A new asymmetrical zeroth-order resonant antenna with improved efficiency and widened impedance bandwidth is proposed. It comprises a rectangular patch, two shorted stubs and a coplanar waveguide feed to have composite right/left-handed features and an easy fabrication process. The asymmetry helps to combine three resonance frequencies and extends the bandwidth up to 61.6%. The efficiency of the proposed antenna is >85% over the entire bandwidth of 4.41–8.34 GHz. The proposed antenna has a compact size, which can provide an omni-directional radiation pattern suitable for wireless applications. These features are verified by CST Microwave Studio and compared with the experimental results.

The application of Babinet's principle to the ultra-wideband (UWB) differential planar printed antenna is presented. This electromagnetic technique is applied to a classical monopole UWB antenna, and can also be applied to most types of printed antennas. It permits grouping on the same substrate of two antennas close to each other, operating in the same frequency band and with high isolation. In the final prototype, the antenna is co-designed with a transmitter/receiver pulse chip and optimised to operate in the Federal Communications Commission frequency band with a return loss <−6 dB. The overall dimensions of the differential antenna are 56 × 30 mm².

A compact coplanar waveguide (CPW) fed electrically small antenna suitable for 2.4 GHz WLAN applications is presented. The resonance of the structure is achieved by implanting a chip inductor in between the centre conductor of an open-ended CPW transmission line and a short metal strip. The total electrical size of the fabricated antenna is 0.08λ ₀ × 0.0479λ ₀. The reflection and radiation characteristics of the antenna and a method to enhance the gain are discussed.

The polarisation transforming properties of a compact frequency selective surface (FSS) are reported. Although known designs made with meander lines employ several cascaded surfaces with many interconnects, the design methodology presented uses a single planar surface of very closely arrayed metallic meander lines impedance loaded with rectangular metal patches. A simulated axial ratio of ≤ −3 dB was obtained over a 40% bandwidth, from 2.10 to 3.2 GHz and optimised for 2.75 GHz. Simulation of normal and oblique incidence in both planes of polarisation shows a ±30° wide range of angles of incidence for polariser performance. A unit cell of the FSS circular polariser is λ_o /12 in size at the optimised operating frequency. The FSS circular polariser was manufactured and attached to a horn antenna aperture and measurement confirmed axial ratio performance.

A simple method is proposed for achieving short-range multiple-input multiple-output (MIMO) transmission without complicated signal processing. This is achieved by a spatial division that cancels interference between signal streams by using π-phase-shifted antenna directivities in high speed parallel transmission systems at 60 GHz. The method achieves spatial division transmission without MIMO detection even for antenna elements with wide beamwidths. This makes it easy to design antenna elements and reduces signal processing costs in the transmitter and the receiver. Moreover, the method provides almost the same channel capacity as that of the complicated MIMO transmission methods. The optimal antenna array length is derived to maximise the capacity and the validity of the proposed method is confirmed through numerical and experimental analyses at 4.85 GHz.

A compact CPW-fed triple band-notched antenna for ultra-wide bandwidth (UWB) application is presented. By introducing an stepped impedance resonator-defected ground structure (SIR-DGS) and fork-shaped stubs, three sharp notches are achieved at frequencies of 3.5, 5.68 and 7.48 GHz. The proposed antenna working from 2.8 to 11.3 GHz with a voltage standing wave ratio of <2 is demonstrated, except for the selected notched bands. The main advantage of the antenna is that the frequency of the notched band can be tuned easily in a wide frequency range. A good agreement between the measurement and the simulation was achieved. The measurements confirm that the technique introduces negligible antenna radiation pattern distortion, except at the triple-notched frequencies.

An inverter-based ultra-low-power, low-noise, single-ended to differential continuous-time variable gain amplifier is presented for 2–6 MHz second harmonic cardiac imaging ultrasound probes in a 65 nm CMOS technology. The proposed variable gain amplifier (VGA) consists of three equal inverters and resistor arrays which form a feedback loop. To improve both the power and noise performances, the inverters operate in the sub-threshold region by adopting a 0.5 V supply voltage. By doubling the input transconductance of the VGA, the noise figure (NF) is enhanced further. The 6-b thermometer resistor arrays achieve a gain range of the VGA from 0 to 22 dB. The total power consumption is 55 μW, NF is 2.7 dB referred to an 8 kΩ source resistor at a centre frequency of 4 MHz. 8-b thermometer calibration codes are added into the inverters to force the static operating point of the inverters to half of the supply voltage, which could increase the immunity of the second harmonic distortion (HD2) to the process variation. The HD2 of the proposed VGA is −60.6 dB at the 330 mV peak-to-peak output swing. The active size is 154 × 102 μm.

