Presented is a fat dipole antenna having square radiating elements fed by a broadband balun based on coplanar-waveguide (CPW)-to-slotline field transformation followed by multi-section slotlines for impedance transformation. The measured results of the broadband balun show a passband of 500 MHz to 5 GH and insertion loss less than 1 dB. The amplitude imbalance is approximately 0.38 dB and the phase imbalance is less than 4.5^o over the entire operation range. In this design, by co-designing the radiating element with the broadband balun, a fat dipole antenna covering from 3 to 5 GHz with an omnidirectional radiation pattern and a maximum gain of 2 dBi is designed, while retaining a size of 38.9 × 80.6 mm.

Presented is an electrically small multiband monopole antenna based on complementary split-ring resonators, which are used to reduce antenna size. The antenna is fed by a three-stage microstrip line and provides 13, 17 and 16% impedance bandwidth performance covering the 3.5 GHz WiMAX and 2.4/5.2 GHz WLAN bands. Also, the proposed antenna exhibits almost an omnidirectional radiation pattern in the H-plane and a dipole-like radiation pattern in the E-plane. The return loss and radiation pattern measurements of the fabricated antenna are in very good agreement with simulation results.

Amplitude shift keying (ASK) and pulse position modulation (PPM) can be used to provide self-synchronised data and clock for wireless implantable neural recoding systems, and therefore simplifies the circuit architecture. Presented is a novel ultra-low-power clock and data recovery (CDR) circuit based on ASK-PPM. The CDR includes a 1 MHz gain-boosted front-end amplifier with an inductive load for ASK demodulation, and a charge-bump based circuit to extract and recover clock and non-return-to-zero (NRZ) data from the demodulated PPM signal. To increase the input data rate range that is able to be processed by this CDR, the reference voltage required for data recovery is adaptively generated, also from the PPM. The proposed design is fabricated in a standard CMOS 0.25 µm process. It exhibits an input sensitivity of − 13 dBm with 1 MHz input, covers an input data rate range between 19 and 76 kbit/s, and consumes only ∼ 100 µW of power.

A breast cancer imaging system with time-domain measurements is proposed. This system is based on complementary metal-oxide semiconductor circuits such that it is very cost-effective and the image is acquired in a very short time. It consists of ultra-wideband transceivers and a master controller for system control and data collection. Simulation results with a 3D FDTD hemi-spherical breast model show the proposed system can reconstruct the breast image in the presence of possible disturbances such as clock skew, clock jitter, and white noise.

Developed is a new form of ball lens onto the tip of an optical fibre for endoscopic imaging and its feasibility is evaluated. This miniaturised fibre-optic imaging probe extends its working distance without inserting a physical spacer between the cleaved fibre and the microlens as done in conventional probes. The experiment was analysed and showed that the results concur with the simulation results. Finally, this optical probe was tested with a spectral domain optical coherence tomography and the images of thin layers of the phantom specimen are taken for comparison.

A wide-range (25 ns), high-resolution (48.8 ps) clock phase shifter developed and fabricated using a 0.13 µm CMOS technology is presented. To achieve both wide range and high resolution with good linearity while minimising area and power consumption, a coarse–fine two-step phase-shifting structure is proposed. A delay-locked loop with an unconventional clock input setup is used in the fine-shifting stage to ahieve the high resolution. Test results show that the standard deviations of the differential nonlinearity and integral nonlinearity are 4.7 ps and 4.3 ps, respectively.

Presented is the error that occurs while estimating subthreshold leakage power of parallel transistor stacks in CMOS gates using leakage power models when there is no consideration of the manufacturing variations, i.e. device geometry related effects in width. For the purpose, efficient support vector machine based macromodels for characterising the transistor stacks of CMOS gates are reported, considering process parameter variations impacting e.g. length, threshold voltage, oxide thickness, supply voltage, temperature and width of the transistors. The experiments show that maximum error can go up to ∼ 15% for AOI22 and OAI22 gate under nominal values of varying parameters without considering manufacturing variations in the width.

A high-speed, large dynamic range f _S /4 bandpass ΔΣ-modulator using a first-order error-feedback loop is proposed. The internal quantiser is realised using a high-speed pipelined ADC. Error feedback is achieved by exploiting the implicit latency in a pipelined ADC. The proposed architecture achieves an SNR of 94 dB with an OSR of 32.

