Dedicated short-range communications (DSRCs) is a novel short- to medium-range wireless protocol designed for automotive use. The DSRC signals are circularly polarised and allocated in the 5.8 GHz band. Described is the development of a monolithic and compact patch antenna with left-hand circular polarisation intended for the on-board unit equipment of a DSRC system. The 40×60×2.455 mm fabricated prototype exhibits a circularly-polarised gain of about 5.52 dBc with a cross-polar discrimination of about 20 dB.

A new compact dual wideband CPW-fed inverse triangular-shape antenna is presented. The lower and upper bands are widened from 68.76 and 18.33% to 96.23 and 28.7%, respectively, using a triangular patch-trapezoidal ground combination. The antenna performance is further enhanced by backing it with a dual-band uni-planar EBG reflector. The final design has been prototyped and measured. It sufficiently covers the frequency bands 1.5–5.63 GHz and 9.52–13.06 GHz with enhanced gain up to 5.2 dB. The radiation efficiency is improved to approximately 72%.

A wideband quasi-Yagi antenna utilising a microstrip-to-coplanar stripline balun is presented. The antenna is designed to be compact, directional and wideband so that it can be used to form a circular array for a microwave imaging system. The feeding structure of the antenna includes an ultra-wideband balun that is formed using a pair of microstrip to slotline transitions. The simulated and measured results of the proposed antenna indicate more than 70% fractional bandwidth with more than 10 dB return loss, more than 18 dB front-to-back ratio and around 4 dBi gain.

Conventional linear regression analysis methods used to measure muscle fatigue based on mean power frequency (MPF) can yield results that are dependent on the epoch of each MPF value. An improved algorithm containing two window segmenting methods, overlapped and non-overlapped schemes, is provided to investigate the effect of window length on linear regression analysis during sustained isometric constant force muscle contraction. The minimal length of each segment can be determined by making the slope sign changes to be zero. In conclusion, the minimal length is independent of the epoch of each MPF value.

An efficient VLSI architecture of a lossless ECG encoding circuit is proposed for wireless healthcare monitoring applications. To reduce the transmission and storage data, a novel lossless compression algorithm is proposed for ECG signal compression. It consists of a novel adaptive rending predictor and a novel two-stage entropy encoder based on two Huffman coding tables. The proposed lossless ECG encoder design was implemented using only simple arithmetic units. To improve the performance, the proposed ECG encoder was designed by pipeline technology and implemented the two-stage entropy encoder by the architecture of a look-up table. The VLSI architecture of this work contains 3.55 K gate counts and its core area is 45987 µm² synthesised by a 0.18 µm CMOS process. It can operate at 100 MHz processing rate with only 36.4 µW. The data compression rate reaches an average value 2.43 for the MIT-BIH Arrhythmia Database. Compared with the previous low-complexity and high performance techniques, this work achieves lower hardware cost, lower power consumption, and a better compression rate than other lossless ECG encoder designs.

An all-digital PLL (ADPLL) which employs a ΔΣ delay-locked loop (DLL) to achieve a PVT-insensitive time resolution of the time-to-digital converter (TDC) as well as noise-shaped dithering is implemented in 65 nm CMOS. Experimental results show that the proposed method can achieve spur reduction with slight degradation of in-band phase noise. The 1.8 GHz ADPLL consumes 14.3 mW, while the TDC with the ΔΣ DLL consumes 2.1 mW.

Presented is a novel current-reuse single Miller feedforward compensation (CRSMFC) scheme for multi-stage operational amplifiers (opamps). The proposed scheme introduces an implicit feedforward path into an opamp by reusing the current of the input stage, which creates a left-half-plane zero to cancel out the first non-dominant pole in the transfer function, thus improving phase margin and gain bandwidth product (GBW) without extra power consumption. A three-stage opamp prototype with proposed CRSMFC is realised in standard 0.18 µm CMOS process, and measured results show that it achieves a GBW of 4.4 MHz and a phase margin of 73° while driving a large capacitive load of 250 pF with only 34 µW of power consumption, implying a superior FoM over the state-of-the-art.

A time-to-digital converter (TDC) for a low-power, wide-range all digital phase-locked loop (ADPLL) is presented. The proposed TDC uses an enabling signal with variable duration to achieve low power and wide range. For verification purpose, the ADPLL is fabricated in a 0.11 µm CMOS technology. The ADPLL dissipates 6.02mW at an output frequency of 1.68GHz and its output frequency is measured as 0.24–1.68 GHz from a 1.2 V supply.

