Output Branches Research Articles

Diode and Selective Routing Functionalities Controlled by Geometry in Current-Induced Spin-Orbit Torque Driven Magnetic Domain Wall Devices.

Research on current-induced domain wall (DW) motion in heavy metal/ferromagnet structures is crucial for advancing memory, logic, and computing devices. Here, we demonstrate that adjusting the angle between the DW conduit and the current direction provides an additional degree of control over the current-induced DW motion. A DW conduit with a 45° section relative to the current direction enables asymmetrical DW behavior: for one DW polarity, motion proceeds freely, while for the opposite polarity, motion is impeded or even blocked in the 45° zone, depending on the interfacial Dzyaloshinskii-Moriya interaction strength. This enables the device to function as a DW diode. Leveraging this velocity asymmetry, we designed a Y-shaped DW conduit with one input and two output branches at +45° and -45°, functioning as a DW selector. A DW injected into the junction exits through one branch, while a reverse polarity DW exits through the other, demonstrating selective DW routing.

High power low cross-talk single inductor dual output DC-DC converter for power management in PLL IC

Abstract A single inductor dual output DC-DC converter operating in CCM mode is proposed in this paper. Due to the more accuracy and complete small-signal model in Simulink and the dedicatedly designed compensation parameters, the circuit can broaden the load range of the converter through fast response and reduce the cross-talk effects between output branches during load adjustment. The converter is implemented in TSMC 0.18μm 1P6M BCD process with a chip area of 1.7mm*1.8mm. The simulation results show that the converter has a conversion efficiency of 81.3% at a maximum output power of 5W and the cross-regulation of less than 36mV/A at 1A load regulation.

LHFFNet: A hybrid feature fusion method for lane detection

Lane line images have the essential attribute of large-scale variation and complex scene information, and the similarity between adjacent lane lines is high, which can easily cause classification errors. And remote lane lines are difficult to recognize due to visual angle changes in width. To address this issue, this paper proposes an effective lane detection framework, which is a hybrid feature fusion network that enhances multiple spatial features and distinguishes key features throughout the entire lane line segment. It enhances and fuses lane line features at multiscale to enhance the feature representation of lane line images, especially at the far end. Firstly, in order to enhance the correlation of multiscale lane features, a multi-head self attention is used to construct a multi-space attention enhancement module for feature enhancement in multispace. Secondly, a spatial separable convolutional branch is designed for the jumping layer structure connecting multiscale lane line features. While retaining feature information of different scales, important lane areas in multiscale feature information are emphasized through the allocation of spatial attention weights. Finally, considering that lane lines are elongated areas in the image, and the background information in the image is much more abundant than lane line information, the flexibility of traditional pooling operations in capturing widely existing anisotropic contexts in actual environments is limited. Therefore, before embedding feature output branches, strip pooling is introduced to refine the representation of lane line information and optimize model performance. The experimental results show that the accuracy on the TuSimple dataset reaches 96.84%, and the F1 score on the CULane dataset reaches 75.9%.

0.35 V Subthreshold Bulk-Driven CMOS Second-Generation Current Conveyor

This study describes a high-performance second-generation Current Conveyor (CCII) operating at 0.35 V and achieving rail-to-rail operation at the Y terminal and class AB current drive at the X and Z terminals. The solution utilizes a low-voltage subthreshold bulk-driven CMOS OTA that was experimentally developed earlier, making systematic use of body terminals to improve small-signal and large-signal performance. The circuit has a high open-loop voltage gain and uses cascoded current mirror topologies, resulting in precise voltage and current transfer with bandwidths of 1.33 MHz and 2.13 MHz, respectively. The CCII offers a linear current drive up to 2.5 µA while consuming a total quiescent current of 2.86 µA (758 nA in the output branches), displaying one the highest figures of merit in terms of current utilization for sub 1 V solutions.

16.2-μW Super Class-AB OTA with current-reuse technique achieving 130.3-μA/μA FoM

In this paper, a single-stage Super Class-AB operational transconductance amplifier (OTA) using current reuse technique is presented. It is based on two scaled current mirrors located at the tail current nodes of input pairs, reusing the current through the cross-coupled input pairs by a current gain factor of α, and delivering new currents to the output branch, enhancing the overall transconductance (Gm) by a factor of (1+α). With adaptive biasing and local common mode feedback (LCMFB), slew rate (SR), open loop gain (AOL) and gain-bandwidth product (GBW) are improved. An prototype has been designed in TSMC 0.18-μm CMOS process. Post-simulation shows the positive and negative slew rate of 22V/μs and −38.3V/μs, respectively, a GBW of 3.72 MHz, a DC gain of 65.6 dB with a phase margin of 69.2° for a capacitive load of 70 pF. The proposed OTA consumes a total power of 16.2 μW under a 1.8V supply voltage.

