Search

In this paper, we present a single nano ring resonator using Annular Photonic Crystal (APC) with ultra-compact size for applications in versatile optical network components. The proposed resonator is designed in the hexagonal lattice, made up of Silicon (Si) planar and annular rods, where the annular rods are filled with Silicon Nitride (Si3N4). Despite being pervasive, the proposed structure operates mostly in the C-band wavelength, providing high resonation, low insertion loss, high contrast ratio, and large bandwidth with low loss comprising a single resonator ring with photonic crystal waveguides. This structure is used to realize several high-performance optical network devices like optical ring resonator, 4 × 2 reversible encoder, 1 × 2 power splitter, and a multiplexer. Several parameters and their performances are optimized for this miniaturized photonic device using Finite Difference Time Domain (FDTD) method. This device works in the quantum regime as a slow light device with a better squeeze factor enabling quantum computing. The proposed nanostructure, having a Quality factor (Q) of 396.05, is highly suitable for optical network applications. [ABSTRACT FROM AUTHOR]

Summary Multivalued logic (MVL) offers higher information density as compared with binary logic systems for an equal number of digits. In a ternary microprocessor, memory access is the most time‐ and power‐consuming action. In order to design a ternary computer, the possibilities for designing ternary encoders and decoders must be examined. This paper presents the designs of 9:2 unbalanced ternary encoder (TENC) and 2:9 unbalanced ternary line address decoder (TDEC) based on novel designs of standard ternary inverter (STI), ternary NAND (TNAND), and ternary NOR (TNOR) logic gates using resistive random‐access memory (RRAM) and carbon nanotube field effect transistor (CNTFET) technology. The simulation results indicate an improvement in performance parameters such as power consumption, delay, power–delay product (PDP), and component count of the proposed circuits in comparison with the different existing counterparts. Moreover, a detailed analysis is carried out on the impact of different process, voltage, and temperature (PVT) variations on the performance metrics, including propagation delay, power consumption, and PDP of the 9:2 unbalanced ternary encoder and 2:9 unbalanced ternary line address decoder. The proposed ternary encoder circuit shows an improvement of 62% in delay, 71% in power consumption, 88.6% improvement in PDP, and 53% reduction in CNTFET count, whereas the ternary decoder circuit exhibits an improvement of 6.6% and 94% in delay and 43% and 94% improvement in power consumption, respectively. Moreover, the CNTFET count is reduced by 16.07% and 46%. [ABSTRACT FROM AUTHOR]

Ocean waves play a significant role in coastal dynamics and the management of coastal development activities. However, there are limited instruments to measure the wave height, particularly in developing countries. This study introduces a novel instrument design for simple and cost-effective wave height measurement. The proposed instrument combines a Hall-effect sensor and an accelerometer to capture the rise and fall of a buoy in response to wave movements. The Hall-effect sensor as a rotary encoder detects the rotation of the buoy while the accelerometer measures the tilt of the instrument caused by ocean waves. The instrument is constructed using PVC pipes and incorporates long-range (LoRa) communication technology for real-time monitoring. Experimental tests were conducted at a study site in Banyuwangi Regency, Indonesia, where the performance of the instrument was compared with an ultrasonic sensor-based instrument. The results validate the effectiveness of the proposed instrument design for wave height measurement. The collected data were transmitted via the LoRa communication system, enabling convenient monitoring and analysis of sea-level changes. The valuable contribution of the instrument to the field stems from its precision in measuring the wave height, adaptability to diverse conditions, and potential applicability in shallow waters. [ABSTRACT FROM AUTHOR]

Traditional traffic flow prediction models model historical data in terms of time and space, ignoring the internal potential temporal periodicity of traffic data and the distance characteristics and similarity spatial characteristics of nodes between traffic networks. Based on this, this paper proposed a multi-channel spatio-temporal encoder model MC-STGNN for traffic flow prediction to improve the accuracy of traffic flow prediction. Firstly, it processed the traffic data into a three channel periodic time series, and encoded the overall sequence data with temporal and adaptive spatial positions to extract dynamic correlations between road network nodes. Secondly, it introduced a multi-heads self-attention mechanism with convolutional structure to capture varying degrees of temporal correlation of periodic data to a greater extent. Finally, it proposed a graph generator to generate a new spatiotemporal map, extracting similarity and distance features between road network nodes, and integrating the spatial information of the original map and the new spatiotemporal map using a gated graph convolutional network. It conducted comprehensive traffic flow prediction experiments for an hour on the highway datasets PEMS03 and PEMS08. The experimental results show that the MC-STGNN model has better performance indicators compared to other baseline models, indicating that the MC-STGNN model has better modeling ability. [ABSTRACT FROM AUTHOR]