The susceptibility to electromagnetic interference (EMI) of an amplifier based on dynamic threshold voltage MOS transistors (DTMOSs) has been analysed and improved thanks to an easy modification of the input differential stage. The final opamp has been designed in a 0.18μm standard CMOS process.

Proposed is a parallel array histogram architecture (PAHA) suitable for embedded implementations. The PAHA uses a register array instead of a memory block to store the histogram bins. In each step, M inputs can be processed in parallel to update the histogram bins without any additional latency. Also described is a second version of the PAHA with a flexible number of inputs, potentially avoiding the need for multiple PAHAs in a single application. Implementation results show that the architecture can achieve a super-linear speed-up of 43.75× for a 16-way PAHA when compared to a software implementation in a general-purpose processor.

A novel implicit dynamic system is exploited and investigated for online simultaneous linear equations solving. Compared to the classic gradient-based dynamic system and the recently-proposed Zhang dynamic system, the proposed implicit dynamic system can achieve superior global exponential convergence performance. Theoretical analysis and computer simulation results substantiate the effectiveness and efficiency of the proposed model for the real-time solution of simultaneous linear equations.

Unequal error protection extensions of low density parity check accumulate (UEP-LDPCA) codes are discussed and several potential applications are proposed. Two UEP-LDPCA schemes are analysed and code parameters are optimised using a density evolution algorithm. Simulation results show that the extensions have the significant unequal error protection property. This activates new directions for existing applications.

Proposed is a composite vector selection method based on discriminant analysis for face recognition. By measuring the amount of discriminant information in each composite vector, informative composite vectors can be selected. Then, composite features for face recognition are extracted with only the selected composite vectors. The experimental results show that the proposed method results in improved performance.

Frame rate upconversion (FRUC) is an important post-processing technique to enhance the visual quality of low frame rate video. A major, recent advance in this area is FRUC based on trilateral filtering which novelty mainly derives from the combination of an edge-based motion estimation block matching criterion with the trilateral filter. However, there is still room for improvement, notably towards reducing the size of the uncovered regions in the initial estimated frame, this means the estimated frame before trilateral filtering. In this context, proposed is an improved motion estimation block matching criterion where a combined luminance and edge error metric is weighted according to the motion vector components, notably to regularise the motion field. Experimental results confirm that significant improvements are achieved for the final interpolated frames, reaching PSNR gains up to 2.73 dB, on average, regarding recent alternative solutions, for video content with varied motion characteristics.

A method to construct a hierarchical template tree for pedestrian contour detection by iteratively applying a k-medoid clustering algorithm from the lowest level to the highest level was recently proposed and received much attention. Analysed here is the limitation of the method using lower level medoids as points of the next higher level, and proposed is a method of selecting a medoid from the leaf nodes of sub-trees corresponding to the lower level medoids.

It is shown that object detection can be addressed in the authors' unified framework, where the observed video matrix is decomposed into the low-rank matrix and the sparse matrix. The recovering problem can be solved by the proposed variant of the Douglas-Rachford splitting method, which accomplishes recovery by exploiting the separable structure property of the model. The effectiveness of the proposed object detection scheme is illustrated on two data: simulated data and real sequences applications. The numerical experiments verify that the proposed algorithm has attractive robustness and high accuracy for illumination variation and dynamic texture.

Investigated is the resource allocation problem in OFDMA networks by introducing the sleep mode. In contrast to the existing literature, which mainly focuses on improving network throughput by increasing energy consumption energy efficient resource allocation with sleep mode is considered under which some components of the transmitter can be shut down to save energy. A mix integer programming problem is proposed to minimise energy consumption while satisfying the average data rate requirement. The existence of a unique globally optimal resource allocation solution is demonstrated and an iterative algorithm is developed to obtain it. The simulation result shows that the proposed algorithm can effectively save total energy consumption especially when the network is under low load condition.

Proposed is a novel clustering technique based on kernel methods. The geometric properties of normalised kernel spaces are exploited to automatically detect the correct number of clusters, thus avoiding the requirement of an initial estimate of this parameter, as required instead in many popular algorithms.

A novel and simple algorithm capable of extracting and cancelling harmonic disturbance from signals subject to severe power line interference is presented. The key element consists of the adaptation of a filter to a deterministic harmonic reference closely correlated to the interference. The resulting signal is optimal in the mean squared error sense. Simulation results show that this carefully crafted adaptive scheme can improve by up to 30 dB the signal to interference ratio. Minimal remaining interference and fast convergence are obtained by optimisation of the algorithm's parameters.

