Analytical expressions for the transfer function and reflection coefficients of the surface acoustic wave (SAW) two-port asymmetrical resonator were obtained using the scattering matrix method. This resonator consists of two different interdigital transducers (IDTs) and reflectors located asymmetrically with respect to the centre of the resonator structure. Equations for the normalised amplitudes for forward and backward travelling SAWs in areas between the IDTs and reflectors have been written and solved. Numerical calculations of the above functions are possible by using the analytical expressions for the admittances and the scattering coefficients of the IDTs and the reflection coefficients of the reflectors. The analytical expressions obtained for the two-port asymmetrical resonator can also be used for the analysis and design of asymmetrical one-port resonators and delay lines.

A reconfigurable wideband metamaterial absorber based on lumped components is proposed. The wideband can be tuned with a −10 dB bandwidth of about 1.5 GHz by selecting certain values of the resistors and the inductors. The simulated results demonstrate that the designed absorber has good performance regarding polarisation insensitiveness and wide angles of incident waves for both TE and TM polarisations. The waveguide measurement method is used to verify the simulated result of the wideband metamaterial absorber.

A novel design of miniaturised ultra-wideband (UWB) antenna for oil pipeline imaging is presented. The proposed prototype uses a simple eye-shaped slot radiator and two ground planes at the same side of the dielectric substrate. The proposed antenna design exhibits an enhanced bandwidth of 20.3 GHz from 3.2 to 23.5 GHz (for return loss <10 dB) which provides a wide usable fractional bandwidth of more than 152% with ultra-compact area of 14 mm × 12 mm. The simulated and measured results confirm the benefits of the proposed design and reduction in the antenna size with enhanced UWB performance.

A frequency reconfigurable antenna array is introduced. The array consists of planar radiating elements that may be interconnected through computer-controlled switches to form larger elements radiating at frequencies lower than the resonant frequency of the individual elements. Simulations using a 2 × 2 array of patch antennas connected using either hard shorts or RF shorts are used to establish the viability of the concept. The simulated arrays were fabricated, the measured properties of the arrays being in good agreement with predicted results. Both of the shorting methods resulted in reduction of the resonance frequency to half that of an individual element, while the radiation pattern of the individual element was maintained.

Liquid metal actuated by pressure-driven air bubbles is guided within channels to dynamically reconfigure the length of the radiating aperture and feed line of a slot antenna. The gallium-based liquid metal is held in position by air bubbles, and liquid-metal motion is induced when a pressure differential is created. In contrast to conventional hydraulic or pneumatic actuation of gallium-based liquid metal, the pressure actuation described here is reversible and repeatable, as the liquid metal is enveloped by a thin layer of NaOH solution, which acts as a slip layer and reduces any metal oxidation. The slot antenna achieves a 26% tunable bandwidth with a contiguous range from 1.42 to 1.84 GHz for which the S ₁₁ is <−10 dB. The peak gain ranges from 4.8 dBi in its lowest-frequency state to 4.1 dBi in its highest-frequency state.

As an alternative to classical ultra-high frequency (UHF) radio-frequency identification (RFID) tag localisation methods (e.g. angle of arrival or received signal strength), a new angle-of-activation approach is presented that uses digital multichannel transmit beamforming. By pivoting the main lobe of the interrogator's transmit antenna array utilising the very sensitive response threshold of passive UHF RFID tags, the tag direction is estimated. The presented three-channel digital beamforming hardware pivots the main lobe of a circularly polarised antenna array in a range of ±52° with an angular resolution of 0.02°. It estimates the direction of a circularly polarised tag with a mean measured accuracy of 0.9° in a basic multipath environment.

A novel substrate integrated waveguide (SIW) slot array antenna covered by circularly polarised array patches is proposed. The corner-truncated rectangular patch covering the radiating slot is introduced to generate the circular polarisation. To increase the normalised resonant conductance, the patch is moved from the centre line of the slot. The design procedure is similar to the conventional SIW slot array antenna due to the slot combined with the patch that can be seen as a radiating element. A 4×4 array antenna with uniform distribution is designed, fabricated and measured at the Ka-band. The measured results agree well with the simulations.