Continuous monitoring of artery and vein vessels in the human eye to prevent the loss of eyesight is essential in the diagnosis of diabetic retinopathy. The progression of diabetic retinopathy is echoed by the internal anatomical changes of the retina, such as changes in the artery vein ratio, formation of fake vessels, tortuosity, lesion formation, etc. Among these symptoms, calculation of the artery vein ratio is still a challenging task since the visibility of the artery and vein changes over different regions in the fundus image. The proposed Atrous Depth Concatenated neural network with the enriched encoder (EEDCFCNN) architecture for artery vein classification is based on the deep semantic segmentation architecture. The proposed architecture can achieve an improved result on the public databases DRIVE, INSPIRE, and IOSTAR. [ABSTRACT FROM AUTHOR]

The geometric error of machine tools can usually be reduced or compensated by various methods, but the influence of encoder performance on machining quality is often ignored. The purpose of this work is to analyze the influence of micro-cyclic error in linear encoders on milling surfaces and establish a compensation method to eliminate the encoder error. First, the origin of the micro-cyclic error of the linear encoder is analyzed. The micro-cyclic errors of the horizontal linear axes of an ultraprecision machine tool are preliminarily measured. The errors are also identified through milling tests. Second, a surface topography model for milling surfaces is established with consideration of the encoder micro-cyclic error. Based on the established model, a compensation method for the encoder micro-cyclic error is proposed. Finally, the error distribution model is validated by diamond ball-end milling experiments. Further experimental results confirm that the proposed compensation method can effectively remove the encoder error-related stripes from machined surfaces. The surface quality is considerably improved. [ABSTRACT FROM AUTHOR]

The main disadvantage of encoders currently used to measure the object motion, is the required mechanical coupling with the observed object, that significantly limits its scope of application. The paper presents theoretical results of the contactless (without mechanical contact with the observed object) encoder based on the Doppler effect. The block diagram and the analysis are presented for the proposed contactless encoder. [ABSTRACT FROM AUTHOR]

Pre-training emerges as a new learning paradigm in natural language processing and computer vision. It has also been introduced into sequential recommendation in several seminal studies for alleviating data sparsity issue. However, existing methods adopt the bidirectional transformer as the encoder which suffers from two drawbacks. One is insufficient intention modeling since the transformer architecture is suitable for extracting distributed consumption intention but cannot well catch users' concentrated and occasion consumption intentions. The other is information leakage caused by foreseeing the future item in advance during the bidirectional encoding process. To address these problems, we propose to construct more effective encoders in pre-training for sequential recommendation. Specifically, we first decouple the original bidirectional process in transformer structure into two unidirectional processes which can avoid the information leakage problem and capture the distributed consumption intention. We then employ the locality-aware convolutional neural networks (CNNs) with narrow receptive field for concentrated consumption modeling. We also introduce a random shuffle strategy to empower CNN with the ability of modeling the occasion consumption. Experiments on five datasets demonstrate that our method improves the performance of various types of downstream sequential recommendation models to a large extent, and it also generates the overall better performance than the state-of-the-art self-supervised pre-training methods. [ABSTRACT FROM AUTHOR]

The early automatic detection of brain tumors in MRI scans is a challenging endeavor due to the high resolution of the images. For a very long time, continual research efforts have been floating a new notion of substituting various grayscale anatomic parts of diagnostic pictures with suitable colors. If successful, this would be an effective way for radiologists to circumvent the challenges they now encounter. The coloring of grayscale photos is a complex process that aims to improve the contrast of different sections of an image by changing grayscale images into color pictures with high levels of contrast. It is common for the predictions to be lacking in fine detail when simply a U-Net design is used; to assist alleviate this issue, cross or skip connections may be introduced between the blocks of the network. Instead of creating a skip connection every two convolutions as it now is in a ResBlock, the skip connections cross from a portion of the same size in the downsampling route to a part in the upsampling path. This improves the overall accuracy of the model and performs better when compared to traditional UNet model. [ABSTRACT FROM AUTHOR]

