Semi-transparent wide-band coplanar waveguide slot antennas based on commercial indium tin oxide (CITO) and silver sandwiched indium tin oxide (AgITO) conductor thin films on soda-lime glass substrate are proposed. RF/DC sputtering was functionalised to deposit the AgITO and obtaining sheet resistance (R_s ) below 1 Ω/Sq. coupled with ∼52% transparency. The associated three thin films (AgITO) deposition parameters were optimised using bandgap engineering to obtain a low ohmic interfacial contact resistance. The ultra-thin AgITO and CITO conductor films impact on the proposed antennas bandwidth and gain is introduced. The two antennas were fabricated and measured. The AgITO presented a maximum gain of 3 dBi against −2.9 dBi for CITO antenna at 5.8 GHz coupled with a nearly stable gain over a wide band of about 4 GHz.

In this Letter, a hierarchical conditional random field (HCRF) model-based gastric histopathology image segmentation (GHIS) method is proposed, which can localise abnormal (cancer) regions in gastric histopathology images to assist histopathologists in medical work. First, to obtain pixel-level segmentation information, the authors retrain a convolutional neural network (CNN) to build up their pixel-level potentials. Then, to obtain abundant spatial segmentation information in patch level, they fine tune another three CNNs to build up their patch-level potentials. Thirdly, based on the pixel- and patch-level potentials, their HCRF model is structured. Finally, a graph-based post-processing is applied to further improve their segmentation performance. In the experiment, a segmentation accuracy of is achieved on a haematoxylin and eosin stained gastric histopathological dataset with 560 images, showing the effectiveness and future potential of the proposed GHIS method.

A background calibration technique for time-interleaved analogue-to-digital converter (TIADC) over Sub-Nyquist frequency timing mismatch is proposed. When the input signal only satisfies the Nyquist sampling theorem of the overall TIADC rather than sub-ADC, both directions of estimation based on correlation and compensation based on differentiator may be opposite to the actual value, which results in a failure calibration. In this Letter, the authors propose a calibration direction correction algorithm which can detect and correct the calibration direction and assure an effective calibration in the whole Nyquist frequency of TIADC. Simulation results show that, for a four-channel 12-bit 1 GS/s TIADC, with , signal to noise and distortion ratio (SNDR) is improved from 30.1 to 73.2 dB.

In this Letter, a general compensation method for finite parasitic output impedance of G_m in G_m-C lossless integrator is discussed. The use of a G_m-simulated negative resistor in series with an integrating capacitor in a lossless integrator is shown to minimise the effects of the finite output impedance of G_m-cell. The proposed method is employed in a second-order G_m-C notch filter to improve the stop-band response of the filter.

Aiming at the problem that the traditional prior information-based defogging algorithm fails in some special scenarios, an end-to-end convolutional defogging network based on attention mechanism is proposed. The network consists of two modules: parameter estimation and image restoration. First, multi-scale convolution is used to extract image feature information. Residual network and skip connection methods are used to improve the utilisation rate of shallow network feature information. Secondly, the channel domain attention is used to add weight to the feature image input from the previous network and select useful feature information. Finally, the atmospheric visibility model is combined to achieve image visibility restoration. The experimental results show that the proposed algorithm can effectively improve the visibility of the image and the restoration effect is natural. The objective evaluation index of the benchmark datasets also shows the effectiveness of the algorithm.

The fusion of infrared and visible images has been playing an important role in various scenarios all over the world. For the fusion results from most of the existing techniques in this area, some features, such as the image contrast and edge details, are still needed to be improved. In this Letter, a new fusion method of the infrared and visible image is rendered. In this method, the infrared image is preprocessed to improve the contrast. Then, the two source images are decomposed based on non-subsampled shearlet transform (NSST). The fusion rules based on regional energy and co-occurrence filtering are proposed for the low-frequency and high-frequency NSST coefficients, respectively. Experimental results show that the proposed method can effectively retain the details of the source image, meanwhile improve the contrast of the fused image.

Recently, transformer has been applied to the image caption model, in which the convolutional neural network and the transformer encoder act as the image encoder of the model, and the transformer decoder acts as the decoder of the model. However, transformer may suffer from the interference of non-critical objects of a scene and meet with difficulty to fully capture image information due to its self-attention mechanism's dense characteristics. In this Letter, in order to address this issue, the authors propose a novel transformer model with decreasing attention gates and attention fusion module. Specifically, they firstly use attention gate to force transformer to overcome the interference of non-critical objects and capture objects information more efficiently via truncating all the attention weights that smaller than gate threshold. Secondly, through inheriting attentional matrix from the previous layer of each network layer, the attention fusion module enables each network layer to consider other objects without losing the most critical ones. Their method is evaluated using the benchmark Microsoft COCO dataset and achieves better performance compared to the state-of-the-art methods.

