A novel low-profile single-feed circularly polarised (CP) microstrip antenna (MSA) loaded by a coplanar parasitic epsilon-negative (ENG) transmission line (TL)-based annular ring with enhanced bandwidth is proposed. By exciting the TM₁₁ modes on both the driven patch and the coupled ENG TL-based annular ring, two minimum axial-ratio (AR) points are generated. It is the combination of the two minimum AR points that result in the bandwidth enhancement. An antenna prototype operating at the WLAN 5.2 GHz band is exemplified to validate the proposed concept. The measured results show that the proposed antenna with a low profile of 0.035λ ₀ obtains an overlapped bandwidth (return loss <−10 dB and AR < 3 dB) of 4%, which is 2.3 times wider than the conventional CP MSA. Moreover, the proposed antenna exhibits higher gain and lower backward radiation as compared with conventional CP MSAs.

Two design strategies are presented to create dual-band frequency selective surfaces based on multi-arm Archimedean spirals. A single-band element is first presented as the basis for the surface. The individual arm lengths of the spirals are utilised as the primary design parameter to synthesise the passband and stopband responses. The first approach uses a lattice of four-arm spirals with alternating sizes, and the second combines these into an eight-arm uniform lattice of elements. Measured and simulated results are included for a fabricated S-band prototype of each topology to demonstrate the basic design strategies.

An exact explicit expression is derived for the Sommerfeld integral describing the surface transverse magnetic field generated by an infinite line source placed on a plane air–earth boundary. The spatial distribution of the transverse magnetic field is given in series form, as a linear combination of exponential functions. The accuracy and convergence rate of the proposed formulation are investigated through numerical simulations.

A parabolic reflector antenna with an azimuthally deformed Cassegrain subreflector is proposed to effectively generate arbitrary orbital angular momentum (OAM) modes. The Cassegrain dual-reflector antenna was fabricated for 18 GHz and measured in the near-field range. The near-field to far-field transformed radiation phase around a full azimuth shows that the proposed antenna generates fields with the l = 1 OAM mode. Simulated and measured feed reflection coefficients are below −10 dB for 15.1–21.2 GHz, even though the distance between a horn feed and a deformed subreflector is 15.6 mm, thus it is expected that the proposed structure can be used for a broadband low-profile antenna.

For the first time, in real experiments it is reported that an electronic neural network sustains the propagation of a low-frequency information for an appropriate setting of a high-frequency perturbation amplitude. The bench of the electrical network is confirmed by numerical predictions.

A low-power, high-sensitive current converter for implantable electrochemical glucose sensor tags based on radio frequency identification is presented. The current converter can work in the incremental sigma delta (ΣΔ) converter mode or the dual-slope converter mode. The two modes can be switched automatically according to the input current value for the target measuring range from 10 fA to 1 nA. The dual-mode converter offers a 10 fA-input current sensitivity, 100 dB dynamic range and consumes 4.26 μW of power from a 1.5 V supply.

The first demonstration of a pulsed electron paramagnetic resonance (EPR) detection system based on the recently announced particle EPR (pEPR) technology for superparamagnetic iron oxide nanoparticles (SPIONs) applications is presented. The SPIONs have large magnetisation, which allows one to detect the electron resonance under a low magnetic field and at room temperature. However, the broad linewidth of the superparamagnetic particles leads to a short relaxation time which requires a quick damping of the transmit signal. The presented experiments on a commercial SPION prove the feasibility to detect a pulsed electron response from broad linewidth particles using the pulsed pEPR technology.

A super class-AB bulk-driven (BD) operational transconductance amplifier (OTA) with improved slew rate (SR) is presented. The proposed OTA utilises bulk flipped voltage follower for the BD input stage and a nonlinear current mirror for the active load to enhance the SR of conventional BD counterpart. Both the traditional and the proposed OTA, have been simulated on the UMC 180 nm process, and results show that the proposed BD OTA achieves a 350% SR enhancement with only 25% increased power dissipation compared with the conventional one.

A triple-stage path prediction algorithm is proposed for real-time mission planning. In every planning horizon, each unmanned aerial vehicle (UAV) utilises the A* algorithm first to estimate the path to every task. Then, the estimated result serve as the input of the cluster method to generate the quasi-optimal task assignment. The shortest path to the assigned task is further calculated using the A* algorithm. Finally, it is smoothed to obtain the flyable reference path to guide the UAV by using the cubic B-spline curve. Numerical experiments demonstrate the effectiveness and efficiency of the proposed algorithm.