A wideband low-profile null-filled monopole antenna for aircraft flush-mount applications is presented. The antenna consists of a semicircular monopole and a direct-fed dual loop in a flanged cylindrical cavity. The zenith null in the monopole radiation pattern is filled by horizontal currents of the dual loop. Measurement shows that the proposed antenna has a reflection coefficient of <−10 dB over 4.53–7.43 GHz, 6.45 dBi maximum gain, −0.8 dBi filled null depth and a −3.7 dBi minimum horizontal gain at 6 GHz.

As non-volatile memory based on resistive switching becomes more mature, memristor devices with very fast switching times are becoming more prominent. However, this reported work shows that memristor devices with slow switching times (of about 10 µs) are more appropriate for use in neuromorphic systems. This is done by modelling a series of memristors that differ in their switching time. Simulation of a memristor-based neuromorphic circuit is performed using each of these modelled devices. Devices with a high switching speed require unrealistically small voltage pulses to incrementally change the memristor resistance.

A memristor circuit that can store, maintain and compare the activation of objects in the long-term memory of a cognitive architecture is described. Objects accessed frequently or recently are assigned a high activation and are more likely to be retrieved first in a search that returns multiple matches. By using memristance to store the activation values, the circuit achieves high density and fast retrieval.

The design of a wearable, portable and reconfigurable physical activity and an eight-channel bio-potential data logger, capable of increasing compliance by enabling customised feedback (i.e. calories expenditure and amount of physical activity) is presented while recording clinically meaningful information regarding a subject's health. Here an application of the device to the cardio-vascular system comprising simultaneous recording of ECG and activity in both resting and under-stress conditions is presented (clinical trials are performed under the supervision of expert cardiologists at Prince of Wales Hospital NSW, Australia). The designed device (based around the low-power LPC1768 ARM processor and the bio-potential front-end ADS1298) is an open-source one and is provided under the GPL GNU 3.0 collaborative licence.

Electroencephalography (EEG) is electrical brain activity that can be measured on the scalp with Ag/AgCl electrodes and conductive gel. However, time-consuming preparation procedures, dehydration of the gel, and skin irritation are crucial drawbacks of using such electrodes. Alternative approaches involving the use of spiky dry electrodes have their own drawbacks such as limited skin–electrode contact area, high skin–electrode impedance, and pain. Reverse-curve-arch-shaped dry EEG electrodes for use in increasing the skin–electrode contact area on hairy scalps are presented. The proposed electrode was fabricated from sterling silver using a three-dimensional printer. To increase the contact area between the skin and an electrode, an electrode was designed to have reverse-curve arches which were arranged in a row on the electrode base. The curvature of the arches was designed to match the curvature of the scalp to maximise the contact area and disperse the pressing force. To validate the proposed electrode design, comparison experiments for EEG and skin–electrode contact impedance were conducted, and the proposed electrode was found to perform better than a commercially available finger-type dry electrode.

A 10 Gbit/s serial link receiver with an offset-calibrated continuous-time linear equaliser, an adaptive one-tap half-rate speculative decision feedback equaliser (DFE) and a phase-interpolator-based clock and data recovery is presented. Adaption of the DFE is achieved by using a novel mixed-signal implementation of the sign-sign least mean square algorithm to save the cost of hardware and power. Fabricated in 65 nm CMOS technology, the receiver can totally compensate 24.85 dB channel loss at a bit error rate of 10⁻¹². The active chip area is 0.08 mm² and the total power consumption is 57 mW from a 1.2 V supply.

A high-resolution phase frequency detector (PFD) is designed for high-frequency signal detection and low jitter phase locked loop applications. The proposed PFD eliminates the reset path delay and usage of any latches, minimise the dead zone to near zero by generating narrow pulses at each input rising edge. In addition, the designed PFD completely removes unwanted output glitches, accepts inputs with a large difference in frequency, and also has the ability to drive a large capacitive load with minimal impact on performance. The proposed PFD is designed in 90 nm CMOS technology with a 1.2 V power supply. Simulation results indicate that the proposed design can operate over a wide range of frequencies from 10 kHz to 6 GHz and can detect phase differences for inputs as small as 125 fs for all frequencies of operation and for all process corners. The simulated power consumption is 75 µW at 166.6 MHz with an input phase difference of 125 fs.

