A novel design addressing the antenna bandwidth issue for future communication standards on handsets is presented. It consists of a tunable-antenna-pair for operation with a tunable front-end. The antennas are narrow-band and frequency-reconfigurable. This Letter focuses on the low communication bands. Measurements of the design are shown in band 12, and exhibit a receiver–transmitter isolation above 25 dB at 700 MHz.

A Fabry-Pérot cavity (FPC) array antenna providing both very high gain and circularly-polarised (CP) behaviour is proposed. To increase antenna gain and to obtain the CP characteristic, a superstrate is introduced above 2 × 2 feeding microstrip patch antennas (MPAs), which converts linearly-polarised (LP) incident waves to CP waves. Input phase difference between MPAs is designed to be exactly 90° to generate the high-quality CP characteristic in cooperation with the superstrate. By using the FPC, very high gain can be obtained with only a small number of feeders, which also provides another advantage of low mutual coupling among the neighbouring MPAs. Although the proposed antenna is very simple in its structure, it readily provides very high gain, high-quality CP property and very low axial ratio in a wide frequency range, which are great advantages compared with conventional CP antennas.

A new method is presented to estimate the loss resistance of antennas located slightly above a lossy medium. This technique is more specifically developed for the context of body area networks. This method is based on the estimation of the antenna resistance above the medium and above a perfect electric conductor.

To solve the problem of performance degradation owing to imprecise knowledge of the array steering vector and inaccurate estimation of the sample covariance matrix, a novel algorithm based on beam-space steering vector estimation for robust adaptive beamforming is presented. By using the complementary set of the spatial sector in which the actual steering vector lies, a beam-space transformation matrix can be constructed to remove the signal of interest from the sampling covariance matrix. The beam-space steering vector is estimated by solving a non-convex quadratically constrained quadratic optimisation problem. The effectiveness and robustness of the proposed algorithm is verified by the simulations.

A small-area low-ripple chopper instrumentation amplifier (IA) using a sample-and-hold circuit is proposed. The proposed chopper IA employs a ripple-reduction loop (RRL) to suppress chopper ripple which is generated by the up-modulated offset and 1/ƒ noise. The conventional RRL is not area-efficient since a sensing capacitor occupies a large chip area. Thus, the RRL using the sample-and-hold circuit to reduce the chip area is proposed. The proposed chopper IA is fabricated by using a 0.18 μm CMOS process technology and has a smaller chip area of 0.054 mm² than that of the state-of-the-art design. The measured input-referred ripples with and without the proposed RRL are less than 3.12 mV and 67 μV, respectively.

A wide tuning range wavelength swept laser for high axial resolution optical coherence tomography reported. Here, a parallel configuration of a quantum dot semiconductor optical amplifier (SOA) and a quantum well SOA is utilised for broadband optical source. The total scanning range of over 210 nm (192 nm at 10 dB) from 1153 to 1366 nm is reported. To investigate the feasiblity of the swept light source, demonstrated is an optical frequency domain imaging (OFDI) system. The system sensitivity of 101 dB is reported and axial resolution of 6.0 µm in air is achieved. Finally, ex vivo OFDI imaging of human breast tissue is performed.

Proposed is the use of the contourlet as a sparse transform which is combined with the fast iterative shrinkage/threshold algorithm (FISTA) for compressed sensing magnetic resonance imaging reconstruction. The proposed method not only inherits the simplicity and effectiveness of the original FISTA but also has the sparse curve representation ability of the contourlet. Simulation results validate the superior performance of the proposed method in terms of reconstruction accuracy and computation time.

To assist the mobility handicapped, it is necessary to support their body weight in part using an assistive device. Since the human leg can be modelled by a spring during the gait, the assistive device (or robot) should have a certain level of compliance in order to achieve natural gait assistance. A pneumatic cylinder is added in parallel to a leg (which is also a spring) for antigravity with compliance. By simulation, the compliant gravity assistance effectively reduces the required forces and torques at the joints of leg. Using this idea, a prototype of an assistive robot named SoftGait is developed and experiments are performed to verify its efficacy.

In practical ultrasonic images, post-processing that involves incoherent frame averaging such as persistence, spatial compounding and frequency compounding creates a layer of smoothed out electronic noise that is analogous to fog in photographic images. A novel approach to enhance ultrasound image contrast based on a modified dark channel model used to dehaze photographs is presented. Local minimum intensity is used to highlight image contrast, control influence of noise layer and improve detail resolution by different levels. The experimental results show that the proposed method has better performance over the existing methods in improving the quality of the ultrasonic image.