Traditional recommendation algorithms map text content through topic model or mean value of word vectorization. For the issue that existing methods cannot make full use of text information or ignore word order information, this study proposes a multi-task learning recommendation method for scientific literature. Based on the multi-task learning framework, an encoder is designed and a GL model is established. The GL model is trained to combine content recommendation and text metadata prediction, which improves the sparsity of traditional collaborative filtering and regularizes the collaborative filtering model. Finally, an evaluation test is carried out on public and private data sets respectively, and the superiority of the proposed method is demonstrated by comparing with the existing classical methods. [ABSTRACT FROM AUTHOR]

Clinically, for observing the healing of the patient's wound, doctors need to insert a cotton swab into the deepest part of the wound to detect the depth of the wound. This measurement method will cause discomfort to the patient. Therefore, obtaining wound depth information directly from wound images is very important for doctors to understand the degree of wound healing. In this paper, we propose the generative adversarial network of chronic wound depth detection (CWD2GAN) to generate wound depth maps of four different shades of color according to the changes of the wound area in the chronic wound image. In CWD2GAN, the generator, which can generate the wound depth map, is composed of three parts: encoder, decoder, and concatenation. And, the discriminator uses the concept of cGAN. It can not only judge whether the generator produces an image but also know that this image is a depth map. In experimental results, the accuracy, sensitivity, specificity, and precision of CWD2GAN are 84.8%, 84.6%, 84.9%, and 86.3%, respectively. The results indicate that our proposed method can accurately generate the different depths layer in a chronic wound image, and reduce the pain caused by invasive testing for patients. [ABSTRACT FROM AUTHOR]

The brain is the center of human control and communication. Hence, it is very important to protect it and provide ideal conditions for it to function. Brain cancer remains one of the leading causes of death in the world, and the detection of malignant brain tumors is a priority in medical image segmentation. The brain tumor segmentation task aims to identify the pixels that belong to the abnormal areas when compared to normal tissue. Deep learning has shown in recent years its power to solve this problem, especially the U-Net-like architectures. In this paper, we proposed an efficient U-Net architecture with three different encoders: VGG-19, ResNet50, and MobileNetV2. This is based on transfer learning followed by a bidirectional features pyramid network applied to each encoder to obtain more spatial pertinent features. Then, we fused the feature maps extracted from the output of each network and merged them into our decoder with an attention mechanism. The method was evaluated on the BraTS 2020 dataset to segment the different types of tumors and the results show a good performance in terms of dice similarity, with coefficients of 0.8741, 0.8069, and 0.7033 for the whole tumor, core tumor, and enhancing tumor, respectively. [ABSTRACT FROM AUTHOR]

3D image reconstruction using multi-view imaging is widely utilized in several application domains: construction field, disaster management, urban planning, etc. The 3D reconstruction from the multi-view image is still challenging due to the high freedom and inaccurate reconstruction. This research introduces the hybrid deep learning technique for reconstructing the 3D image, in which the C-dual attention layer is proposed for generating the feature map to support the image reconstruction. The proposed 3D image reconstruction uses the encoder–decoder–refiner which is utilized for reconstruction. Initially, the features are extracted from the AlexNet and ResNet-50 features automatically. Then, the proposed C-dual attention layer is utilized for generating the inter-channel and inter-spatial relationship among the features to obtain enhanced reconstruction accuracy. The inter-channel relationship is evaluated using the channel attention layer, and the inter-spatial relationship is evaluated using the spatial attention layer of the encoder module. Here, the features generated by the spatial attention layer are combined to form the feature map in a 2D map. The proposed C-dual attention encoder provides enhanced features that help to acquire enhanced 3D image reconstruction. The proposed method is evaluated based on loss, IoU_3D, and IoU_2D, and acquired the values of 0.0721, 1.25 and 1.37, respectively. [ABSTRACT FROM AUTHOR]

The aim of this study was to examine the validity and reliability of an Inertial Measurement System integrated into a secondary pulley (IMS) for determining linear velocity during flywheel squat exercises. Thirty-one male participants who were highly experienced in a flywheel resistance exercise training performed flywheel squat exercises with three incremental loads, and mean velocity (MV), mean propulsive velocity (MPV) and max velocity (Vmax) of the exercises were simultaneously recorded with a validated linear encoder and the IMS, in two different sessions. Validity was analyzed using ordinary least products regression (OLP), Lin's concordance correlation coefficient (CCC), and Hedge's g for the values from the linear encoder and the IMS. Test-retest reliability was determined by coefficient of variation (CV), Intraclass correlation coefficient (ICC), and standard error of measurement (SEM). Results showed a high degree of validity (OLP intercept = −0.09–0.00, OLP slope = 0.95–1.04, CCC = 0.96–0.99, Hedge's g < 0.192, SEM = 0.04–0.08) and reliability (CV < 0.21%, ICC > 0.88, SEM < 0.08). These results confirm that the IMS provides valid and reliable measures of movement velocity during flywheel squat exercises. [ABSTRACT FROM AUTHOR]