A three octave ultra-wideband 3-port circulator is proposed and implemented using 110 nm CMOS technology. The circulator operates across the frequency range from 0.5 to 4 GHz with supply voltage of 3 and − 3 V. The measured insertion loss is from 1.3 to 4.9 dB, return loss is better than 12 dB and isolation is more than 12 dB.

A new design is presented to decrease the contact wear between a piezoelectric transformer and its heat transfer structure by inserting a layer of polypropylene membrane between them. Without significant degradation of performance, the wear problem is well suppressed with notable extension of working life. Compared with the former wear rate of 2.72% per day on the electrodes in the first 2.5 working days, the wear rate with polypropylene membrane is decreased to only 0.03% per day within an accumulated working time of 110 days. Moreover, with the insulating layer, structure design and package become more flexible as the metal electrodes and radiator are isolated.

The raster scan method in visual information processing is the rectangular pattern of image capture and reconstruction which systematically covers the area progressively, one line at a time starting from the top left corner. This traditional scanning method might work well for still image processing but for moving images, a scanning method such as diagonal snake scan can provide a better object correlation in a sequence of image frames. A novel adaptive scanning method for object-oriented visual information processing is proposed. Here, image pixels are collected starting from the centre of the image area and progressively extended outwards similar to the water drop ripple effect. Simulation results have shown that the proposed work uses up to 33% less processing time to cover the region-of-interest in an image compared to the raster scan method. Preliminary simulation results have also shown that with the proposed adaptive scanning methodology, there is a significant improvement in the performance of the object-oriented visual compression.

A novel bit by bit row/column interleaver (BBRCI) design technique is proposed to significantly increase the minimum Hamming distance (MHD) of block accumulate codes (BACs), making them suitable for optical applications. The proposed BBRCI design method employs a separate row/column interleaver (RCI) for the transmission of each bit position of the outer block code. The numbers of rows of these individual RCIs are properly selected to increase the MHD of the resulting BAC and to reduce the frame size. It is demonstrated that BACs constructed with a BBRCI can be designed to have significantly higher MHD than similar BACs constructed with uniform interleavers (UI) and S type interleavers. Simulations show example BACs constructed with BBRCIs that perform significantly better than those BACs constructed with a UI.

A new bound on the frequency hopping (FH) sequence set with respect to the size of the frequency slot set q, the sequence length N, the family size M, the maximum periodic Hamming correlation H _m is established. The new bound is tighter than the Singleton bound on the FH sequence set when gcd (H _m + 1, N) ≠ 1.

An inductor-less on-chip radio frequency bandpass filtering (RF-BPF) technique with tunable centre frequency is proposed to replace the bulky and expensive external SAW filter. The centre frequency of the BPF not only tracks the local oscillation (LO) frequency (f _LO ), but also can be shifted from f _LO to a controllable offset for low-IF applications. The RF-BPF has been embedded into a single-ended-input differential-output radio frequency amplifier (RFA) and implemented in 65nm CMOS. The RFA draws 3.5 mA DC current from a 2.5 V power supply. Measurement results show that the BPF achieves 16 dB attenuation at 150 MHz offset from the 1.5 GHz centre frequency, and 7 MHz tunable range without changing the LO frequency.

A tri-band superconducting filter at L-band comprising three high-pole single-band sub-filters with sharp band-edges is presented to meet the high selectivity requirement. Matching junctions with high-impedance branch lines are used to connect the sub-filters and to reduce the channel interactions. MgO and LaAlO₃ substrates are utilised in the design and fabrication procedure to reduce the filter size and achieve a high performance. The whole filter has a compact size of 43 × 39.1 mm, and the measured results match well with the simulations.

A novel microstrip diplexer with low insertion loss, high selectivity and isolation performance is proposed through the combination of two compact filters composed of short-circuit centred stepped impedance resonators and, a side-coupled shorted microstrip line, which are designed for GPS and UWB application, respectively. Due to the intrinsic characteristics and mutual loading effect of the two filters, multiple transmission zeros can be created to improve the out-of-band rejection and isolation of the diplexer. More compact size is obtained without an extra junction matching networks. Simulated results of the fabricated diplexer agree well with measured ones.