A simple-but-efficient calibration-spur reduction technique is introduced that utilises the digital-and-analogue hybrid return-to-zero (HRZ) method in a current-steering DAC employing a round-robin-fashioned background current-source calibration scheme. A simulation with a 12 b 500 MS/s current-steering DAC showed 9.9 and 7.8 dB improvements in the spur-free dynamic range and signal-to-noise + distortion ratio, respectively, owing to the reduced calibration spur by the proposed HRZ technique.

This reported work addresses the deployment problem of unmanned aerial vehicles (UAVs) for multiple scattered tasks. The objective is to minimise the duration of all UAVs in reaching their assigned task locations so that the overall time needed to accomplish a mission is minimised. Taking speed and energy constraints into account, it is advantageous to transport UAVs by a large-scale ground vehicle in the shortest time. However, computation of the deployment points requires too much time for an optimal operation to accomplish a task. Therefore, an approximate solution is pursued by an algorithm for efficient deployment. The feasibility of the proposed algorithm is demonstrated through simulations.

The parabolic equation (PE) can be solved in one dimension by using the alternating direction implicit (ADI) scheme. In this way, the reduced scattered fields can be calculated line by line in each transverse plane. Thus both the CPU time and the memory requirement can be largely reduced. In this reported work, the ADI-PE method is adapted to graphics processing units (GPUs) implementation with the CUDA. The processes of both the matrix-filling and the equation-solving can be accelerated by the CUDA-based GPU. A couple of numerical results are given to demonstrate that the speedup of over 87 times can be obtained by the GPU implementation when compared with the CPU one.

For analysing the scattering problem in conductive medium with the new unconditionally stable associated Hermite (AH) finite-difference time-domain (FDTD) method, the formulation of uniaxial anisotropic perfectly matched layer absorbing boundary condition (UPML-ABC) and the total-field/scattered-field (TF/SF) implementation scheme are deduced. By virtue of AH differential transformation matrix and the eigenvalue transformation, frequency-dependent parameters of the lossy medium can be well treated in AH-UPML, without using any additional auxiliary variables. Numerical verification for the scattering of buried conductor in lossy half-space is performed. Compared with Mur's ABC, UPML increases the absorbing performance by more than 80 dB in AH-FDTD analysis.

In this Letter, a novel perceptual rate distortion optimisation (RDO) method based on the free-energy principle is proposed. The human visual system is not sensitive to the distortion hidden in disorderly regions due to the disorderly concealment effect, i.e. the free-energy principle. Based on the human visual perception, we introduce a free-energy based just-notable-difference (FEJND) model into the RDO process to adaptively adjust the Lagrange multiplier and quantisation parameter values. Experimental results show that the proposed method conceals more distortions in disorderly regions and achieves 8.46% bit-rate reduction over the JM reference software at the same perceptual quality level.

A new remotely sensed hyperspectral image data classification algorithm which integrates the adaptive mean filter and jump regression in a variational framework is introduced. First, the adaptive mean filter is used to build the posterior probability distributions in each subpixel, and the jump detection method is then used to provide the content-aware information to adjust the smoothing extent of total variation in image discontinuous areas. Experimental results on real hyperspectral datasets show the relatively good performance of the proposed algorithm in terms of the overall accuracy, average accuracy and kappa statistic.

A low-complexity distributed cyclic redundancy check (CRC) architecture for the CRC-aided early stopping unit is proposed. In the previous distributed CRC unit, the general high-order Galois field (GF) multiplier occupies almost the area of the CRC unit and requires high-hardware cost and long critical path-delay. Accordingly, a computation algorithm based on GF arithmetic is analysed and an optimal CRC unit with the small order of the GF multiplier and newly designed linear feedback shift register is proposed. The proposed CRC architecture is implemented in 65 nm CMOS process for radix-2² and radix-2⁴ parallel turbo decoders based on LTE-Advanced. In the radix-2² system, reductions of about 57.1% of gate count, 31.7% of critical path-delay and 44.1% of power consumption are achieved compared with the previous work.

Owing to the limited synchronisation time and hardware complexity of frequency-hopping multiple access (FHMA) systems, partial Hamming correlation of the frequency-hopping sequence (FHS) set applied by the FHMA system plays an increasingly important role in measuring the performance of the system. A new lower bound on the average periodic partial Hamming correlation which is tighter than the existing one in a particular case is introduced. The new lower bound includes some earlier theoretical lower bounds on the average Hamming correlation of the FHS set as special cases. A necessary and sufficient condition for obtaining optimality with regard to the new bound is also presented.