Real-time multi-object detection model for cracks and deformations based on deep learning

The paper proposes a deep learning-based multi-object real-time detection model for concrete cracks and structural deformations. The model improves the single-stage object detection framework, You Only Look Once version 7 (YOLOv7), by incorporating convolutional block attention mechanisms and global attention mechanisms into its backbone and neck networks, respectively. It also establishes dual output branches for cracks and deformations within the output module to enable multi-object detection capabilities. Utilizing transfer learning strategies, the model effectively detects concrete cracks and structural deformations with a limited dataset. The results demonstrate that the improved YOLOv7 model significantly improves the detection of non-continuous cracks and reduces noise in complex environments, indicating strong generalization and robustness. The improved model exhibits a 4.53% increase in crack detection accuracy over the original and achieves low peak relative errors for deflection deformation in concrete beams and in-plane deformation in concrete slabs at just 0.22% and 3.05%, respectively.

Design of the Clockless Logic Gates Based on Quantum Phase Slip Junction

In this manuscript, clockless circuits of the two basic logic gates, namely AND and OR gates are designed by using the quantum phase slip junction (QPSJ). In the previously designed (existing) circuits of these gates, there is an internal clock pulse source in the output branch which is necessary for proper gate operation. When the existing AND and OR gates have to be connected to another gate, this internal clock pulse becomes in series between their outputs and the input of the next gate. As the result, the internal clock pulse must be provided by a separate source from, but synchronized to the system clock pulse (which is needed for synchronization of the inputs). This increases the design complexity. By using the proposed clockless AND and OR gates, the mentioned design complexity is removed. In addition, the proposed clockless gates have less number of QPSJs than the existing ones which allows less occupied area on the chip, higher operation frequency and smaller dynamic power dissipation.

A Dual-Band Dual-Output Filtering Power Amplifier Based on High-Selectivity Filtering Impedance Transformer

In this brief, a novel filtering impedance transformer with good selectivity and high termination impedance is proposed, of which close-formed design equations are derived and effectively verified. Then, a dual-band dual-output filtering power amplifier (PA) operating at 2.4-2.6 GHz and 3.4-3.6 GHz is designed based on the proposed filtering impedance transformer. The dual-band filtering PA contains a diplexer-like output matching network, which can separate two band signals into corresponding output branches. EM-simulated results of the diplexer-like output matching network show that the isolation between the two output ports is better than 30.7 dB. Finally, for demonstration, the dual-band dual-output filtering PA using a packaged 10 W transistor is fabricated, and the measured drain efficiencies are 45.3%-50.2% and 41.7%-53.2% at lower and higher bands, respectively. Also, a good dual-band filtering response is obtained. A good agreement between simulated and measured results is observed.

Modeling of Switching Power Losses in Cascaded H-Bridges With Unipolar PWM

This paper studies the extraction of an equivalent resistance for calculation of the switching power losses in cascaded H-bridge multilevel inverters with unipolar pulse width modulation. The concept of local and global rms currents and switching power losses are introduced in each H-bridge considering different frequencies of the inverter output and the switching network, i.e., transistor and diode. Energy conservation law has been utilized to show that the equivalent averaged local and global resistances for switching power losses are functions of load power factor angle, load rms current, and modulation index. This dependency is then reduced to only load rms current when the equivalent resistances are transformed to the inverter output branch based on the reflection rule. Consequently, a resistive model of switching power losses of the inverter is deduced, which depends neither on the knowledge of the physics of the switch, nor on complicated nonlinear equations of the semiconductor devices, and loop inductances. Double-pulse tests are conducted to acquire reliable data on the switching characteristics of the devices for different operating points.

Unsaturated polyester resin/polymethylmethacrylate waveguide-based refractive index sensor with dual-wavelength temperature compensation

This paper demonstrates an optical waveguide based- refractive index (RI) sensor using the temperature compensation method. The optical waveguide was formed using a polymethylmethacrylate sheet as the cladding material and unsaturated polyester resin as the core material. The sensor design consists of two input waveguide branches, a sensing area and an output branch. Two light emitting diodes with wavelength of 530 nm and 660 nm were used as light sources. In this work, temperature compensation was done by dual-wavelength technique in which RI and temperature sensitivities were measured at two different wavelengths at 530 nm and 660 nm. Based on the RI and temperature sensitivities, temperature compensation was implemented. Experimental findings indicated that the average relative error of the uncompensated measurement using the light source of 530 nm and 660 nm were 0.4372% and 0.2749%, respectively. Meanwhile, the average error of the temperature compensation method was 0.0344%. Hence, the temperature compensation method provides measurement error up to 92% lower compared to the uncompensated method. As such, the proposed dual-wavelength compensation method could effectively improve the RI measurement accuracy.