ΣΔ ADCs show promising perspectives in the field of compressive sensing (CS). A major limitation related to the first-order ΣΔ is its high oversampling ratio (OSR) that is required for a specific bit resolution. An extension to the high-order incremental ΣΔ for the implementation of a dedicated image sensor is presented. By extending the concept of the ΣΔ CS ADC, optimal working points can be reached between the OSR and the compression ratio. Simulations of a particular fourth-order ADC combined with the authors column-based Bernoulli CS scheme show a severe relaxation on the ADC master clock.

The non-local means (NLM) algorithm computes the weighted average of the difference between two local patches centred at each neighbour and the centred pixel, which is adapted for the adaptive mean filter of noise suppression. However, computation of the weighted average of the local patch difference is very time consuming. To solve this problem, a fast weighted average computation using an integral image for the NLM, called integral NLM (INLM), is proposed. The performance of the INLM algorithm is estimated and the experimental results show that the outcomes of INLM are totally the same as those by the classical NLM, while the running time is reduced significantly.

In face recognition, the self-quotient image (SQI) is a widely used preprocessing step to compensate for illumination. To improve SQI performance, a specially designed edge-weakening guided image filter (EWGIF) is proposed. An edge aware coefficient in the EWGIF makes it aware of edges and able to change the smoothing level automatically according to the image contents. Through the improved SQI, most of the face informatic areas are well preserved, while inconsequential parts are dropped. Experimental results on the extended Yale B and the Gray FERET databases show that face recognition accuracy is significantly improved with the EWGIF.

The Cramer-Rao bound (CRB) gives a lower variance bound that an estimator can achieve. Presented is a mathematical approach to estimate it for specific radar signals, namely Costas signals and their modified versions. The modified versions design consists in widening Costas frequency hops and replacing rectangular pulses with other waveforms (namely linear frequency modulated pulses, phase codes and other Costas codes). The calculation is based on approximations of the ambiguity function (AF) around the mainlobe. The CRBs are then derived based on relationships between the AF and these bounds.

A new circuit is described which applies a configurable voltage across an RTD while the current flowing through it is measured with a current mirror. The circuit also allows working with voltages above the IC supply voltage to cope with the high power RTD dissipation normally required in thermal anemometers. The circuit is periodically calibrated to cancel the errors and amplifier offset and therefore improves measurement accuracy. Experimental measurements of the circuit fabricated using 0.35 μm AMS technology show the functionality and improved power efficiency.

A high-linearity CMOS power amplifier (PA) operating at 2.4 GHz with an integrated diode lineariser is presented. A diode-connected N-type MOS transistor is used to increase the V _gs of the power transistor as input power increases for linearity improvement. The CMOS PA fabricated in 180 nm CMOS technology achieved 32 dB small signal gain, 30.2 dBm P _sat, 28.2 dBm P _{1 dB}, and a 31.1% peak power added efficiency. By adopting the proposed diode lineariser, the output 1 dB is improved from 24.9 to 28.2 dBm.

A 60 GHz down-conversion mixer for a superheterodyne receiver with an IF frequency of about 12 GHz is reported. On the basis of the double-balanced Gilbert cell, the mixer comprises two series LCR networks to enlarge the IF bandwidth (BW) and a cross-coupled pair used as positive feedback to the raise conversion gain (CG) at the load stage. By adjusting the bias of the cross-coupled pair, the mixer features a variable CG. The mixer core only consumes 2.8 mA from a 1.2 V supply when the cross-coupled pair is turned off, whereas the corresponding a 3 dB IF BW is from 6.5 to 17.5 GHz with a 3 dB CG, which is believed to be the widest IF BW ever reported for a 60 GHz down-conversion mixer with an IF frequency of about 12 GHz. The CG can be enhanced by driving the cross-coupled pair, which consumes another 2.4 mA, i.e. 5.2 mA in all for the mixer core. The CG will reach 7.5 dB and the corresponding 3 dB IF BW is from 8 to 16 GHz, where the CG and BW product against power consumption is also believed one of the best results ever reported for a 60 GHz down-conversion mixer with IF frequency of about 12 GHz.

A one-third triangular resonator with two electric walls and two virtual magnetic walls is proposed for the design of a broadband filter. The structure is realised by taking out one-third of an equilateral or isosceles triangular resonator of the substrate integrated waveguide and therefore it is called the one-third-mode substrate integrated resonator (OTMSIR). The characteristics of the OTMSIR cavity are studied. Based on this resonator, a filter composed of six OTMSIR cavities is designed, fabricated and tested. The measured results agree well with the simulated ones.