A new approach is proposed to design a phase-locked loop (PLL) as an adaptive system in order to extract the desired sinusoidal component of the input signal, and estimate its amplitude and phase. In this method, the Lyapunov stability theorem is used to obtain adjustment rules for parameters, i.e. the stability of the system and convergence of the parameters are guaranteed in the sense of Lyapunov. Simulation results show the effectiveness of the proposed structure, and demonstrate that the input signal level and adaptation gains do not change the performance of the system. The proposed PLL structure can be utilised for fast and accurate applications in signal processing, power electronics, communication systems, and control.

A new incomplete decay signal model is proposed to describe the incomplete decay effects in a time-correlated single-photon counting based fluorescence lifetime imaging microscopy system. On the basis of this model, a MUltiple SIgnal Classification algorithm to correctly estimate fluorescence lifetimes even in the presence of incomplete decay effects is modified. Monte Carlo simulations were carried out to demonstrate the performances of the proposed approach, and the results show that the proposed method is insensitive to the laser pulse rate and has a larger lifetime dynamic range compared with previously reported approaches. To the authors’ knowledge, it is the first non-fitting method that can resolve all lifetime components individually of a multi-exponential decay suffering from incomplete decay effects.

In this Letter, a comparison of the electrical performances of a standard and a double-sided 3D-radiation detectors using TCAD simulations is performed. The transient simulation is used to investigate the charge collection performance by swamping the detector with a uniform concentration of EHPs (75 e-h pairs/µm), chosen to give the same total number of carriers as a minimum ionising particle. Both detectors can be biased with the same voltage level of the readout circuitry in the standard 0.35 µm BiCMOS technology, but the standard full 3D structure is preferred for its high collection charge which means high signal to noise ratio. Finally, the impact of a temperature (300−400 K) pulse with fast transition times is studied. The impact has no significant effect on the collection current waveform of the standard but it has a big influence of the current waveform of the double-sided 3D-detectors. This makes the standard 3D-detector favourable than the double-sided 3D-detector in a harsh environment.

Resonant slot-antenna-integrated uni-travelling-carrier photodiodes were developed for enhancing terahertz (THz)-wave output powers while maintaining relatively broad bandwidths. Their output powers were two to three times larger at their peaks than those of a non-resonant bowtie-antenna-integrated device. The output powers were enhanced at frequencies from 900 GHz to 1.6 THz for a narrow-slot device and from 350 to 850 GHz for a wide-slot device. Typical output powers were 3.5 μW at 1.25 THz and 28 μW at 700 GHz for a photocurrent of 10 mA with a bias voltage of only −0.4 V.

The performance of a millimetre-wave wavelet generator under picosecond-pulsed trigger operation conditions has been evaluated experimentally. The reactively-loaded resonant tunnelling diodes metal–oxide-semiconductor field-effect-transistor circuit is shown to generate well-formed wavelets for control pulse-lengths down to approximately the free-running carrier oscillation period. Transient circuit simulations have also been implemented to provide additional insight into the observed circuit dynamics.

From a qubit in a noisy state affected by an uncontrolled decohering environment represented by a thermal bath at temperature T, estimation of the norm of its Bloch vector is considered, equivalent to estimation of its purity or of its linear entropy. The performance in estimation is assessed by the quantum Fisher information. When the qubit can be prepared with an optimal orientation precisely matched to the thermal noise, the estimation performance always degrades with an increasing temperature of the bath. On the contrary, for a non-optimal orientation of the qubit, the possibility of an enhancement of the estimation performance with an increasing temperature of the bath is demonstrated. Such behaviour shows that increased decoherence is not necessarily associated with poorer informational performance, and can be compared with stochastic resonance or useful noise effects previously reported in classical (non-quantum) estimation.