In molecular communication, some molecules stay longer in the medium and impinge on later transmissions. The bit error rate (BER) of a diffusion-based molecular communication channel which has memory is analysed. The effect of four previous symbols on the current symbol is considered. Human insulin hormone is used as the messenger molecule and the first passage time of a molecule has been defined as the delay. Channel memory has been found to have a profound influence on the BER.

The use of a rollback buffer (RB) for hiding the rollback latency in log-based eager hardware transactional memory is proposed. The RB allows a transaction to abort without performing rollback, but still makes the transaction's old values immediately available. In effect, the rollback latency almost disappears. When running the Stanford transactional applications for multi-processing benchmark on a 16-core processor that implements the LogTM-SE, the speedup (decrease in execution time) achieved with a 2 KB RB is 15.8% on average.

It is shown that the area of the pinched hysteresis loop of the current-controlled ideal memristor represents the quantity ‘content’, which was introduced into the theory of nonlinear systems by Millar in 1951. Two parts of the content are identified which correspond to distinct parts of the area below the v–i characteristic of the memristor: one is related to the power conditions and the other to the instantaneous state of the memristor memory. It is demonstrated for a memristor driven by the sinusoidal current that the power part of the content depends on the fundamental harmonic of the voltage, whereas the memory part of the content is given only by the higher harmonics of the voltage, and only by the even harmonics in the case of entirely closed loops. The analogous conclusions also hold for the voltage-controlled memristor.

A new linearisation technique for active mixers is introduced to enhance the third-order input intercept point (IIP3). This technique is based on the cancellation of the third-order Volterra kernel in the output current of the transconductance stage by using the interaction between two nonlinear systems. In this scheme, a nonlinear circuit is employed prior to the conventional mixer to improve the total IIP3 by adjusting its second-order Volterra kernel magnitude and phase. Moreover, the added circuit increases the conversion gain of the mixer. The simulation results show that both the IIP3 and the conversion gain simultaneously improve by 13 and 11.4 dB, respectively, compared with the conventional active mixer.

A constrained kernel regression model is proposed to solve the problem of one-dimensional (1D) pose estimation. Unlike the traditional kernel regression model, a circular constraint is applied to the output of the regression function, i.e. using 2D coordinates on a unit circle as output instead of 1D pose angles from 0 to 360°. The experimental results show that with this constraint, the performance of kernel regression on the 1D pose estimation can be improved significantly, and the constrained kernel regression model can run in real-time.

A novel full-reference image quality assessment metric employing statistical local correlation as an indicator for quality quantification is presented. The local correlation is extracted in a wavelet domain and is pooled into an objective quality score. Simulation and testing results demonstrate that the proposed method not only outperforms the other methods in terms of high accuracy of image quality prediction, but also is consistent with the subjective evaluations.

A novel temporal contrast filter (TCF)-based method was developed to detect supersonic small infrared targets. The proposed hysteresis threshold-based detection followed by the TCF can enhance the accuracy of the target position, the robustness to background clutter and the velocity range of moving targets compared with the conventional temporal variance filter-based method.

In traditional compressed sensing methods, imposing sparsity alone usually does not produce the most visually pleasing reconstructed videos. Thus, by leveraging more prior information extracted from the temporal redundancy, a regularised reweighted basis pursuit denoising method with estimated support and signal value is proposed for a compressively sampled video. Moreover, an effective alternating direction method of multipliers is presented to solve the optimisation problem. Simulation results show that the proposed method compares favourably with conventional algorithms in the recovery performance.