An integrated downconversion CMOS mixer incorporating a comprehensive compensation scheme is presented which aims to minimise second-order intermodulation distortion (IMD2). Unlike previously reported IMD2 calibration schemes which tune only one nonlinear factor at a time, the presented solution allows simultaneous adjustment of several different factors thus achieving a better compensation. The mixer has been implemented in UMC 0.18 µm CMOS to verify the proposed scheme and for comparison with alternative compensation methods. Measurements show that the solution described can improve the input intercept point (IIP2) by over 20 dB while maintaining good amplification and noise performance. IMD2 calibration results are presented and show useful advantages over other approaches. To the best of the authors' knowledge, this scheme for IMD2 calibration has not been previously reported.

A paired T-element that assists in successfully controlling a dual-rail delay-insensitive asynchronous circuit system that implements multiple states containing acknowledge networks, that are interdependent, is presented. A T-element allows triggering of next-state actions concurrently with negation of its acknowledge input. However, this introduces a timing assumption that can cause system failure if it is violated. With little additional overhead, the paired T-element is designed to combine the functionality of two standard T-elements and allow adjacent reads to take place when there are inter-state dependencies. This component successfully overlaps advantages of the performance optimisation of the original T-element with the safety of the solution to this control scenario, making it an important element to be used in dual-rail delay-insensitive asynchronous circuit systems.

The impact of dynamic variability due to low-frequency fluctuations on the operation of CMOS inverters, which constitute the basic component of SRAM cell, is investigated. The experimental methodology to characterise the effect of dynamic variability in a CMOS inverter is first established based on fast I–V measurements of the load current following the application of a ramp input voltage V _in(t). It is shown that, for small ramp rise times, the load current characteristics I _DD(V _in) exhibit a huge sweep-to-sweep dispersion due to low-frequency noise. The impact of such dynamic variability sources on the inverter's output characteristics V _out(V _in) is finally demonstrated, revealing a 20% noise margin reduction for the smallest inverter cell.

NanoFS is a novel file system for embedded systems and storage-class memories (like flash) and is specially designed to be directly implemented in hardware. NanoFS is based on an original internal layout intended to achieve an optimal hardware implementation of the file system's file lookup and data fetch operations. File system specification on a sample reader module completely implemented in a programmable device is introduced.

The power grid is a particular case for electromagnetic compatibility and interferences risk management because it presents a significant coupling area. The profusion of an electronic equipment connected to the network requires a control of the couplings to prevent any incident induced by a weak immunity of electronic devices to natural electromagnetic interferences in the environment or induced by EW-, HPEM-sources. It is of fundamental interest to estimate the maximal voltage (or current) magnitude induced on the power socket. The influence of loads and the transfer of energy between the power grid and the appliances require the characterisation of the electric and electronic devices. A new approach to achieve this task is proposed.

Part-based adaptive appearance model has been extensively used in increasingly popular discriminative trackers. The main problem of these methods is the stability plasticity dilemma. Embedding holistic appearance information in the part-based appearance model which is learned online to alleviate this problem is proposed. Specifically, the object is represented by sparse multi-scale Haar-like features and the appearance model is constructed with a naive Bayes classifier. Unlike the conventional methods, the classifier is trained by positive and negative samples that are weighted according to their similarity with the holistic appearance model, which is kept constant during the updating procedure. The constant holistic appearance information providing some constraints when updating the part-based appearance model makes the tracker more stable. The online updating procedure of the part-based appearance model makes the tracker adaptive enough to appearance changes. Experimental results demonstrate the superior performance of the proposed method compared with several state-of-art algorithms.

Preserving hue is an important issue for colour image enhancement. Here, a hue-preserving gamut mapping method with high saturation is proposed. Experimental results with the Macbeth colour chart and natural images show vivid colour with higher subjective image quality.

Sparse subspace clustering (SSC) has achieved state-of-the-art clustering quality by performing spectral clustering over an ℓ¹-norm based similarity graph. However, SSC is a transductive method, i.e. it cannot handle out-of-sample data that is not used to construct the graph. For each new datum, SSC requires solving n optimisation problems in O(n) variables, where n is the number of data points. Therefore, it is inefficient to apply SSC in fast online clustering and scalable grouping. An inductive spectral clustering algorithm called inductive SSC (iSSC) is proposed, which makes SSC feasible to cluster out-of-sample data. iSSC adopts the assumption that high-dimensional data actually lie on the low-dimensional manifold such that out-of-sample data could be grouped in the embedding space learned from in-sample data. Experimental results show that iSSC is promising in clustering out-of-sample data.