This Letter reports a method to obtain fine-grained energy consumption analysis of programs running on the low-power internet of things (IoT) devices. In the energy estimation domain, most state-of-the-art solutions focus on the coarse-grained approach to monitor the energy consumption of a device or an application. However, few solutions addressed energy monitoring to analyse fine-grained energy consumption of a program, especially for IoT appliances. Therefore, the authors present a fine-grained gem5-based energy profiling tool to help developers to evaluate the energy efficiency of hot spots in their programs. The authors demonstrate its usability by applying it to profiling garbage collection and property lookup in Jerryscript, a JavaScript engine specifically designed for resource-constrained IoT devices, and discover that these two operations are consuming, as they account for 17.67 and 12.26% of total energy, respectively. Therefore, this tool can aid the developers to gain insight into the energy efficiency of their software.

This Letter presents a new second-order discrete map model, which is derived from a simple sampling switch-based memristor-capacitor circuit. The memristor-based map model has infinite unstable and critical stable fixed points, and exhibits chaotic and hyperchaotic behaviours via a period-doubling bifurcation scenario. These complex dynamical behaviours are confirmed by the phase portraits, iterative sequences and basins of attraction.

In this Letter, performance of multiple users in downlink of non-orthogonal multiple access (NOMA) network considering the general shadowed fading model which was proposed in recent years is studied. In particular, main metrics including the outage probability, ergodic rate related to ordered channels are derived in novel and exact expressions for arbitrary values of the fading parameters. Moreover, numerical simulations are carried out in order to exhibit further insights on how the system performance is affected by the fading parameters.

A wideband filtering phase shifter is proposed using a cascaded single line and coupled line as a shunt stub of the centre-coupled line. Such kind of a shunt stub provides the possibility to achieve wideband impedance matching, wideband phase shift, and multiple controllable transmission zeros (TZs) at the same time. The coupling between the shunt stubs of the practical layout can generate extra TZs and enhance the frequency selectivity. Compared with the state-of-the-art designs, the proposed one has wider bandwidth, more TZs and medium size, and the former cannot juggle bandwidth and size. Two prototypes at 1.76 GHz are designed with the relative operating bandwidth for 45°±3° (90°±4.2°) phase shift of 75% (76%), five TZs and the minimum return loss of 14.7 dB.

A continuously tunable phase shifter based on a substrate integrated waveguide (SIW) is proposed for high-performance applications at millimetre-wave frequencies. The proposed phase shifter consists of two slots cut on the SIW's broad wall and a high dielectric constant (HDC) slab over the slots. The phase shifting mechanism is based on perturbing the electric field of the SIW-slot by putting a HDC slab over two slots. The fabricated phase shifter exhibits linear phase-frequency response behaviour with an average insertion loss less than 2 dB over a phase shift range of in the operating frequency band (29–31 GHz).

A broadband rectangular waveguide TE₁₀–TE₂₀ mode converter with high mode purity and conversion efficiency is designed, fabricated and measured in this Letter. Based on the broadband characteristic of ridged twist waveguide and its characteristic of polarisation direction torsion of electromagnetic wave, this mode converter achieves broadband and high-efficiency TE₁₀–TE₂₀ mode conversion. Simulation results reveal that the return loss of the proposed mode converter is >20 dB and insertion loss is <0.04 dB in whole Ka-band. Two prototype TE₂₀ mode converters were fabricated. Back-to-back millimeter-wave measurements by vector network analyser were performed to verify the design. The measured performances show return loss of the proposed mode converter is >20 dB and insertion loss is <0.1 dB in whole Ka-band, the corresponding mode conversion efficiency is >97.7%.

A novel out-of-plane integration of multilayer strip-resonators is proposed. This kind of structure retains the merit of traditional in-plane integrated ones – simple and compact-sized, meanwhile, achieves the same amount of coupling with smaller horizontal space between resonators. Further miniaturisation of the previous bandpass filters (BPFs) can be realised by re-designing them based on the proposed structure. This Letter gives an example of an S-band BPF. A 52.4% size reduction is achieved by out-of-plane integration of resonators, while maintaining a comparable performance to the previous in-plane integrated design.