Reflectarray antennas composed of rectangular grooves with a digitised phase-shift element are designed for the millimetre-wave region. The digitised metal antennas have been realised by quantising the phase shift of each rectangular groove. For 4 bit quantised antennas, the far-field patterns are very similar to those of the non-digitised case and especially show that the measured gain of the 2 bit reflectarray is only 1 dB lower than that of a non-digitised reflectarray antenna. These results will contribute to saving manufacturing cost and time and the guidance of the phase shifter for an active metal reflectarray antenna for millimetre-wave wireless networking or imaging systems.

A compact microstrip tri-band lowpass-bandpass filter is described. Two different sets of λ/4 uniform-impedance open stubs generate the desired transmission zeros (TZs) to divide a lowpass frequency response based on a radial stub loaded high-impedance microstrip line into a tri-band lowpass-bandpass frequency response. A lowpass passband with a 3 dB cutoff frequency of 0.66 GHz, and two bandpass passbands centred at 1.57 and 2.45 GHz for global positioning system and wireless local area network applications are obtained in this design. The fabricated tri-band lowpass-bandpass filter occupies a compact circuit area of 0.065 λ_gc × 0.085 λ_gc. Seven TZs around the three passbands result in high band-to-band isolation, good passband selectivity and a wide upper stopband.

Stub-loaded split-ring resonators (SRRs) are presented for the design of a compact tri- to dual-band bandpass filter (BPF). The length of the two concentric split-rings having splits on the opposite sides, the gap between the two concentric split-rings and the length of the stub determine the positions of the three passbands. Switchability has been achieved by controlling the closing and opening of the inner ring of the SRRs using pin diodes. The responses of the BPF depend upon the condition of the switch. Moreover, lumped coupling capacitors are connected between the two SRRs to enhance the bandwidths of the passbands of the BPF. It also permit a greater coupling-gap between the SRRs which further reduces the challenges of fabrication.

A new dual-mode differential bandpass filter on a squared microstrip ring resonator with a slotline feeding scheme is presented. This filter consists of a microstrip dual-mode ring resonator, four identical slotlines equally spaced along the ring and two differential-mode microstrip line to slotline transitions. By perpendicularly installing the slotline between the differential feeding line and the ring resonator, the coupling strength between them can be dramatically increased as desired. More importantly, only the differential modes can propagate along the slotline, so the proposed filter can naturally block the common modes (CMs) from propagation and achieve good CM suppression. To verify the design concept, a filter has been designed and fabricated. Good agreement between simulated and measured results is attained.

A compact varactor-tuned microstrip lowpass filter (LPF) with a large cutoff frequency tuning range and an ultra-wide stopband is presented. By employing a pair of parallel coupled-lines and a low-impedance stub loaded with a varactor diode, the introduced filter can successfully realise the tunable cutoff frequency function. Besides, both radial patch resonators and stepped-impedance resonators are introduced to achieve an ultra-wide stopband in all the states in this design. A varactor-tuned microstrip LPF is designed, fabricated and measured. The measured results show that the proposed LPF provides not only a large cutoff frequency tuning range of 91.3% from 0.85 to 2.28 GHz, but also an ultra-wide stopband which extends to 20 GHz.

100 Gbit/s operation of a 1.3 μm, 4-channel directly modulated laser (DML) array monolithically integrated multimode interference (MMI) coupler is demonstrated for the first time. An optical modulation amplitude exceeding −0.7 dBm and clear eye openings are achieved at 55°C under simultaneous operation.

Temperature-stable oxide-confined 980 nm vertical-cavity surface-emitting lasers for optical interconnects are presented. For the first time 40 Gbit/s 4-PAM data transmission is experimentally demonstrated at temperatures of 100 and 120°C.

On the basis of the tuned four-wave mixing generated in a tellurite microstructured optical fibre, flexible and efficient wavelength conversion can be realised. The signal at 1494 nm can be converted to any wavelength in the band from 1598 to 1641.5 nm by tuning the pump wavelength. A 27.6 dB maximum idler conversion efficiency has been achieved.