The forthcoming high efficiency video coding (HEVC) standard substantially improves the coding efficiency compared with the H.264 standard. It also brings in much higher computational complexity. Transcoding from H.264 to HEVC is an important step to enable gradual migration to HEVC. A low-complexity H.264 to HEVC transcoding algorithm based on motion vector clustering is proposed. Experimental results show that the proposed algorithm can reduce up to 40% of the computational time compared to the original method, while maintaining a high rate-distortion performance.

Traditionally, frequency-domain equalisers have been used to equalise the single-carrier block transmission (SCBT) signals. However, to support the frequency-domain equaliser, two fast Fourier transform modules are required, which will increase the circuit complexity and chip area. A low-complexity time-domain linear equaliser for SCBT is proposed, which is less complicated than conventional frequency-domain equalisers, at the expense of an acceptable degradation in performance. For instance, for 60 GHz conference room non-line-of-sight channels, the complexity can be reduced by half at the expense of a 1.5 dB performance loss.

A new compact Ka-band rat-race coupler (RRC) is proposed and fabricated in the 0.15-μm GaAs pHEMTs process. Since the developed RRC is implemented by asymmetrical coplanar-waveguide T-structures, it has a compact circuit footprint and a return-loss bandwidth similar to that of the conventional coupler. Besides, a short-circuited quarter-wave stub is attached into the ring of the RRC to flatten the output phase responses. The developed coupler has a 47.28% 10-dB return-loss bandwidth and only occupies a 20.89% circuit size of the conventional one. By considering the 180° ± 5° phase difference, the frequency range covers from 28.73 to 39.53 GHz with a 31.6% relative bandwidth. Over this frequency range, the amplitude imbalance varies from − 0.269 to 0.8 dB.

A common base five-stage D-band amplifier developed in a 0.18-μm SiGe HBT technology is presented. An inductive positive feedback technique is employed to enhance the gain, leading to 12.2 dB peak gain at 135 GHz with a power consumption of 67.2 mW. Measured output saturated power was − 5.8 dBm. The total chip area including pads is 1.54 × 0.56 mm².

A superconducting dual-band bandpass filter with a mixture of bandpass and bandstop sub-filters is presented. The bandpass sub-filter is designed with λ/2 double spiral resonators mid-tapped by λ/4 connecting lines which act as admittance inverters. The tap position of each resonator is adjusted to meet the required susceptance slope parameter. The bandstop sub-filter is designed with λ/2 folded resonators embedded in and coupled with the centres of the λ/4 connecting lines. The dual-band filter is designed and fabricated on an LaAlO₃ substrate with a compact size of 30.4 × 10.3 mm. The measurements show good dual-band response and match well with the simulations.

A novel compact 2.45 GHz wideband lumped balun with out-of-band suppression using a tail inductor has been proposed. By adding an extra transmission zero in even-mode around operation frequency, the tail inductor can improve phase balanced performance and enlarge operational bandwidth of the balun. Besides, the tail inductor together with a band-stop resonator can realise out-of-band transmission zero in even-mode. The measured results illustrate a relative 41% bandwidth, namely from 2.35 to 3.35 GHz. Over the operation frequency, a less than 0.5 dB amplitude imbalance and a less than ± 7° phase error have been demonstrated. The out-of-band suppressions of two output ports at 5.3 GHz are 24.5 and 27.5 dB, respectively.

K-band 0°/180° phase shifter MMIC based on balanced structure which can be applied to BPSK or QPSK modulator with dc-offset cancellation is presented. The proposed balanced phase shifter consists of four Lange couplers, two quarter-wave transmission lines and four switches for open and short terminations. The proposed structure implemented in GaAs 0.15-µm low noise pHEMT process ensures stable 0/180° phase shift while common-mode signals due to the mismatch in reflections by open and short are fed to the isolation port. The proposed phase shifter is verified by the phase shift of 179 ± 1.5° with the insertion loss of 4.5 ± 0.8 dB from 17.5 to 22.5 GHz. The fabricated phase shifter size is 2.2 × 2.85 mm including test pads.

A novel fifth-order wideband bandpass filter using open/shorted coupled lines is proposed. A wideband passband with eight transmission zeros from 0 GHz to 2f ₀ (f ₀ is centre frequency of the passband) can be achieved respectively due to the superposition of signals of the two transmission paths. A wideband bandpass filter prototype with 3-dB fractional bandwidth of 62% (2.12–3.98 GHz) is designed and fabricated.