A simple effective ensemble method is proposed in which individual classifiers are combined with the weight coefficients obtained by decomposition for the 0-1 matrix. The 0-1 matrix is introduced to denote individual classifiers of the ensemble and is constructed based on the prediction labels of individuals and the true labels. The weight coefficients of individuals are obtained by singular value decomposition for the 0-1 matrix based on linear mapping. In particular, the square of elements of the right singular vector corresponding to the maximum singular value are as the weight coefficients of individuals, and it is proved theoretically that it minimises the upper bound of ensemble error. Experimental results illustrate that the proposed method improves the performance of classification compared against standard ensemble strategies.

The active thermal nondestructive testing and evaluation method is a rapidly growing testing procedure for a quick and remote inspection procedure for fibre-reinforced plastics. Conventional modulated lock-in thermography significantly contributed to this field by allowing usage of low peak power controlled stimulations followed by phase based detail extraction procedures. But demand of repetitive experimentation required for depth scanning of the test object limits its applicability for realistic critical applications and demands multi-frequency low power stimulations for better resolution and sensitivity for sub-surface defect detection. Frequency modulated thermal wave imaging and coded excitation thermal wave imaging methods permitting multi-frequency stimulations cater for these needs and facilitate depth scanning of the test object in a single experimentation cycle. Recently introduced three-dimensional pulse compression is an alternative to phase based analysis for these stimulations by providing enhanced defect detection even in noisy environmental and experimental conditions. Defect detection capability and sizing by these non-stationary thermal wave imaging methods are highlighted using the pulse compression approach. The present experimental study has been carried out on a carbon fibre reinforced plastic specimen with flat bottom holes.

During running and walking the human centre of mass experiences a symmetric acceleration along the mediolateral direction. This reported work shows how to exploit this knowledge to correct misalignments of the axes of a trunk-mounted accelerometer with respect to the body axes. After vertical alignment, based on the gravitational component of the signal, the technique computes the virtual rotation angle of the axes lying in the horizontal plane. The chosen angle minimises the autocorrelation of the signal along the mediolateral direction.

The bandstop behaviour of a stepped impedance microstrip ring is studied. The ring is composed of four sections of microstrip line, which are equal in electrical length, but different in impedance. When the impedances of the lines meet the condition of rotational symmetry, a bandstop response could be obtained. At the central frequency of the stopband, the electrical length of the lines is 90°. By tuning the lines' impedances, the bandwidth could be finely adjusted, typically from 0.5 to 40%. Prototypes are designed, fabricated and measured. Good agreements are observed between the measured and the simulated data. The measured data shows that the 3 dB bandwidth is 24%, with a rejection level of −32 dB at the central frequency 1.99 GHz. In addition, the measured matching within the passband is good, as predicted.

A compact dual-wideband bandpass filter using an asymmetrical resonator is presented. The proposed resonator consists of three uniform microstrip lines of different lengths, and provides not only three transmission poles but also two transmission zeros for each passband. Thus, the bandpass filter has wide bandwidths and sharp cutoff responses. Theory as well as experimential results are given to validate this design concept.

A compact ultra-wideband (UWB) bandpass filter (BPF) with triple-notched bands and wide stopband is presented. The basic UWB BPF comprises four folded shunt quarter-wavelength short-circuited stubs separated by connecting lines with lengths of λ_g0/4 and λ_g0/2, respectively. The proposed coupled triple-mode stub loaded resonator is found to have the advantage of introducing triple-notched bands simultaneously. Five stepped-impedance shunt stubs are chosen to suppress the harmonic passband. The predicted results are compared with measured data and good agreement is reported.

A novel microstrip bandpass filter with compact size is proposed. With two short-circuited stubs, one open-circuited coupled line using symmetrical feeding, and two spur lines, a wide passband can be obtained. Detailed design and synthesis procedures are provided for filter development. Moreover, the electromagnetic simulator IE3D is used; the prototype of the bandpass filter is consequently fabricated and measured. Agreement between measured results and the theoretical prediction validates the proposed structure.

A class B totem pole power amplifier is presented. The amplifier has a small signal bandwidth of 23MHz and is capable of driving a 1MHz sinsuiodal signal into a 7.5Ω load resistance at an output power of 3.7W. When amplifying the AC envelope component of a 20MHz bandwidth 3GPP Long Term Evolution (LTE) signal, 2W is produced into the 7.5Ω load. By optimising the supply rails based on the characteristics of the envelope signal, an efficiency of 31% is achieved. When combined with a switched mode power supply, it forms an envelope modulator that can achieve 61.6% efficiency at 9.3W output power with an output peak-to-average power ratio of 8.5dB.