The design and fabrication of metamorphic periodic heterostructures containing InAsSb layers with controllable modulated Sb composition and well-regulated band alignments are reported. The bandgap energy of ordered alloy can be much smaller than that in bulk InAsSb with any Sb content. The modulation period is determined by the thicknesses of the strain compensated InAsSb _x /InAsSb _y pairs grown on a virtual GaIn _z Sb substrate with a given lattice constant. The experiment shows that the deviation of the modulation period from ∼2.3 to ∼5.5 nm leads to the shift of the maximum of 20 K photoluminescence from 12.9 to 19.6 µm. It is concluded that the growth of strain compensated InAsSb _x /InAsSb _y on virtual substrates allows developing narrow bandgap III–V materials with various bandgap energies.

The noise power estimation for pulsed interference detection in global navigation satellite system (GNSS) receivers is considered. Previous effective consecutive mean excision (CME) related algorithms in this area work in an inter-block iterative fashion and the convergent process is usually quite time-consuming and needs a large storage space. A block-flow manner based on the backward CME (BCME) algorithm to overcome these problems is proposed. Numerical results show that the method decreases the computation and storage requirements of BCME, and possesses closer convergence properties.

To suppress ground clutter and barrage noise jamming for airborne radar effectively and efficiently, a cascaded method using the sparse representation technique is proposed. Based on the sparsity of jamming in the spatial domain, the jamming subspace is first estimated using the orthogonal matching pursuit algorithm with only one snapshot. Then, the jamming is rejected by projecting the received signal on the orthogonal complement subspace of the jamming. Lastly, reduced-dimension space-time adaptive processing algorithms are utilised directly to suppress clutter. Simulation results demonstrate that the novel method could achieve great performance as well as a significant saving of computation and training samples.

A novel linear-time approximation for the marginal association probabilities in the JPDAF is proposed. The key idea is to combine two existing approximations that are based on (computationally cheap) many-2-1 and 1-2-many association models. Simulations demonstrate the benefits of the novel approximation, called many-2-many approximation, in comparison with existing linear-time approaches.

Radar research using transmitters of opportunity (commensal radar (CR)) has experienced a resurgence in popularity. Most published results to date used a multiple transmitter, single receiver site configuration with combinations of bearing, range and Doppler for target localisation. Doppler-only tracking has received the least attention. This reported work demonstrates with field measurements, using a single frequency modulation broadcast radio transmitter and multiple, spatially distributed, low-cost CR receivers, that Doppler-only tracking is achievable with target position accuracies of 150 m, despite a course range resolution.

An unsupervised classification algorithm utilising both polarimetric scattering mechanisms (PSMs) of hybrid-polarity data and the Wishart classifier is proposed. The initial scattering categories of the proposed algorithm are derived from the roll-invariant m–χ classification algorithm. Pixels with no clearly defined dominant PSM are excluded, and the resulting categories are expanded into a specified number of classes. These derived classes are taken as training samples of the Wishart classifier. The effectiveness of the proposed algorithm is validated with the dataset over San Francisco.

A three-terminal breakdown voltage over 3 kV is reported on AlGaN/GaN high-electron-mobility transistors (HEMTs) grown on silicon (Si) (111) substrate with a buffer thickness of 5.5 µm. The breakdown voltage of the 1.5 × 50 μm² devices has been improved by more than 200% by applying a local substrate removal all around the drain of the transistors in order to suppress the parasitic substrate conduction phenomenon. This establishes a new record breakdown voltage for GaN-on-silicon lateral power devices while maintaining a low specific on-resistance of about 10 mΩ · cm².

In this reported work, amorphous InGaZnO thin film transistors (TFTs) with symmetric and asymmetric structures are fabricated. A constant positive gate bias for a time of 1–5 × 10⁴ s was then applied to the TFTs to investigate their performance variation. The mobilities of the TFTs were not significantly degraded by a long stress time of 5 × 10⁴ s. Also, decreases of sub-threshold swing and off-state current were observed for a stress time of 10⁴ s. Although significant threshold voltage variation was found for the a-IGZO TFT with a symmetric design during the stress test, it could be reduced by 56% for the TFT fabricated by the proposed asymmetric design.