A novel wide-stopband dual-band bandpass filter with a modified coupled line and lowpass filters is proposed. The modified coupled line is realised with two open stubs added to the conventional coupled line's open ends. With this proposed circuit, additional transmission zeros will appear around two passbands. Moreover, these predictable transmission zeros can be utilised to tune the two passbands’ bandwidths. Furthermore, with the lowpass filter embedded, located at the input/output port, a wide stopband can be obtained. Finally, agreement between measurement and the theoretical prediction validates the proposed structure.

A planar balun based on using parallel-coupled microstrip lines is presented. It is designed to transfer the unbalanced microstrip line to a balanced coplanar stripline as needed in many applications such as centre-fed dipole antennas. The balun includes a T-junction of stepped-impedance and tapered microstrip lines and two pairs of stepped-impedance parallel-coupled lines. The designed balun covers the band from 3 to 11 GHz with less than 0.4 dB insertion loss.

A 1.9 µm hybrid silicon/III-V laser based on a wafer bonding technique is reported. The gain materials are InGaAs multiple quantum wells grown on InP substrate, with strain compensation between barriers and wells. The III-V wafer is bonded to a silicon-on-insulator wafer with processed silicon waveguides and transition tapers. Laser emission with a threshold current of 95 mA at room temperature and 45 mA at 5 °C is demonstrated.

Energy-efficient oxide-confined 850 nm vertical-cavity surface-emitting lasers (VCSELs) for optical interconnects are presented. Error-free data transmission at 40 Gbit/s is achieved with a record low dissipated heat per bit (energy efficiency) of only 108 fJ/bit. At 36 Gbit/s across 200 m and at 40 Gbit/s across 50 m of OM3 plus multimode optical fibre error-free operation is achieved with energy efficiencies of 127 and 119 fJ/bit, respectively.

Single spatial mode GaSb-based type-I quantum well diode lasers operating near 3 µm at room temperature were designed and fabricated by chlorine-free dry etching technique. The etching profile was optimised to minimise the optical loss. The 5.5 µm-wide ridge lasers demonstrated the same slope efficiency as that of 100 µm-wide multimode devices and generated 37 mW of continuous-wave output power at 17°C.

A start-up circuit, used in a micro-power indoor light energy harvesting system, is described. This start-up circuit achieves two goals: first, to produce a reset signal, power-on-reset (POR), for the energy harvesting system, and secondly, to temporarily shunt the output of the photovoltaic (PV) cells, to the output node of the system, which is connected to a capacitor. This capacitor is charged to a suitable value, so that a voltage step-up converter starts operating, thus increasing the output voltage to a larger value than the one provided by the PV cells. A prototype of the circuit was manufactured in a 130 nm CMOS technology, occupying an area of only 0.019 mm². Experimental results demonstrate the correct operation of the circuit, being able to correctly start-up the system, even when having an input as low as 390 mV using, in this case, an estimated energy of only 5.3 pJ to produce the start-up.

In this reported work mathematical morphology theory is applied to implement the instantaneous power p − q control strategy of a shunt active power filter. The simulation results indicate that the implemented control strategy is able to completely compensate for the harmonic components in the load current and reactive power required by nonlinear loads in the power system.

In synthetic aperture radar (SAR) images, the speckle noise often corrupts salient information that is of interest (e.g. textures, small hard targets, object boundaries, etc.). To tackle such an issue, proposed is a novel anisotropic diffusion filter that manages to simultaneously fulfil competing requirements: noise reduction on homogeneous regions, weak edges preservation, and keeping hard targets intact. The capabilities of the proposed filter were proved by comparing it with another two non-linear diffusion filters applied on a Lena image corrupted by a multiplicative speckle noise and on a real SAR image acquired by the Cosmo-SkyMed satellite constellation.

Fifty years after tunnelling through semiconductor heterojunctions was originally investigated, the present authors are the first to demonstrate the required reproducibility, in wafer, between wafers in a given growth run, and from run to run, of the electrical properties required for manufacturing a microwave and millimetre-wave detector based on electron tunnelling through a thin semiconductor tunnel barrier layer.

The unscented transform (UT) is applied to the calculation of bit error rates of noisy communication channels. The UT is faster than and as accurate as the Monte Carlo method. The technique was successfully applied to several modulation schemes with different noise statistics. The UT has also the advantage of being completely defined by the moments of the random process, instead of demanding full knowledge of the probability distribution.