A surface plasmon resonance (SPR) sensor based on hollow-core photonic crystal fibre (HC-PCF) filled with silver nanowires is designed. The analyte and silver nanowires are full filled in the air holes of the HC-PCF to realise the SPR sensing, which is more convenient than silver coated in operation. The designed sensor is analysed through numerical simulations and demonstrated by experiments. All the results show that a blue-shift is obtained with increase of the analyte refractive index (RI), and the silver nanowires concentration has no effect on spectral sensitivity. The highest average spectral sensitivity of 14 240 nm/RIU is obtained by experiments, which is higher than that previously reported for the same type of sensors. The sensor is useful for detecting small analyte RI changes, and can also provide a reference for the implementation and application of PCF-SPR sensors with high sensitivity.

Quantum-dot cellular automata (QCA) seek potential benefits over CMOS devices such as low-power consumption, small dimensions, and high-speed operation. Two prominent QCA concerns of wire crossing complexity and circuit robustness are addressed by developing a three-step bilayer logic decomposition (BLD) methodology to design QCA-based logic circuits. The partitioning of QCA computing operations into logic layers realises considerable improvements in complexity, area, and modularity metrics. Moreover, since larger circuits are divided into two increasingly disjoint sub-planes, verification of the functionality of the design becomes compartmentalised. Design capability of the proposed approach is illustrated and analysed by implementing an area-efficient full comparator (FC) based on a novel logic realisation. The resulting 1-bit FC achieves 32% improvement in complexity metrics in comparison with the previous optimal QCA-based FC. The related waveforms used in verification of the BLD-generated FC which are obtained by the QCADesigner simulation tool are discussed as a motivating example of the BLD methodology.

Fibre links that are a few hundreds of metres long are needed for in-building networks. However, the transmission of radio-frequency (RF) signals through low-cost, easily installable plastic optical fibre (POF) is severely limited by their bandwidth-length products. Digitised radio over fibre (DROF) converts the RF signal into a baseband signal through bandpass sampling prior to its transmission, thereby reducing the fibre bandwidth required for the RF signal's transmission. For the first time, a POF-based DROF link relying on the Ethernet protocol is proposed and the initial experimental results for a link that employs a 143 m long step-index POF and uncoded RF modulation are presented. A 5 GHz RF signal through the POF is transmitted. The same method can be employed for higher RF carriers as well. The RF bit rate was ∼30% of the protocol-dependent DROF bit rate.

Enhanced unipolar orthogonal frequency division multiplexing (eU-OFDM) is a novel unipolar multicarrier modulation scheme used in intensity modulation and direct detection systems for optical wireless communication. It improves the spectral efficiency of unipolar OFDM at the expense of lower power efficiency. A novel symbol detection method exploiting the intrinsic signal structure of the received signal is proposed. Analysis and simulations show that the power efficiency of eU-OFDM could be improved up to 3 dB in the high signal-to-noise ratio regime using the proposed method.

Data are presented on 901 nm lithographic vertical-cavity surface-emitting lasers (VCSELs) demonstrating high efficiency for small VCSEL sizes, and stable beam patterns. Power conversion efficiency >40% is obtained for VCSELs ranging in size from 6 to 2 µm diameter. The 2 µm diameter VCSELs produce output powers in excess of 6.5 mW, and produce single-lobed far-field radiation patterns over the full range of operation.

A new design of a fibre-optic probe system is presented which can be used to simultaneously detect scattering signals from multiple channels for spatially offset Raman spectroscopy and Raman tomography in reflection mode. The fibre setup contained 32 source fibres with an auto-controlled optical switch and 40 detector fibres, which were deployed in a large-area planar interface. Phantom experiments were conducted using a prototype of the Raman instrument, and have demonstrated the practical applicability of the system design. The presented area-detection instrument provides an efficient platform for various reflectance applications of Raman modalities over a large area. It could also be potentially adopted for other diffuse optical-based spectroscopic and tomographic applications.

The linewidth enhancement factor (LEF) of an InAs/InGaAs quantum dot Fabry-Pérot laser in a wide wavelength range from 1110 to 1300 nm, including ground state (GS) and exited state (ES) bands, is studied. LEF spectra were derived from amplified spontaneous emission spectra measured below the threshold in the pulse regime. The ES optical transition is characterised by significantly lower values of the LEF (≤0.54) as compared to the GS (≥1.21). Moreover, a zero LEF is observed within the ES spectral band. At sufficiently high currents, a near-zero LEF (|α| ≤ 0.1) is achieved in a wide spectral interval from 1146 to 1175 nm, in which the optical gain is not less than 9 cm⁻¹.