Based on the low rank representation (LRR) and the sparse representation (SR), a composite LRR with SR graph LRRSR for semi-supervised label propagation is proposed. The LRRSR aims to capture both the global structure of the data by a low rank constraint and the local structure of the data by a sparse constraint simultaneously. A composite framework is applied to fuse the two graphs. Then, a label propagation framework is used to transmit the labels from the labelled samples to the unlabelled samples. It is applied on several face image datasets and the experimental results demonstrate its good performance for face classification with a limited number of labelled samples.

A novel passive wireless temperature sensor is presented, which aims at operating in harsh environments. For this purpose, the tag is designed quasi-chipless and it uses a harmonic radar sensing principle: it combines the backscattering on the first harmonic of the transmitted frequency with frequency-position encoding of the measured value. The harmonic radar approach brings significant advantages for the wireless readout, since it isolates the undesired radar clutter from the useful tag signal. The tag signal carries the desired temperature information encoded in the position of a resonance transmission peak. The tag is built up planar and consists of two patch antennas, a temperature-dependent bandpass filter and a diode-based harmonic generator. It works at 1.59 GHz and backscatters at 3.18 GHz. Wireless indoor measurements have proven the concept. The sensor has shown a temperature sensitivity of 1.2 MHz/K in the range of 22–109 °C.

A novel X-band differential bandpass filter based on substrate integrated waveguide is proposed. By making an aperture etched in the middle common ground of the vertically stacked cavities, the common-mode (CM) signals can be highly suppressed. Compared with conventional structures, the proposed design possesses a relatively simple and effective measure for the rejection of the CM signals. Based on simulated and measured results, good differential-mode responses can be observed along with a CM rejection of more than 45 dB in a wide frequency range.

A novel compact ultra-wideband (UWB) bandpass filter (BPF) with triple sharply notched bands and good selectivity is proposed using a parallel U-shaped defected microstrip structure (UDMS). The initial UWB BPF comprises interdigital coupled lines and an E-shaped multiple-mode resonator to achieve two transmission zeros on both sides of the passband, thus improving skirt selectivity. Then, three parallel UDMSs are introduced, which have the properties of achieving the triple band-notched characteristics and providing a high degree of adjusting freedom. To validate the design theory, a microstrip UWB BPF with three notched bands centred at 5.2, 5.8 and 8.0 GHz, respectively, is designed and fabricated. Both the simulation and the experimental results are provided which have good agreement.

A simple and successful design for an arbitrary phase shift at two operational frequencies by using a T-section with only a single dual-band stub-loaded line is presented, and the design equations are derived analytically. For verification, a proposed dual-band phase offset line is fabricated and measured at 1618.25 and 2491.75 MHz of the ‘COMPASS’. The results show that there is good agreement between the measured and the electromagnetic simulations.

The concept of a microwave transmission line suspended in air, and a methodology for electromagnetic and thermal optimisation of these structures in X-band (8–12 GHz) are reported. A reliable printed circuit process allows us to obtain compact, low loss and reproducible stripline structures. The measured stripline circuits demonstrate losses from only 4.2 to 5.2 dB/m at 9 GHz. The use of periodic ceramic contacts increases the power capability up to 50 W. The final structure is low loss and compact with high power handling in order to be used in power combination in the X-band.

An ultra-wideband power divider with out-of-band filtering and DC blocking functions is presented. The divider consists of three parallel-coupled lines, a short-circuited shunt stub and a resistor. The divider was implemented in a two-layer printed circuit board with a high dielectric constant. The measurements show a minimum insertion loss of −0.4 dB in the middle band and a rejection of −17 to −18 dBc at 1.5 and 11.5 GHz, respectively.

A via-swap 180° phase inverter on the slotted ground is presented. The phase inverter with a 90° line replaces the conventional 270° line built into the microstrip rat-race hybrid. This results in a size reduction and the widening of the bandwidth. The experiments show that the proposed hybrid has a good performance from 1.5 to 2.5 GHz (f_o = 2 GHz). The bandwidth is increased from 10 to 50% and a 43% size reduction is obtained compared with the conventional rat-race hybrid.

