification and accuracy of the diode model, in which approximate materials are set to be the same permittivity in the 3D diode model and only four parameters are required in modelling the non-linear part of diode model. To validate the accuracy of the proposed models, one K band tripler and two W band triplers are designed and tested. The measured results agree well with the simulated results. It is envisaged that the method can be very useful for GaAs Schottky diode modelling.

Timing estimation is the very first step and the most important part in the wireless communication system. In conventional long term evolution (LTE) system, a ZC-based primary synchronisation signal (PSS) waveform is used to deal with the timing issue due to its ‘perfect’ autocorrelation property. However, in the Internet of things (IoT) scenarios, the initial frequency offset caused by the inaccurate crystal oscillator can be very large due to the low-cost requirement. ZC sequence is sensitive to the frequency offset especially the very large frequency offset. A new PSS waveform with frequency offset tolerance is proposed to improve the timing performance by using a couple of chirp signals. Simulation results demonstrate that the proposed PSS waveform could work well under very large frequency offset and is better than the narrow band (NB)-IoT PSS with differential detector.

In disaster recovery, ground network consisting of mobile terminals of rescue vehicles and civilians may be disconnected due to severe environment. In this situation, multiple unmanned aerial vehicles (UAVs) form an aerial network to enhance the connectivity of the ground network. When two nodes in ground network are disconnected, they adopt UAVs as relays to transmit data. Otherwise, they communicate with device-to-device manner. With this framework, the coverage probability of these two kinds of users is analysed. Besides, the optimal altitude of UAVs to maximise the capacity of ground network is achieved. A guideline is provided for the design of multiple UAVs support wireless network.

Hybrid visible light communication/infrared communication (VLC/IRC) systems are becoming the leading way for supplying high-speed Internet connectivity to airplane passengers. A robust IR uplink transmission scheme is presented for in-flight VLC/IRC systems which has the potential to support 10 Mbit/s transmission for each user in the cabin with uncoded bit error rate ≤10⁻⁶ under the assumption of line-of-sight communication scenario, also in the presence of misalignments between transmitter and receiver due to random movements of the passengers.

A compact wideband bandpass filter with a very wide stopband for terrestrial radio (TETRA) band applications is proposed. The proposed filter employs two types of resonators: one is tri-section half wavelength stepped impedance resonator (SIR) and the other is stub-loaded shorted ring resonator. The proposed filter has adopted SIRs and meandering technique in achieving miniaturised size. Spur-lines and tapped feeding techniques are infused in the filter to suppress harmonics so as to achieve wide stopband. Cross-coupling is introduced in the filter for acquiring good selectivity. This highly selective filter has a miniaturised size of 0.09 × 0.12λ _g (28 × 36.8 mm²) with a wide stopband up to 8.75f ₀ (f ₀ is the centre frequency of filter) with 25 dB harmonics suppression level.

The authors investigate the spectral flatness measure (SFM)-based spectrum sensing technique for cognitive radios. This scheme exploits the fact that under Gaussian noise, the noise-only observations have flattened spectrum, i.e. more white, when compared with that of the observations containing the incumbent or primary signal; hence, an increased SFM under the null hypothesis. Under the null hypothesis, the authors derive the asymptotic distribution of the test statistic, and the asymptotically optimal detection threshold, with a constraint on the probability of false-alarm. Furthermore, the authors show that this test is robust to the noise variance uncertainty (NVU) and is related to a test based on the entropy in the observed sequence. Through extensive Monte-Carlo simulations, the authors show that the test based on SFM performs better than the existing energy detector and the blind detector, under realistic signal and fading models, all in the presence of NVU. The authors also highlight the practical utility of this technique based on experimental results.

A novel wideband balanced-to-unbalanced filtering unequal power divider with wide stopband and isolation is proposed. The wideband filtering response with high frequency selectivity can be realised easily using the modified two-port coupled lines with short- and open-circuit stubs, thus generating two transmission zeros near the passband. The complete closed-form equations are presented. A fabricated prototype with 1 : 2 power dividing ratio is simulated and measured to validate the proposed design.