Mirco-motion analysis of ballistic targets is a useful tool for discriminating different targets in the aerospace. A micro-motion model combining inertial parameters with a target radar coss-section is presented. The relation between target micro-motion states and inertial parameters is deduced. By using the time-frequency transform method, the micro Doppler spectrogram is obtained. Simulation results show that the proposed method can be used for evaluation of ballistic targets recognition and discrimination.

By employing open circuited stubs with a length ratio of 1:2:3 separated by three unit elements, it is shown that a pseudo-elliptic frequency response can be obtained with a compact structure. It has an electrical performance comparable to that of a Cauer filter, but is realisable in etched form for very wide bandwidths and very high rates of cutoff.

The fabrication of highly conducting carbon nanotubes (CNTs)/polydimethylsiloxane (PDMS) composite is presented with the aim of electric heating unit applications. High shear processing techniques were used to separate highly entangled multi-walled carbon nanotubes with high aspect ratio (∼5000). No notable agglomerates or phase separation between the CNTs and PDMS matrix are observed, and the CNTs are well dispersed. It is observed that the approximate percolation threshold is found to be below 0.1 vol.%, which is a lower value than previously reported. The resulting conductivity of the CNT/PDMS composite is about 2.23 S/cm at 5.7 vol.% CNT loading. The fabricated CNT/PDMS composites can be quickly heated from room temperature to 130 °C within 90 s by applying a DC voltage of 5 V. The proposed CNT-polymer composite system with obtained results could be used as a basis for light-weight and high effective heating unit applications.

Presented is a study on a single-drive dual-parallel Mach-Zehnder modulator implementation as a single sideband suppressed carrier generator. High values of both extinction ratio and sidemode suppression ratio were obtained at different modulation frequencies over the C-band. In addition, a stabilisation loop had been developed to preserve the single sideband generation over time.

A report is presented on four-wave-mixing-based all-optical time-domain demultiplexing of a 170.8 Gbit/s signal down to 42.7 Gbit/s in a chalcogenide GeAsSe microstructured fibre. The high nonlinearity of the fibre allows error-free and low power penalty operation with only 56 mW of total input average power.

Proposed is a multi-core fibre amplification method that can reduce crosstalk, in which the optical amplified signals in two adjacent cores propagate in opposite directions. By employing the proposed method, confirmed is a crosstalk improvement of more than 4 dB using the outer six cores of a seven-core erbium-doped fibre.

A high phase sensitivity of 25 dB at the nonlinear phase shift of 0.8 rad in a sideband-assisted dual-pump phase-sensitive amplifier is achieved. Here the sidebands play an important role to evoke larger phase-sensitive attenuations for achieving higher phase sensitivity at such a lower nonlinear phase shift. This feature is utilised to improve the efficiency in the phase regeneration of 21.5 Gbit/s DPSK signals. The phase error of 22.6° was squeezed to 2.5° by operating such a sideband-assisted dual-pump phase-sensitive amplifier at 0.8 rad nonlinear phase shift. The operations at low nonlinear phase shift enable utilisation of the state-of-the-art silica-based highly nonlinear fibre with low noise and low dispersion slope by avoiding stimulated Brillouin scattering induced by higher pump power.

Delay and packet delay variation are key parameters for real-time applications and packet-layer synchronisation. Demonstrated is an Ethernet integrated hybrid network enabling adding packet-traffic to a 10 Gbit/s Ethernet circuit without impacting packet-traffic delay, delay variation or packet loss. A scheduling algorithm increasing the throughput efficiency of integrated hybrid networks is demonstrated.

A receiver that incorporates an all-optical clock recovery approach for synchronisation suitable for use in incoherent OCDMA transmission is demonstrated. The developed solution was implemented and tested in a multiuser environment using the 2D-WHTS coding scheme on incoherent OCDMA transmission with 2.5 Gbit/s data rate. The receiver with built-in all-optical clock recovery was tested by taking the BER for the received data when synchronised with the all-optically recovered clock from the incoming OCDMA traffic and when an RF synthesiser was used to the generate the clock. Improvement of ∼7.5 dBm was observed with the all-optical clock recovery approach. The related eye diagrams and the wavelength spectrum were also recorded.