The article addresses the issue of mobile robotic platform positioning in GNSS-denied environments in real-time. The proposed system relies on fusing data from an Inertial Measurement Unit (IMU), magnetometer, and encoders. To get symmetrical error gauss distribution for the measurement model and achieve better performance, the Error-state Extended Kalman Filter (ES EKF) is chosen. There are two stages of vector state determination: vector state propagation based on accelerometer and gyroscope data and correction by measurements from additional sensors. The error state vector is composed of the velocities along the x and y axes generated by combining encoder data and the orientation of the magnetometer around the axis z. The orientation angle is obtained from the magnetometer directly. The key feature of the algorithm is the IMU measurements' isolation from additional sensor data, with its further summation in the correction step. Validation is performed by a simulation in the ROS (Robot Operating System) and the Gazebo environment on the grounds of the developed mathematical model. Trajectories for the ES EKF, Extended Kalman Filter (EKF), and Unscented Kalman Filter (UKF) algorithms are obtained. Absolute position errors for all trajectories are calculated with an EVO package. It is shown that using the simplified version of IMU's error equations allows for the achievement of comparable position errors for the proposed algorithm, EKF and UKF. [ABSTRACT FROM AUTHOR]

This paper presents a low-area 8-bit flash ADC that consumes low power. The flash ADC includes four main blocks—an analog multiplexer (MUX), a comparator, an encoder, and an SPI (Serial Peripheral Interface) block. The MUX allows the selection between eight analog inputs. The comparator block contains a TIQ (Threshold Inverter Quantization) comparator, a control circuit, and a proposed architecture of a Double-Tail (DT) comparator. The advantage of using the DT comparator is to reduce the number of comparators by half, which helps reduce the design area. The SPI block can provide a simple way for the ADC to interface with microcontrollers. This mixed-signal circuitry is designed and simulated using 180 nm CMOS technology. The 8-bit flash ADC only employs 128 comparators. The applied input clock is 80 MHz, with the input voltage ranging from 0.6 V to 1.8 V. The comparator block outputs 127 bits of thermometer code and sends them to the encoder, which exports the seven least significant bits (LSB) of the binary code. The most significant bit (MSB) is decided by only one DT comparator. The design consumes 2.81 mW of power on average. The total area of the layout is 0.088 mm2. The figure of merit (FOM) is about 877 fJ/step. The research ends up with a fabricated chip with the design inserted into it. [ABSTRACT FROM AUTHOR]

Depression is a health disorder that affects the population, regardless of their age or social status. The World Health Organization (WHO), considers it the greatest generator of incapacity worldwide. Depression increases the possibility of suicide, being the latter, the second trigger of death in people between fifteen and twenty-nine years of age. It negatively impacts different levels of the person: family, work and school and affects its ability to face daily life, aggravated preexisting medical conditions. Young or re-tired person and pregnant or postpartum women, are the groups most vulnerable to suffering from depressive disorder. In this paper, we apply two different classification techniques, namely: Bidirectional ***Encoder Representations from Transformers (BERT) and Support Vector Machines (SVM) in order to identify depression patterns in the Distress Analysis Interview Corpus DAICWOZ. We compare the models obtained and determine their robustness, using performance metrics. The results show that the approach BERT has good performance over the SVM model, reaching an accuracy of almost 90%. [ABSTRACT FROM AUTHOR]

Wideband cutting force sensing is a key technology for process monitoring. Sensorless cutting force estimation using the internal servo information of a machine tool with ball-screw-driven stages has been studied owing to its high maintainability and ease of introduction. In the motor current-based method, the cutting force estimation along the stationary axis is challenging, and the estimation bandwidth is significantly limited owing to the low sensitivity of the motor current in the high-frequency range. The dual-inertia model-based load-side disturbance observer (LDOB) can estimate the cutting force along the stationary axis using the relative position obtained from the rotary and linear encoder. The linear encoder is installed relatively near the cutting point and has a high sensitivity in the high-frequency range. However, this approach is not applicable to machine tools with complicated structural dynamics. To address this challenge, we propose a cutting force estimation method along the stationary axis using the Kalman filter (KF) based on a multiple inertia model derived solely from the relative position signal. The dynamics, depending on the stage position of the feed drive, were modeled using linear interpolation. Through end milling tests, we confirmed that the cutting force estimation accuracy along the stationary axis of a machine tool with multiple inertia dynamics was significantly improved by the proposed method compared to the current and LDOB-based methods. Additionally, the wideband cutting force could be estimated using the proposed method for bandwidths up to 1000 Hz. [ABSTRACT FROM AUTHOR]