Optical single-sideband (SSB) modulation is theoretically realised by applying two modulation signals with a 90° phase difference to the dual-electrode type electro-optic (EO) modulator. A branch-line coupler (BC) as the microwave 90° hybrid was directly prepared on the EO modulator substrate, LiNbO₃, to create two modulation signals from a single input signal. Output ports of the BC were connected to the modulation electrodes on the substrate. Thereby, an EO optical SSB modulator was realised with a single-tip and a small-sized configuration. The SSB modulator was designed and fabricated for around 10 GHz modulation frequency and 1550 nm light wavelength. The optical SSB modulation operation with a sideband suppression ratio of more than 20 dB was confirmed by the experiment.

InAs-based quantum cascade lasers (QCLs) operating near 20 µm are demonstrated. A double metal waveguide fabricated by transfer of the laser active region onto a host InAs substrate has been employed in these devices. The threshold current density of 1.6 kA/cm² at 80 K is lower than that of InP-based QCLs emitting in the same spectral region. The lasers operated in pulsed mode up to 170 K.

A nanosecond pulse erbium-doped fibre laser (EDFL) is demonstrated based on nonlinear polarisation rotation effect using a highly concentrated EDF in a ring cavity. The EDFL generates a stable multi-wavelength comb with a constant spacing of 0.22 nm in the 1533.6 nm wavelength region at a pump power of 87.6 mW. A stable mode-locked operation also commences with a constant repetition rate of 77 MHz with pulse width that varies from 5.3 to 6.8 ns as the pump power is increased from 87.6 to 168.9 mW. With careful adjustment of the polarisation controller, a stable square pulse is obtainable at a pump power of 128.3 mW with a pulse width of 6.5 ns and an average output power of 5.0 mW, which translates to the pulse energy of 35 pJ. The square pulse is generated based on the nonlinear polarisation switching effect inside the laser cavity.

The first demonstration of a silicon-based metal–insulator–semiconductor (MIS) nanoplasmonic disc resonator monolithically integrated on a CMOS compatible platform is presented. The nanoplasmonic disc resonator is integrated with Au/SiO₂/Si MIS nanoplasmonic waveguides and conventional silicon-on-insulator photonic waveguides. The device has a high plasmonic quality (Q) factor of 82.3, one of the highest reported for an integrated silicon-based nanoplasmonic device, with a compact device footprint of only 4 µm².

An all-digital field programmable gate array based electromagnetic subtraction technique for direct UWB pulse generation is presented. Using this technique, it is possible to generate Gaussian pulses, monocycles or doublets with high flexibility using a single printed circuit board layout. User-controlled parameters include amplitude, pulse-width, pulse-type, repetition rate or even modulation. Measurements indicate a Gaussian pulse with 1.24 V peak amplitude and a 10% pulse-width of 670 ps can be transmitted at a repetition frequency of 100 MHz. The corresponding − 10 dB bandwidth is 2.8 GHz.

A reconfigurable fully digital transmitter with carrier-frequency (CF) pulse-width modulation (PWM) is implemented in 180 nm CMOS for low CF applications. Through delay lines, baseband phases are mapped to position delays of 1/4 period of a carrier clock. With delay lines based CF PWM, pre-corrected baseband envelopes are converted to pulse-widths of the position-modulated carrier clock. Both the pulse-widths and the positions of the modulated carrier clock are amplified and reshaped to the sine-wave carrier with baseband components by the sequent switched-mode power amplifier and a bandpass filter. The circuit-level simulation results verify that the proposed implementation achieves reconfigurable CF and 20 dB spectrum optimisation, when compared with the existing intermediate-frequency PWM based designs.

A low-complexity low-power multi-mode quasi-cycle low-density parity check decoder architecture for 60-GHz Gbit wireless communication is presented. A novel, dynamic column shifting scheme is introduced for a multi-mode architecture that provides a low complexity and fixed throughput across all rates. Novel low-complexity local switch architecture and its control values are described to implement the dynamic shifting scheme. Post-layout results show that the proposed architecture has low power consumption at high throughputs. It reduces about 57% memory and 31% area requirement compared to previously reported architecture.

Radio frequency communications are not always possible in hostile electromagnetic environments, and the molecular diffusion channel is considered as an information channel. A binary concentration modulated channel is considered and treated is the uncertain arrival time of particles as the main source of inter-symbol interference. The analysis derives the bit-error-rate and data rate expressions for a diffusion channel with zero drift. The minimum error rate and maximum data rate are derived analytically. It is also found that in order for reliable communications to be possible, the delay time must be greater than a threshold that is proportional to the square of the communication distance.