A current-based, real-time and fast symmetry-analysis-based diagnosis method with a correlation coefficient for an open-circuit fault in a voltage-source inverter is proposed. The proposed method includes three steps: data reconstruction, features extraction, fault detection and isolation. To achieve fair robustness, real-time and fast performance, sliding windows with length of a period current samples are designed for data reconstruction. Correlation coefficients of each of the two phase currents are designed to describe the symmetry of the inverter both in a healthy and a faulty condition, and mean values of current samples during a period are used to accurately isolate the faulty switch or switches. The proposed method is current-based sharing the current sensors that are used for the current regulators, hence it is low-cost. The high effectiveness and merit of the proposed method are validated by experiments.

An adaptive pulse width modulation (PWM) strategy is proposed for four-switch three-phase inverter (FSTPI) to improve the performance and simplify the algorithm. The FSTPI has a split DC-link and one phase of the AC load is connected to the neutral point. The conventional space vector PWM has a limited performance in the low-frequency region since the phase currents are unbalanced due to the neutral-point voltage fluctuation caused by the phase current flowing through the capacitors. To solve this problem, an adaptive PWM considering the neutral-point voltage fluctuation is proposed. Moreover, sector identification is unnecessary thus the algorithm is simplified. Experimental results demonstrate the validity and feasibility of the proposed strategy.

A method for powering an on-line condition-monitoring device installed on a high-voltage transmission tower is presented. The technique uses the harvesting plate and a transformer both installed away from the transmission line to harvest the electric field energy. The circuit model of the method is analysed and the power equation given. The method has been experimentally validated. When the stray capacitance is 40 pF, about 645 mW was obtained with the load resistance 0.9 kΩ once 50 kV AC was applied on the electric field generator.

An inertial electromagnetic generator realised with magnetic spring and ferrofluid to scavenge energy from low-frequency human motion is presented. The magnetic spring is formed by magnetic attraction force between the fixed and the moving permanent magnet (PM). The kerosene-based ferrofluid is used on the moving PM and suspends the moving PM due to the fluid-magnetic pressure to reduce the mechanical friction as a fluid lubricant. The resonance frequency of the generator 3.5 Hz is experimentally validated with the sinusoidal excitation. The generator can generate an open-circuit voltage of ∼10 V and 0.26 mW average power at a load resistance of 465 Ω on resonance condition with the acceleration of 0.55778 m/s², and the corresponding power density of the generator is 40.35 μW/cm². The generator can be fastened to the belt on the backside of waist to harvest human vibration for walking and low running, and the power lever of the generator can reach 1.4–3.4 mW, which is a considerable power for low-power electronic devices.

This Letter presents a fluorescent light energy harvesting method using a capacitive touch screen to charge a phone battery. The proposed idea utilises additional power circuitry, i.e., AC to DC and DC to DC converters to provide DC power to the load. Fluorescent light used in the experiment had a 29.56-V peak-to-peak voltage at 38 kHz frequency. Experimental measurements show that the fluorescent light energy harvesting system can generate 835 μW of power.

A positive sequence voltage memory filter for numerical digital relaying applications is proposed. It is designed based on converting an analogue first-order lowpass filter into an infinite impulse response digital filter using the backward difference method. Its combination with both recursive and non-recursive phasor estimation algorithms is discussed. The obtained results reveal the decaying time constant of the proposed filter transient response is easily set, providing a simple and stable mean to adaptively adjust the positive sequence voltage memory polarisation of numerical relays.

It is shown that the electric characterisation of supercapacitors cannot simply be performed by standard LCR meters, as these devices do not account for the fractional capacitance behaviour. The accurate modelling of their frequency-dependent impedance requires direct impedance measurement using electrochemical impedance spectroscopy. A very simple and cost-effective alternative using a single operational amplifier-based circuit to extract the parameters of a supercapacitor without requiring the use of an expensive electrochemical station or a post-processing data fitting algorithm is presented.

A maximum a posteriori (MAP) algorithm based on Bayesian criterion for high-resolution direction-of-arrival (DOA) estimation in array signal processing is proposed. The generalised Gaussian distribution was considered as the prior information of sources distribution for its wide applicability. The statistic parameter of the generalised Gaussian function can be transformed according to different sources distribution conditions. Only the sparse signal recovery problem is considered. It is found that with the generalised Gaussian sparse constraint, the proposed MAP algorithm provides high DOA estimation performance in the case of limited snapshots. Simulation results are given to verify the effectiveness and efficiency of the proposed algorithm.