A few-layer-thick epitaxial graphene synthesised on both the up-side and the down-side surfaces of SiC wafers in an ultra-high vacuum annealing furnace is presented. The graphene grown on the down-side surfaces of the SiC wafers that contacted with a graphite pedestal resulted in a relatively lower sheet resistance along with less graphene layers, as compared with that of their up-side surfaces. Overall, the sheet resistance of the few-layer-thick graphene grown on the down-side was found to range from 200 to 3 Ω/□, depending on the annealing time.

Phase change memory cells composed of nanocrystalline Ge₂Sb₂Te₅ with a heater diameter of 10 nm and Ge₂Sb₂Te₅ thickness of 100 nm are studied by using two-dimensional finite element simulations with COMSOL Multiphysics. The nanocrystalline Ge₂Sb₂Te₅ is emulated by using a mesh-based model incorporating crystalline grains of random size and location embedded in the amorphous media. The material parameters are modelled with temperature dependency from 300 to 1000 K, including electrical resistivity, thermal conductivity, electric field breakdown and Seebeck coefficient. This model is shown to capture the cycle-to-cycle and device-to-device variability in phase change memory cells.

An ultra-compact (8 µm long) electro-absorption modulator is presented which is laterally integrated with a 980 nm InGaAs vertical cavity surface emitting laser incorporating a bow-tie-shaped oxide aperture. A low driving voltage operation below 400 mV_pp for a dynamic extinction ratio of 6 dB and a large signal modulation of up to 25 Gbit/s is demonstrated. The static extinction ratio of over 6 dB can be obtained by applying an extremely low voltage of −200 mV.

980 nm vertical-cavity surface-emitting lasers (VCSELs) operating error-free at 38 Gbit/s at 85°C with record 177 fJ of dissipated heat per bit are presented. The VCSELs are particularly well suited for short, very-short and ultra-short-reach interboard, chip-to-chip and intrachip optical interconnects, where temperature stability, bandwidth and VCSEL density and the minimisation of waste heat are critical. Error-free operation at 40 Gbit/s at 75°C and at 42 Gbit/s at 25°C is achieved. To date, 980 nm VCSELs are the most energy-efficient of all VCSELs at 85°C at any wavelength.

The transient off-state thermal emission leakage currents of hydrogenated amorphous silicon (a-Si) and indium gallium zinc oxide (IGZO) metal oxide thin-film transistors (TFTs) fabricated on flexible plastic substrates have been examined for the first time. The transient leakage currents resulting from the detrapping of charge in the TFT active layer were observed to decay in a 1/t power law behaviour from 2 to 100 ms after switching the TFT test structures off. The results demonstrated a uniform density of states for both types of thin-film devices and an ∼1.3× higher transient leakage current for the IGZO TFTs. It was also shown that the observed transient leakage currents result in a shot noise at the input to the data-line charge integration amplifiers that is 20 × larger than predicted by using only the conventional static DC TFT leakage current measurements.

The first experimental demonstration of a continuous-wave dual-pump fibre optical parametric amplifier operating near 1 µm is reported. A flat gain band of 23 dB spanning over 16 nm (4.5 THz) with <2 dB variation is reported. This fibre optical parametric amplifier is achieved with a dispersion stabilised microstructured optical fibre having a high figure of merit, i.e. high nonlinearity and low loss.

Reverse-bias operations for a light-emitting diode can quickly screen the weakness of the device. Electromigration around the electrode on the lateral and vertical directions can be observed by optical microscope images, scanning electron microscope images and energy dispersive spectroscopy. The failure process caused by the electromigration can be revealed by the changes of forward-bias and reverse-bias electroluminescence images and the emergence of the reverse-bias EL may be attributed to the diffusion of gold. Furthermore, the increase of the leakage current may supress the forward-bias band to band recombination current especially near the metal contact.

A planar wireless power transfer (WPT) system with printed square spiral resonators is reported. To facilitate its fabrication, both the spiral resonator and its coupling loop are printed on the top and bottom layer of a single FR4 substrate, such that the spiral resonator is driven by the magnetic field of the coupling loop. The utilised coupling loop along with the appropriate matching circuit will improve the overall efficiency. The proposed WPT system which features a measured efficiency of more than 40% is not only compact but also low cost.