The generation of second- and third-harmonic light with single- and dual-pitch periodically poled lithium niobate ridge waveguides for achieving a carrier-envelope-offset-locked frequency comb with low pulse energy is investigated. The wavelengths of both the second- and third-harmonic light converted by short- and long-wavelength components of supercontinuum spectra are precisely adjusted by selecting appropriate pitch size and controlling the temperature of the devices. The output powers of the second- and third-harmonic light are 9.1 and 4.8 µW, respectively, when using a 2/3-octave-wide supercontinuum light generated by a 1.2 nJ laser pulse at a repetition rate of 250 MHz.

Presented is the power-scaling characteristics of microchip vertical external-cavity surface-emitting lasers which include grating apertures for mode selection. Multi-Watt multi-lateral-mode emission at a wavelength of 1060 nm is obtained with sub-linear increase in maximum output power with aperture-size. It is also shown that the polarisation is only set by the grating lines for high-order modes which interact sufficiently with the grating.

It is found that Ca additive effectively suppresses the photo-darkening effect in Yb-doped silica fibre even at 6.0 wt% of high Yb₂O₃ concentration. Ca ion works as a stabiliser to maintain the Yb³⁺ valence state. It is hypothesised that this phenomenon may reflect the local charge balance effect in silica glass. This Yb-Ca co-doped silica fibre can be useful for high-power fibre lasers.

The feasibility of energy harvesting using the human body, which is used instead of a general antenna as an antenna in a low-frequency band to harvest energy included in an ambient electromagnetic wave, is investigated. Using the presented method, it is possible to harvest much more power in comparison with that possible with the previous method, in which a general antenna is used to harvest power in the high-frequency band. Also, this method can be easily applied to mobile devices because no antenna is required.

A method for power sensing in low-power energy harvesting systems is presented. The technique utilises a small sense capacitor between the converter output and the main storage capacitor. The storage capacitor is disconnected from the converter for a short period of time and the change of sense capacitor voltage is measured. This change is proportional to the average output current of the converter for the period of the measurement, thus the relative output power can be determined using only voltage measurements. The method has been experimentally validated in a low-power energy harvesting system that employs a microcontroller-based control implementation.

Presented is the use of a reflectarray-based structure in order to enhance the radar cross-section (RCS) of a flattened dihedral corner reflector. Phase-shifting cells are printed on the panels of the reflector to steer the reflected beam so that a retrodirective radiation is obtained. An experimental validation is given at 9 GHz.

Arsenic (As) diffusion in germanium (Ge) has been studied by implanting As in a Ge substrate with high boron (B) background doping. The high hole density induced by the B doping suppresses negatively charged vacancies (V) in Ge. Under this condition, we have investigated the dependence of As diffusion on the dopant-vacancy pairs As⁺V⁰ by secondary ion mass spectroscopy. After rapid thermal annealing at 600−750°C, the chemical profiles of As do not change in the highly B doped Ge. Experimental results suggest that the As⁺V⁰ pairs are not the diffusion vehicles of As in Ge. Activation energy of the As⁺V⁰ pairs has to be larger than 3.25 eV.

The traps in an AlGaN/GaN metal-insulator-silicon-high-electron-mobility-transistor on silicon are investigated by analysing the random-telegraph-signal (RTS) noise in the drain current. Two different types of RTS noise due to the traps at the gate-oxide and AlGaN have been observed in this reported work. Through the analysis of the two RTS noises, the presences of the two trap types are verified and the trap locations have also been identified.

Efficient wave equation perfectly matched layer (WE-PML) formulations, based on the auxiliary differential equation (ADE) scheme, are presented for truncating scalar dispersive finite difference time domain (FDTD) grids. The formulations maintain the same accuracy as previous WE-PML approaches but with fewer additional auxiliary variables. Moreover, the formulations can be coupled easily with the classical WE-FDTD scheme.

Proposed is a new set-membership approach based on adaptive error bounds to solve multi-hop localisation problems in the cases where the prior information is insufficient and only a very few distance measurements are available, which are commonly encountered in large-scale underwater wireless sensor networks. Unlike most of the existing schemes, the localisation problem is solved in a set-membership framework which is based on the consideration that the distance measurement uncertainty can be cast into an unknown but bounded context. For the first time, an adaptive bounds specification is introduced into multi-hop localisation for eliminating the risk of overbounding or underbounding. To reduce the computational burden, boxes are utilised to approximate the irregular intersection of all feasible solutions. Then the feasible sets and coordinates of ordinary nodes are obtained by using the interval analysis method. Simulation results confirm the effectiveness of the algorithm.