Choroidal vessel segmentation in SD-OCT with 3D Shape-aware Adversarial Networks

Background and Objective:Quantitative analysis of choroidal vessels is important for the diagnosis of choroidal related diseases and the cognition of the choroid. Previous studies focused on segmenting choroidal vessels in enhanced depth imaging optical coherence tomography (OCT) and swept-source OCT modalities. There is still a blank for the segmentation of 3D choroidal vessels in the spectral-domain OCT (SD-OCT) modality. Quantitative analysis of choroidal vessels is crucial for diagnosing choroidal-related diseases and understanding the choroid. Previous studies have focused on segmenting choroidal vessels in enhanced depth imaging optical coherence tomography (OCT) and swept-source OCT modalities. There is still a gap in the segmentation of 3D choroidal vessels in the spectral-domain OCT (SD-OCT) modality. Method:We propose a novel 3D CNN-based method, called Shape-aware Adversarial Networks (SAN), for segmenting choroidal vessels in SD-OCT images. To learn the 3D mapping of SD-OCT images and labels, we built a 3D U-shape network backbone with anisotropic down-sampling and up-sampling operations. To learn the morphological characteristics of choroidal vessels, we constructed an auxiliary output branch to predict a newly designed choroidal vessel shape representation generated by labels. We further introduced a label-conditional adversarial loss to enforce shape learning in a patch-based pipeline. The code for SAN is publicly available at: https://github.com/nicetomeetu21/SAN. Results:The effectiveness of SAN has been verified in data with myopia. The proposed model reported as measures of segmentation performance a mean dice similarity coefficient of 85.96% and a mean Hausdorff distance of 2.54, outperforming other 2D and 3D segmentation methods. We further applied our method in the statistical analysis of the choroidal feature between 100 eyes with low to moderate myopia and 108 eyes with high myopia. The results show that in all Early Treatment of Diabetic Retinopathy Study subfields, the choroid volumes and vascular volumes of the two groups are significantly different (P<0.01). The choroidal vascular indexes (CVI) of two groups are significantly different in most subfields (P<0.01) except for central and nasal inner subfields. The CVI of the high myopia group is higher than that of the low to moderate myopia group in most subfields except for the nasal inner subfield. Conclusions:The results demonstrate that the proposed method can accurately segment choroidal vessels from SD-OCT volumes. The quantitative analysis of choroidal biomarkers in a total of 208 myopia eyes shows the applicability of the proposed method in helping to study the choroid.

Power-Efficient Single-Stage Class-AB OTA Based on Non-Linear Nested Current Mirrors

A novel approach to design low-power area-efficient rail-to-rail output single-stage class-AB operational transconductance amplifiers (OTAs) with enhanced large- and small-signal performance to drive large capacitive loads is presented. It is based on a non-linear nested current mirror at the active load of a splitted differential input pair biased in weak inversion that boosts dynamic currents beyond their quiescent value directly at the output branch. As a result, slew rate, DC gain, gain-bandwidth product, settling time and noise performance are improved without additional circuit elements or power consumption. An OTA prototype has been fabricated in a 180-nm CMOS process, consuming a quiescent power of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2.9~\mu \text{W}$ </tex-math></inline-formula> from a supply voltage of ±0.5 V and a silicon area of 0.001 mm2. Measurement results validate the advantages of the proposal, exhibiting positive and negative slew rates of 110 V/ms and −58 V/ms, respectively, and a gain-bandwidth product of 136 kHz with a phase margin of 90° for a capacitive load of 160 pF.