In this paper, a new design of an all-optical 4 × 2 encoder based on 2D photonic crystal (PC) structures with a very good CR has been presented. Simulations and optimizations were performed to improve the performance of the structure. To achieve the desired results, Nano-resonators (NRs) were used to reduce noise distribution and losses. The symmetrical structure caused the encoder to have the same performance in various conditions. Due to the use of NRs and the symmetrical structure, the size of the structure was about 136.4 µm2. The results obtained from the structure simulation showed the contrast ratio (CR) of 14.88 dB, the response time of 2 ps, and the bit rate (BR) of 5 Tbps. The central wavelength of 1550 nm was used for the simulations. Power applied to the structure in each of the stages was equal to 1 mW/µm2. Structural parameters such as the radius of dielectric rods, the value of lattice constant, and the size of NRs were optimized and the effects of input power intensity, wavelength, and operating frequency on the results obtained from the simulations were investigated. The results indicated the proper design of the structure that can be used in optical integrated circuits. In the present paper, the plane wave expansion (PWE) method was used to extract the photonic bandgap (PBG) and the finite-difference time-domain (FDTD) method was used to obtain the results of the output spectrum. [ABSTRACT FROM AUTHOR]

The Diameter at Breast Height (DBH) (at a height of 1.3 m on the bole of a tree) has been one of the most important indicators during tree measurements in forestry resource inventory. However, the current DBH measurement cannot fully meet the requirement in recent years, due to the low portability, precision, efficiency, applicability, and stability, together with the complex operation, rudimentary software, high costs, and short range. In this study, an innovative device was developed to accurately, efficiently, and conveniently measure the tree DBH suitable for the complex tree shapes and the different diameter classes, while cost-saving in the office-field work survey. The specification of the device was as follows (size: 8.35 cm×5.80 cm×5.55 cm; weight: 230 g; resolution: 0.01 cm; linear range: 0-150 cm; battery capacity: 4 000 mAh input vatage: 3.7 V, output votuge: 5 V; micro-processor chip: STC15W4K48, 8 bits; encoder type: PD-1503-SDI, 12 bits). A Tunnel Magneto-Resistance (TMR) rotary encoder was also combined with the low-cost, small size, and light weight electro-mechanical structure, and high-resolution processing. As such, the measurement device was achieved in the electronization, digitization, portability, and integration of office and filed work for the tree DBH. A supporting system software was also developed, including the embedded software, mobile terminal application, and Web terminal application. In the process of an individual tree survey, the electro-mechanical structure of the device firstly converted the mechanical parameter of tree DBH to the magnetic signal, and then the magnetic signal was converted to an electrical signal. Secondly, the electrical signal was converted into the DBH measurement data using the processing integrated into the embedded software. Thirdly, the operation flow was better applied to measure the trees with special shapes and large diameters using multi-function key combinations. After all individual tree surveys, the DBH measurement data was transmitted by Bluetooth in the device to the Android application, and then uploaded to the database managed by the Web application. The measurement accuracy and operation efficiency of the device were verified to select the 196 standing trees with many tree species and a small sample plot of 42 standing trees in the Botanic Garden of Beijing Forestry University, China. The test results showed that the device presented a higher accuracy to measure the standing trees of different diameter classes than before. The total tree DBH measurement data from different diameter classes (weight: 1 092 g; resolution: 0.001 cm; linear range: 0-500 cm) indicated the mean absolute error (MAE) of 0.08 cm, Mean Absolute Percentage Error (MAPE) of 0.37%, Root Mean Square Error (RMSE) of 0.12 cm, and Relative Root Mean Square Error (RRMSE) of 0.54%, compared with an electronic draw-wire displacement sensor. In addition, a high measurement efficiency was achieved, where the average measurement time per person of each tree was 9.3 s from the efficiency test. The devices demonstrated nearly two times faster than the traditional diameter tape (weight: 42 g; resolution: 0.01 cm; linear range: 0-200 cm), while one time faster than the electronic draw-wire displacement sensor. Additionally, the price of the device was only 260 RMB, due to a 12 bits encoder (price: 135 RMB). In conclusion, this device behaved at a low cost and less labor consumption, fully meeting the technical requirement of accuracy for the Continuous Forest Inventory (CFI) in China. Therefore, the finding can provide broad application prospects in forestry resource inventory. [ABSTRACT FROM AUTHOR]

