Integrated power electronics module based on chip scale packaged power devices

High performance can be obtained for the integrated power electronics module（IPEM） by using a three-dimensional packaging structure instead of a planar structure. A three- dimensional packaged half bridge-IPEM （HB-IPEM）, consisting of two chip scale packaged MOSFETs and the corresponding gate driver and protection circuits, is fabricated at the laboratory. The reliability of the IPEM is controlled from the shape design of solder joints and the control of assembly process parameters. The parasitic parameters are extracted using Agilent 4395A impedance analyzer for building the parasitic parameter model of the HB- IPEM. A 12 V/3 A output synchronous rectifier Buck converter using the HB-IPEM is built to test the electrical performance of the HB-IPEM. Low voltage spikes on two MOSFETs illustrate that the three-dimensional package of the HB-IPEM can decrease parasitic inductance. Temperature distribution simulation results of the HB-IPEM using FLOTHERM are given. Heat dissipation of the solder joints makes the peak junction temperature of the chip drop obviously. The package realizes three-dimensional heat dissipation and has better thermal management.

A complementary integrated Transformer network for hyperspectral image classification

In the past,convolutional neural network(CNN)has become one of the most popular deep learning frameworks,and has been widely used in Hyperspectral image classification tasks.Convolution(Conv)in CNN uses filter weights to extract features in local receiving domain,and the weight parameters are shared globally,which more focus on the highfrequency information of the image.Different from Conv,Transformer can obtain the long‐term dependence between long‐distance features through modelling,and adaptively focus on different regions.In addition,Transformer is considered as a low‐pass filter,which more focuses on the low‐frequency information of the image.Considering the complementary characteristics of Conv and Transformer,the two modes can be integrated for full feature extraction.In addition,the most important image features correspond to the discrimination region,while the secondary image features represent important but easily ignored regions,which are also conducive to the classification of HSIs.In this study,a complementary integrated Transformer network(CITNet)for hyperspectral image classification is proposed.Firstly,three‐dimensional convolution(Conv3D)and two‐dimensional convolution(Conv2D)are utilised to extract the shallow semantic information of the image.In order to enhance the secondary features,a channel Gaussian modulation attention module is proposed,which is embedded between Conv3D and Conv2D.This module can not only enhance secondary features,but suppress the most important and least important features.Then,considering the different and complementary characteristics of Conv and Transformer,a complementary integrated Transformer module is designed.Finally,through a large number of experiments,this study evaluates the classification performance of CITNet and several state‐of‐the‐art networks on five common datasets.The experimental results show that compared with these classification networks,CITNet can provide better classification performance.

Synchronization analysis on cascaded multilevel converters with distributed control

Cascaded multilevel converters built with integrated modules have many advantages such as increased power density,flexible distributed control,multi-functionality,increased reliability and short design cycles.However,the system performance will be affected due to the synchronization errors among each integrated modules.This paper analyzes the impact of the three kinds of synchronization errors on the whole system performance,as well as detailed synchronization implementation.Some valuable conclusions are derived from the theoretical analysis,simulations and experimental results.

Compact temperature-insensitive modulator based on a silicon microring assistant Mach-Zehnder interferometer

On the silicon-on-insulator platform, an ultra compact temperature-insensitive modulator based on a cascaded microring assistant Mach-Zehnder interferometer is proposed and demonstrated with numerical simulation. According to the calculated results, the tolerated variation of ambient temperature can be as high as 134 ℃ while the footprint of such a silicon modulator is only 340 μm2.

Theory study on a photonic-assisted radio frequency phase shifter with direct current voltage control

A photonic-assisted radio frequency phase shifter with direct current voltage control is proposed using a polymer- based integrated Mach-Zehnder modulator. A closed-form expression of radio frequency （RF） signal power and phase is given. Theoretical calculation reveals that by carefully setting the bias voltages, RF signal power variation lower than 1-dB and phase accuracy less than 3° can be achieved and are not degraded by perturbation of modulation index once the bias voltage drift is kept within -3% -- 3%.

Lightwave Modulation Techniques for Advanced Modulation Formats

Arbitrary waveform can be synthesized by two orthogonal components: in-phase and quadrature components. Thus, if we can manipulate the two components independently, various complicated modulation can be achieved. This technique is called vector modulation, which is commonly used in wireless radio communication systems where the carrier frequency is lower than 100GHz. On the other hand, in lightwave transmission systems, the carrier frequency is higher than 100THz, and the stability of the signal sources would be much poorer than in radio systems, so that the simplest modulation format (on-off-keying) was often used in commercial systems. However, recently, various modulation techniques such as quadrature amplitude modulation, etc., have been demonstrated by using precise and rapid lightwave control. In this report, we review recent progress in optical vector modulation technologies which can generate highspeed and complicated optical signals. Integrated optical modulators can synthesize multi-level signals in photonic circuits.