A new low voltage low power fully differential CMOS class AB current output stage with unique current drive capability

ABSTRACT This paper presents a novel low-voltage low power fully differential class AB current output stage with high linearity and unique current drive capability. The proposed circuit is based on a simple and efficient structure to obtain high linearity by minimising the channel length modulation effect and also avoids cascade structure to act best in low-voltage applications. Giving the complete analyses and equations of the circuit, its properties are verified by Cadence using TSMC 0.18 µm CMOS parameters. In order to study its performance at the presence of non-idealities, Pre-layout and Post-layout both plus Monte–Carlo simulations are also performed. Under±0.9 V supply and 35 µA output bias current, it can deliver the ever high output current of ±350 mA with THD of −49 dB and output impedance of 2.8 MΩ, which leads to a unique high current drive ratio (Ioutmax/Ibias) of 10,000 and low consumed power of 150 µw in pre-layout simulation. In post-layout plus Monte–Carlo simulation, the results will be as ±320 mA output current, −42 dB THD, 2.2 MΩ output impedance, 40 µA DC current of the output branch, 8000 current drive ratio and 160 µw consumed power. These remarkable results prove the proposed novel COS not only as the unique one in the world but Longley ahead of the yet second rank COS by some unbeatable ratios as 11300×,14×,4× and3× in F T , I outmax , (Ioutmax/power), and current drive, respectively, that promise to keep it as the lead one for many years in low-voltage/power ultra-high current applications.

Field Assessment of Two Micronutrients (Zinc and Boron) on the Seed Yield and Oil Content of Mustard

In an experimental investigation, we looked into how different zinc (Zn) and boron (B) dosages affected the production and the amount of oil in mustard. Zn and B treatments, respectively, were separated into four levels: 0, 1, 2, 4, and 0, 0.4, 0.6, and 0.8 kg per hectare. We observed considerable Zn as well as B effects on mustard yield and its oil content. The maximum (1.6 Ton per hectare) and minimum (1.3 T ha−1) seed output, the maximum (1.9 T ha−1) and minimum (1.5 T ha−1) stover production, the maximum (4.9) and minimum (3.99) number branches plant−1, and the longest (114.6 cm) and shortest (87.44 cm) plant height were observed from 4 kg of Zn ha−1 and 0 kg of Zn ha−1, respectively. On the other hand, we discovered that applying 0.8 kg and 0 kg of B ha−1, respectively, resulted in the highest (1.6 T ha−1) and lowest (1.3 T ha−1) seed yield, the highest (1.8 T ha−1) and lowest (1.5 T ha−1) stover yield, the maximum (4.75) and minimum (4.02) number of branches plant−1, and the longest (118.7 cm) and shortest (85.15 cm) plant heights. The maximum seed (1.9 T ha−1) and stover output (2.0 T ha−1), tallest plant (140.9 cm), and most branches per plant−1 (5.47) were obtained when 4 kg of Zn ha−1 was given with 0.8 kg of B ha−1. However, while Zn and B were not applied, the lowest output for all of the crop attributes assessed was observed. As a result, for mustard, 4 kg of Zn ha−1 combined with 0.8 kg of B ha−1 may be advised as an effective approach in terms of seed yield and oil content.

MyoPS-Net: Myocardial pathology segmentation with flexible combination of multi-sequence CMR images.

Myocardial pathology segmentation (MyoPS) can be a prerequisite for the accurate diagnosis and treatment planning of myocardial infarction. However, achieving this segmentation is challenging, mainly due to the inadequate and indistinct information from an image. In this work, we develop an end-to-end deep neural network, referred to as MyoPS-Net, to flexibly combine five-sequence cardiac magnetic resonance (CMR) images for MyoPS. To extract precise and adequate information, we design an effective yet flexible architecture to extract and fuse cross-modal features. This architecture can tackle different numbers of CMR images and complex combinations of modalities, with output branches targeting specific pathologies. To impose anatomical knowledge on the segmentation results, we first propose a module to regularize myocardium consistency and localize the pathologies, and then introduce an inclusiveness loss to utilize relations between myocardial scars and edema. We evaluated the proposed MyoPS-Net on two datasets, i.e., a private one consisting of 50 paired multi-sequence CMR images and a public one from MICCAI2020 MyoPS Challenge. Experimental results showed that MyoPS-Net could achieve state-of-the-art performance in various scenarios. Note that in practical clinics, the subjects may not have full sequences, such as missing LGE CMR or mapping CMR scans. We therefore conducted extensive experiments to investigate the performance of the proposed method in dealing with such complex combinations of different CMR sequences. Results proved the superiority and generalizability of MyoPS-Net, and more importantly, indicated a practical clinical application. The code has been released via https://github.com/QJYBall/MyoPS-Net.

Segmentation of Clinical Target Volume From CT Images for Cervical Cancer Using Deep Learning.