The authors present the implementation of two amplifiers in a commercial flexible amorphous Indium Gallium Zinc Oxide (a-IGZO) thin-film transistor (TFT) technology: a Cherry Hooper (CH) amplifier and a 4-stage common source amplifier. The CH amplifier is designed as a pre-amplifier for wireless receivers. It is optimised for a high gain-bandwidth product (GBW). From a supply voltage of V it provides 19.4 dB gain and has a -bandwidth of 5.3 MHz, while consuming 0.2 mW. It has a GBW of 49.35 MHz, which is more than a factor two better than previously reported a-IGZO TFT amplifiers. The 4-stage common source amplifier is designed as output buffer, has a very wide range of operating conditions and strong robustness against manufacturing tolerances. From a supply voltage of V it provides 28.9 dB gain, has a -bandwidth of 1.2 MHz, and a GBW product of 33.43 MHz, while consuming 14.2 mW. Both circuits can operate from a supply voltage between 2.5 and 10 V.

The lateral diffusion length (L _h) of minority carriers in LWIR InAs/GaSb superlattice detectors grown by metalorganic chemical vapour deposition was evaluated. The deeply-etched PNn device exhibits a diffusion-limited behaviour at 80 K, with a dark current density as low as 9.1 × 10⁻⁶ A/cm² at −0.1 V and a 50% cut-off of 10.1 μm. In shallow-etched pixels with a common absorber, both the photo-current and the dark current show a size-dependent behaviour. L _h deduced from the two methods are 211 and 251 μm, respectively, which are longer than those in superlattice materials grown by molecular beam epitaxy.

Herein, for the first time, the role of double-layer electron transporting layer (ETL), i.e. ZnO/SnO₂ on the efficiency of lead sulfide quantum dots solar cells is demonstrated. Using this approach, a device with power conversion efficiency of 10.6% is achieved mainly due to the enhancement of open circuit voltage (V _OC). The authors reduce the loss-in-potential by 80 mV using ZnO/SnO₂ double-layer ETL.

A novel ultrathin barrier AlGaN/GaN hybrid-anode-diode (UTB-HAD) with in-situ Si₃N₄-cap passivation is experimentally demonstrated. The forward turn-on voltage (V _on) of the UTB-HAD is determined by the intrinsic threshold voltage of the two-dimension electron gas (2DEG) channel, which can be precisely controlled by tailoring the as-grown AlGaN-barrier thickness. The typical V _on as low as 0.48 V is obtained by using the UTB AlGaN/GaN with a barrier thickness of 4.9 nm. The MOCVD has grown in-situ Si₃N₄-cap and the LPCVD-Si₃N₄ bilayer passivation stack is developed to effectively restore the 2DEG in the UTB AlGaN/GaN heterostructure and simultaneously improve the dynamic characteristics of the diode. The UTB-HAD and the novel passivation scheme are of great potential for power applications.

Similarity measures play a significant role in adaptive filtering. Previous work such as correntropy and kernel risk-sensitive loss (KRSL), has successfully improved the technology of adaptive filtering in terms of robustness against outliers, fast convergence speed and high filtering accuracy. Based on KRSL, a newly raised similarity measure, complex KRSL (CKRSL), was proposed by extending KRSL to the complex domain. It successfully gains superior performance than other similarity measures in adaptive filtering algorithms. However, the minimum CKRSL (MCKRSL) algorithm may result in poor performance when the parameters are not properly chosen. In this Letter, an adaptive parameter selection is proposed to help the MCKRSL algorithm improve performance while overcoming the uncertainty in artificial selection. The proposed MCKRSL with variable parameters (MCKRSL-VP) algorithm updates the risk-sensitive parameter and kernel width by making the iteratively squared bias as small as possible. A moving average scheme is further used to smoothly update the risk-sensitive parameter and kernel width. Finally, the authors verify that MCKRSL-VP performs better than other algorithms by simulations.

In this Letter, the Logical Jaya optimisation is proposed as an extension of the Jaya optimisation algorithm to improve the cognitive workload (CW) assessment technique where channel selection for the EEG signal act as a binary optimisation problem. Channel selection is very crucial, time-consuming and requires expertise, specially when brain cognitive load is considered. The proposed approach is designed such that it not only improves the performance of the assessment model of CW but also reduces the computational cost. The approach also helps in the automation of brain analysis. The results obtained show that performance is improved by 22% than existing approaches to an average of >90% accuracy in different scenarios. The channels obtained using the approach also provided accurate active brain regions during CW analogous to previous studies.