Efficient evanescent coupling design for GeSi electro-absorption modulator

Efficient coupling from the silicon waveguide to the GeSi layer is the key to success in the GeSi electro-absorption （EA） modulator based on evanescent coupling. A lateral taper in the upper GeSi layer has room for increasing the modulating efficiency and alleviating the sensitivity of the extinction ratio （ER） and insertion loss （IL） to the length of the active region. The light behavior and the effect of the taper are explored in detail using the beam propagation method （BPM）. After optimization, the light can nearly be totally confined in the GeSi layer without any oscillation. The modulator with the designed taper can achieve low IL and high ER.

Recent progress in hydrogen:From solar to solar cell

Hydrogen,meeting the requirements of sustainable development,is regarded as the ultimate energy in the 21st century.Due to the inexhaustible and feasible of solar energy,solar water splitting is an immensely promising strategy for environmental-friendly hydrogen production,which not only overcomes the fluctuation and intermittency but also contributes to achieving the mission of global“Carbon Neutrality and Carbon Peaking”.However,there is still a lack of a comprehensive overview focusing on hydrogen progress with a discussion of development from solar energy to solar cells.Herein,we emphasize several solar-to-hydrogen pathways from the basic concepts and principles and focus on photovoltaic-electrolysis and photoelectrochemical/photovoltaic systems,which have achieved solar-to-hydrogen(STH)efficiency of over 10%and have extremely promising for large-scale application.In addition,we summarize the challenges and opportunities faced in this field including configuration design,electrode materials,and performance evaluation.Finally,perspectives on the potential commercial application and scientific research for the further development of solar-to-hydrogen are analyzed and presented.

112 Gbit/s transmitter optical subassembly based on hybrid integrated directly modulated lasers

Based on the hybrid integration technology, an ultra-compact and low cost transmitter optical subassembly module is proposed. Four directly modulated lasers are combined with a coarse wavelength division multiplexer operated at the O-band. The bandwidth for all channels is measured to be approximately 3 GHz. The 112 Gb∕s transmission is experimentally demonstrated for a 10 km standard single mode fiber（SSMF）, in which an optical isolator is used for avoiding the back-reflected and scattered light to improve the bit error rate（BER） performance. A low BER and clear eye opening are achieved for 10 km transmission.

Constructing irreducible integrable modules for extended baby TKK algebra

In this paper, we give explicit realizations for the irreducible integrable modules, which were clas- sified in Chang and Tan [Pacific J Math, 2011, 252: 293-312], of the extended baby TKK algebra. Moreover, conditions for these modules to be unitary are determined.

Specificity-preserving RGB-D saliency detection

Salient object detection(SOD)in RGB and depth images has attracted increasing research interest.Existing RGB-D SOD models usually adopt fusion strategies to learn a shared representation from RGB and depth modalities,while few methods explicitly consider how to preserve modality-specific characteristics.In this study,we propose a novel framework,the specificity-preserving network(SPNet),which improves SOD performance by exploring both the shared information and modality-specific properties.Specifically,we use two modality-specific networks and a shared learning network to generate individual and shared saliency prediction maps.To effectively fuse cross-modal features in the shared learning network,we propose a cross-enhanced integration module(CIM)and propagate the fused feature to the next layer to integrate cross-level information.Moreover,to capture rich complementary multi-modal information to boost SOD performance,we use a multi-modal feature aggregation(MFA)module to integrate the modalityspecific features from each individual decoder into the shared decoder.By using skip connections between encoder and decoder layers,hierarchical features can be fully combined.Extensive experiments demonstrate that our SPNet outperforms cutting-edge approaches on six popular RGB-D SOD and three camouflaged object detection benchmarks.The project is publicly available at https://github.com/taozh2017/SPNet.

Integrable Representations for the Extended AlPine Lie Algebra Coordinated by Quantum Tori in Two Variables

In this paper we classify the irreducible integrable modules for the core of the extended affine Lie algebra of type Ad-1 coordinated by Cq with finite-dimensional weight spaces and the center acting trivially, where Cq is the quantum torus in two variables.

Optical generation of UWB pulses utilizing Fano resonance modulation

In this paper,we reported an integrated method to generate ultra-wideband(UWB)pulses of different orders based on a reconfigurable silicon micro-ring resonator-coupled Mach–Zehnder interferometer.Under proper operating conditions,the device can produce Fano resonances with a peak-to-valley extinction ratio of above 20 dB.UWB monocycle and doublet signals with picosecond pulse widths are produced when the microring resonator is modulated by square and Gaussian electrical pulses,respectively.With our Fano resonance modulator on silicon photonics,it is promising to foresee versatile on-chip microwave signal generation.