Introduction: Segmentation of clinical target volume (CTV) from CT images is critical for cervical cancer brachytherapy, but this task is time-consuming, laborious, and not reproducible. In this work, we aim to propose an end-to-end model to segment CTV for cervical cancer brachytherapy accurately. Methods: In this paper, an improved M-Net model (Mnet_IM) is proposed to segment CTV of cervical cancer from CT images. An input and an output branch are both proposed to attach to the bottom layer to deal with CTV locating challenges due to its lower contrast than surrounding organs and tissues. A progressive fusion approach is then proposed to recover the prediction results layer by layer to enhance the smoothness of segmentation results. A loss function is defined on each of the multiscale outputs to form a deep supervision mechanism. Numbers of feature map channels that are directly connected to inputs are finally homogenized for each image resolution to reduce feature redundancy and computational burden. Result: Experimental results of the proposed model and some representative models on 5438 image slices from 53 cervical cancer patients demonstrate advantages of the proposed model in terms of segmentation accuracy, such as average surface distance, 95% Hausdorff distance, surface overlap, surface dice, and volumetric dice. Conclusion: A better agreement between the predicted CTV from the proposed model Mnet_IM and manually labeled ground truth is obtained compared to some representative state-of-the-art models.

A radiation hard bandgap voltage reference for the ARCADIA project

This work presents the design and characterization results of a radiation hard bandgap reference circuit fabricated in a 110 nm CMOS technology for the Main Demonstrator chip of the ARCADIA project. The design, based on a current-mode approach in order to be able to output a smaller than 1.2 V reference voltage, employs diode-connected MOSFETs instead of BJTs to enhance the radiation hardness and a second amplifier to improve the current mirror of the output branch and therefore the line regulation of the circuit. This paper describes the features of the circuit and its measured results.

Synthesis of Dual-Output DC−DC Converters With Coupled Inductor From a Generalized Output Branch

A generalized output branch with coupled inductor is proposed in this article, with which a diversity of single-input dual-output (SIDO) configurations can be synthesized through substituting it for the original inductor in the conventional single-input single-output (SISO) converters. Furthermore, for each configuration, multiple SIDO converters will be generated after realizing the auxiliary voltage source in the generalized output branch with two different nodes. Because the number of magnetic components remains unchanged in comparison with the SISO converter, high power density can be achieved in the derived SIDO converters with coupled inductor. Moreover, the cross-regulation problem is effectively alleviated, and soft-switching operation is realized for switches. Hence, both optimized steady-state and dynamic performance can be obtained. In this article, three example SIDO configurations based on buck, boost, and buck−boost converters are derived, and an example family of 12 SIDO converters from buck converter is demonstrated in detail. Afterwards, a unified operation principle and performance analysis suitable for all these 12 converters are also introduced, which is helpful for the topology comparison. Finally, experimental verification of the theoretical analysis is also conducted.

JSH-Net: joint semantic segmentation and height estimation using deep convolutional networks from single high-resolution remote sensing imagery

ABSTRACT Semantic segmentation for high-resolution remote sensing imagery is a pivotal component of land use and land cover categorization, and height estimation is essential for rebuilding the 3D information of an image. Because of the higher intra-class variation and smaller inter-class dissimilarity, these two challenging tasks are generally treated separately. This paper proposes a fully convolutional network that can tackle these problems simultaneously by estimating the land-cover categories and height values of pixels from a single aerial image. To handle these tasks, we develop a multi-task learning architecture (JSH-Net) that employs a shared feature representation and exploits their potential consistency across tasks, resulting in robust features and better prediction accuracy. Specifically, we propose a novel skip connection module that aggregates the contexts from the encoder part to the decoder part, bridging the semantic gap between them. In addition, we propose a progressive refinement strategy to recover detailed information about the objects. Moreover, we also proposed a height estimation branch on the head of the model to utilize shared features. The experiments we conducted on ISPRS 2D Labelling dataset verified that our network provided precise results of semantic segmentation and height estimation from two output branches and outperformed other state-of-the-art approaches.

A compact multiplexer for linear accelerator systems

This article presents a compact multiplexer for rapid power distribution in linear accelerator high-power waveguide systems. The compact multiplexer consists of a single-input multiple-output (SIMO) microwave network and several bandpass filters. When the input microwave frequency is within the passband of any filter, the input microwave is matched to transmit to the corresponding output port. An X-band compact multiplexer, with one input port and three output ports, is designed, fabricated, and tested. The simulated transmission efficiency of each output branch is over 91%, and the simulations agree well with the test results. The proposed compact multiplexer is attractive for high-power radio-frequency (RF) systems with rapid power distribution demands and provides a potential solution for rapid multiangle irradiation field switching technologies, such as FLASH radiotherapy and stationary computed tomography (CT) technology.