Tunnel transistors are one of the most attractive steep subthreshold slope devices which are being investigated to overcome power density and energy inefficiency exhibited by CMOS technology. These transistors exhibit asymmetric conduction which can cause sustained noise voltage pulses (bootstrapping) within digital tunnel FET circuits leading to delay degradation. A minor modification of the complementary gate topology to avoid the bootstrapping problem is proposed and its impact on speed at the circuit level is shown. Speed improvements up to 33% have been obtained for 8-bit ripple carry adders when implemented with the solution.

A new methodology for sparse signal acquisition using adaptive selective compressive sampling (ASCS) is presented. By employing the estimated prior information, the measurement matrix in the ASCS method can be adapted in order to selectively sense the sparse signal. The proposed ASCS method has an inherent characteristic of noise suppression, thus provides fewer noisy measurements. The experimental results demonstrate the effectiveness of the proposed scheme.

Recent studies have shown that when state-of-the-art probabilistic linear discriminant analysis (PLDA) speaker verification systems are developed with out-domain data, the mismatch between development data and evaluation data significantly degrades speaker verification performance. An unsupervised cross-domain variation compensation (CDVC) approach to compensate the domain mismatch is proposed. This approach is based on the assumption that the inter-domain variability is an additive factor with normal distribution in the i -vector space. The effect of the approach on the domain adaption challenge of the JHU 2013 speaker recognition workshop is tested. Applying the CDVC approach on evaluation i -vectors, the out-domain PLDA system achieves a relative performance improvement of 61.9% in equal error rate.

Carrier frequency offset (CFO) due to Doppler shift or misaligned frequency between the transmitter and the receiver has a serious impact on both orthogonal frequency division multiplexing (OFDM) and single carrier (SC) communication systems. Inter-carrier interference coefficients of OFDM system produced by CFO have been proposed and applied in analysing exact bit error rate (BER) upon diverse channel models. In this Letter, interference of hybrid carrier communication system based on weighted-type fractional Fourier transform impaired by CFO is investigated, which is a weighted form of multi-carrier and SC schemes, and then exact BER expression of BPSK modulation HC system is derived and verified by simulations.

A CMOS 65 nm substrate crosstalk noise sensor with exceptional performance characteristics was implemented. The sensor is integrated and fabricated onto the same die with a pin-grid array packaged ZigBee transceiver. It provides a gain of 6.5 dB in an operating bandwidth from 1 MHz to 4.5 GHz and a −1 dB gain compression point for an input signal amplitude of 124 mV. Its superior substrate noise sensing capacity is demonstrated using measurements in an advanced wireless communication system on chip, having a programmable CMOS control logic of 120 kGate acting as the substrate noise transmitter.

A dynamic power allocation scheme is proposed to maximise the energy efficiency (EE) for multi-antenna transmissions when both the imperfect channel estimation and feedback are considered. The lower bounded ergodic EE is given by the temporal correlation analysis from the Kalman estimation and the differential codebook-based feedback of the Gauss–Markov channel, based on which both the powers of data and training are adaptively optimised. Simulations show that the proposed dynamic power allocation scheme outperforms the traditional power allocation schemes.

A high-speed encoder is proposed for quasi-cyclic low-density parity-check codes. By merging some sub-matrices of a parity-check matrix H in an approximately lower triangular form, a compact encoding process is obtained, reducing pipeline stages from six to three. Moreover, well-designed circuits are used to implement back-substitution and sparse-matrix–vector multiplication. The low-density parity-check (672, 336) code in IEEE 802.15.3c shows that the proposed encoder is easy to implement, runs fast, and requires no memory.

The problem of physical layer security in a cognitive MIMO wiretap channel is studied. The problem is formulated as a secrecy rate maximisation (SRM) problem, and solved by a successive convex approximation (SCA) method. With the SCA method, the non-convex part of the SRM problem is approximated by its first-order Taylor expansion. Then, relying on solving a series of convexified optimisation problems, an iterative precoding algorithm is developed.