An optimised design of a radio frequency energy harvesting antenna is presented. The antenna is based on a compact ferrite rod which, together with the electronics, can directly replace batteries in suitable applications. The antenna is optimised such that the energy available for the applications is maximised, while considering constraints such as the device geometry and the Q-factor. That the antenna can power a wireless sensor node is shown from the ambient medium wave transmissions.

A simple control strategy to reduce the total harmonic distortion in single-stage high-order power factor correction AC–DC circuits is presented. A variable hysteresis window is used to improve the system performance near the zero crossing of the input current by increasing the switching frequency near this region.

Synthetic aperture radar (SAR) processing is assumed to start with a demodulated baseband signal and attempt to solve ground reflectivity. An expression of two-dimensional SAR echo data in the discrete domain is presented. To meet the requirements of the processing power and reduce the storage, a compressive sensing framework is adopted to solve the ground reflectivity. By a small number of signal samplings, the imaging is constructed efficiently and has a higher resolution. Promising results were obtained by several experiments.

A simplified analytical drain current model for the long-channel pinch-off junctionless (JL) double-gate field effect transistor (DGFET) is proposed. The drain current is generally expressed in terms of the mobile charge density and it has been expressed in terms of the pinch-off voltage and the gate capacitance. The proposed model is validated against TCAD simulation results, and is able to predict the behaviour of the JL DGFET for different bias conditions and device dimensions, accurately. The transconductance and the transconductance-to-drain-current ratio are important parameters for analogue circuits, and have been calculated and exhibit good agreement with the TCAD results.

The autoconfiguration and management features of the Customer Premises Equipment (CPE) WAN Management Protocol (CWMP) are powerful and have been extensively used by service providers and telecommunication services. The capability of the auto configuration server (ACS) in the CWMP technology is limited by its management structure, the maximum connection and the network bandwidth. Typically, these concerns are solved by adding more ACSs. Nevertheless, the ACS that has stronger management power is pricy too. On the other hand, the computing power of the CPE having lower costs is increasing rapidly. To utilise the benefits of this change, a new design by combining a sub-ACS concept with a dynamic adaptive multi-level grouping algorithm mechanism is created, that is to recalculate a healthy rate of the ACS while a new CPE is joining the system. If the healthy rate of the ACS is below the healthy criteria, a new group of CPEs and its corresponding sub-ACS are built. The newly formed sub-ACS responds to manage the new group. By leveraging the new model, the management capability of a CWMP system can be raised to 278%. Consequently, the infrastructure of establishing both the cost and the operation cost is reduced, yet the efficiencies of the entire remote management system can be improved.

New analytical closed-form expressions are derived for the probability density function and the cumulative distribution function of the squared Rician shadowed variates. In particular, the application of such results in the performance analysis of wireless communication systems is investigated.

The dynamic and adaptive radio protocol (DARP) is a new protocol proposal with some interesting features such as dynamic roles and the use of virtual sub-networks. The wireless sensor network state of art is discussed and some desirable features are presented in order to adapt these networks to new scenarios. These necessities are quite important to expand the applicability of wireless sensor networks and for this reason the DARP is proposed.

A joint relay selection and power allocation scheme to maximise energy efficiency with restriction on the minimum spectral efficiency requirement for an amplify-and-forward two-hop relay network is presented. A closed-form solution of the low-complexity power allocation for such a system has been derived, which is easy to implement and has the potential to find wide application in green communications. Some numerical examples are provided to validate the analysis.

The outage probability of spatial modulation (SM) multiple input multiple output (MIMO) systems incorporated with transmit antenna selection (TAS) over Rayleigh fading channels for binary phase shift keying signals is analysed. A MIMO system with N _t transmit and N _r receive antennas in the Rayleigh fading environment is considered. S < N _t antennas are selected at the transmitter side to maximise the received signal power. A closed form expression for the outage probability of the SM MIMO systems with TAS is derived and validated using the simulation results.

Cooperative localisation can be classified into distributed and centralised algorithms. However, thorough theoretical analysis on the discrepancy and correlation of their performances has not yet been given by existing researches. Presented are different Cramér-Rao lower bounds for these two kinds of algorithms and the performance relationship is investigated through their own lower bounds. Furthermore, it is theoretically proved that after infinite iterations, distributed and centralised algorithms have the identical lower bound.