A novel asymmetric isolated gates dopant segregated Schottky barrier MOSFET (AIG DS-SBMOS) for the suppression of ambipolar leakage current (I _AMB) using gate engineering is reported. The AIG DS-SBMOS consists of a dopant segregated source/drain with two asymmetric isolated gates (control gate, fixed gate) having different metal work functions. The control gate is used to control current conduction by modifying the Schottky barrier height and width at the source/channel junction; while, the fixed gate modulates the drain side Schottky barrier to the suppress I _AMB caused by hole injection in the off-state condition. In contrast to the conventional SBMOS and DS-SBMOS that suffer from ambipolarity, the proposed device is free from severe I _AMB. Moreover, the proposed AIG DS-SBMOS has a high I _ON/I _OFF ratio of 10⁷ with nearly the same I _ON and subthreshold swing as that of the DS-SBMOS. The proposed device structure is especially beneficial for nanowire-based transistors.

With the rapid advancements in electronics and in communication infrastructures, the deployment of low-complexity devices for power system monitoring has become an active area of research. For the large-scale deployment of monitoring devices, robust and economical remote data acquisition systems for power system monitoring are required. Keeping this in mind, the scheme of a remote leakage current acquisition system for high-voltage insulators that may be employed for remote monitoring purposes is described. A low-complexity method is adopted for the extraction of harmonic information from the leakage current signal. Unlike existing remote insulator-monitoring techniques that employ a dedicated fibre-optic network and satellite communication, the mobile communication voice channel is used to communicate the data to a central location for analysis. This provides an economical solution for the remote monitoring of insulators that may be used for large-scale deployment.

The effective capacity of communication systems over generalised κ–μ shadowed fading channels is investigated. A novel and analytical expression for the exact effective capacity is derived in terms of the extended generalised bivariate Meijer's-G function. To intuitively reveal the impact of the system and channel parameters on the effective capacity, closed-form expressions are derived for the effective capacity in the asymptotically high signal-to-noise ratio regime. The results demonstrate that the effective capacity is a monotonically increasing function of channel fading parameters κ and μ as well as the shadowing parameter m, while it decays to zero when the delay constraint θ → ∞.

The numerical solution of the dense linear complex valued system of equations generated by the method of moments (MoMs) generally proceeds by factoring the impedance matrix into LU decomposition. Depending on available hardware resources, the LU algorithm can be executed either on sequential or parallel computers. A straightforward parallel implementation of LU factorisation does not yield a well distributed workload, and therefore it is the computationally most expensive step of the MoMs process, especially when adapting to the GPU technology. Some performance improvement of LU decomposition can be achieved by applying a hybrid approach to the parallel processing model. In this reported work, the problem of accelerating an out-of-core-like LU solver on a heterogeneous low-cost single GPU/CPU computing platform is addressed. For this, a variable panel-width tuning scheme combined with a hybrid panel-based LU decomposition method is employed, which is something of a novelty in the development of dense linear algebra software. To demonstrate the efficiency of the proposed approach some numerical results are provided.

Novel authentication procedures for interworking wireless local area networks and evolved packet system are proposed for the 3GPP TS 33.402 (released 12 September 2014) standard. The novel solutions mainly exploit public-key fully homomorphic encryption schemes and one-way functions. This not only mitigates potential security threats, but significantly reduces the authentication load of the standard, especially in the case of 3GPP-IP access.

A low-complexity method for symbol detection in multiple-input multiple-output communication systems with spatial modulation is proposed. The proposed method finds a solution by performing beam search, where the beam search algorithm is modified to apply an additional branch-and-bound technique in order to reduce the search space. Simulation results show that the proposed method achieves a near-optimal bit error rate, and its complexity is much lower than that of previous methods.

A low-complexity transmit antenna selection (TAS) scheme is proposed for spatial modulation systems to provide a better trade-off between computational complexity and system performance. The proposed scheme can achieve a performance similar to the optimal TAS scheme in the low SNR region, but with much lower complexity compared with existing schemes in the literature.

A new concept of burst generator architecture dedicated to impulse-radio ultra-wideband wireless communication systems is presented. Bursts are generated by using four elementary baseband pulse generators operating in parallel. Thus, the resulting waveform is the amplitude and phase addition of time-overlapped pulses. The circuit has been implemented in 65 nm CMOS technology under a 1.2 V supply voltage. Simulation results are presented exhibiting a power consumption of 7.24 mW.

Considering the effects of primary user traffic, timing misalignment and multiple arrival and departure issues in a sensing period should be addressed for spectrum sensing in future overlay cognitive radio (CR) networks, such as femtocell networks. To achieve this goal, division of the whole sensing duration into multiple windows is proposed. For each window, a two-step detection scheme, i.e. coarse and fine detection, is proposed to improve the performance of the conventional energy detector (ED) in the overlay CR scenario.