This work describes a 4-bit Flash ADC with low power consumption. The performance metrics of a Flash ADC depend on the kind of comparator and encoder used. Hence openloop comparator and mux-based encoder are used to obtain improved performance. Simulation results show that the simulated design consumes 0.265mW of power in 90nm CMOS technology using cadence-virtuoso software. The circuit operates with an operating frequency of 100MHz and a supply voltage of 1V. [ABSTRACT FROM AUTHOR]

• This model utilizes and combines as much information as it can. • The loss function we proposed can take good care of the area edges. • This model can find more hidden feature information or abstract information that is difficult to be found by naked eyes. • The hidden information significantly assists in identifying healthy and mildly diseased samples. As a common cause of hydronephrosis in children, ureteropelvic junction obstruction (UPJO) may lead to a series of progressive renal dysfunction. Ultrasonography is a primary screening of UPJO, yet its further examinations are laborious, time-consuming, and mostly radioactive. The deep learning based automatic diagnosis algorithms on UPJO or hydronephrosis ultrasound images are still rare and performance remains unsatisfactory owning to limitation of manually identified region of interest, small dataset and labels from single institution. To relieve the burden of children, parents, and doctors, and avoid wasting every bit information in all datasets, we hence designed a deep learning based mutual promotion model for the auto diagnosis of UPJO. This model consists of a semantic segmentation section and a classification section, they shared a mutual usage of a transformation structure by separately training the encoder and decoder and loop this circle. Thorough comparative experiments are conducted and situations are explored by ablation experiments, results shown our methods outperformed classic networks with an accuracy of 0.891 and an F1-score of 0.895. Our design can jointly utilize different supervisions and maximize the use of all the characteristics of each dataset, and automatically diagnose the severity of UPJO on the basis of ultrasound images by first segmentate then classify the images, moreover, not only is the final result excellent, but also the midway segmentation result is also very accurate and have smooth edges that are convenient for doctors to recognize with their naked eyes. All in all, our proposed method can be an important auxiliary tool for smart healthcare. [ABSTRACT FROM AUTHOR]

Human pose estimation from images is an exceedingly challenging task due to occlusion, blur, illumination, and size variations. Several research works have addressed this problem by designing novel encoders that extract discriminative features from input images. However, the design of decoders is an understudied problem as most approaches use the conventional transposed convolution for feature upsampling. In this paper, we propose a novel transformer-based pose estimation architecture, HumanPoseNet with a novel decoder design. The encoder is a transformer with dual-attention for highlighting important spatial and channel regions. The decoder contains a novel Patch Expansion module for feature upsampling which is computationally more efficient than the traditional transpose convolution. The decoder also contains an Attentional Feature Refinement Block that is integrated with the attention mechanism to extract refined features. Extensive experiments conducted on the MS COCO benchmark dataset proves that HumanPoseNet beats other state-of-the-art models, achieving an average precision (AP) score of 77.3. A qualitative analysis is conducted and the results prove that HumanPoseNet is able to predict accurate pose keypoints, even in occluded/blurred images. An ablation study is done to verify the complementary contributions of Patch Expansion and Attentional Feature Refinement Module. HumanPoseNet is easy to train as the model converges swiftly in merely 50 epochs while existing state-of-the-art methods are usually trained for 200 epochs or more. [ABSTRACT FROM AUTHOR]

• This study proposes a multi wavelength optical image feature fusion method for improving the effectiveness of subdural hematoma image reconstruction. • CT images of real patients was used in this study to establish a 3D three-layer tissue brain tissue model. • The average VRE and average AMD of hematoma reconstruction in the model which is using this method are only 0.66 % and 0.0052 mm−1. • Compared with the model which is directly using optical information as input, the method make the average VRE and average AMD of hematoma reconstruction respectively decreased by 80.0% and 27.8%. Diffuse optical tomography (DOT) under medical image guidance can be used to reconstruct images of subdural hematomas for monitoring purposes. However, when the shape of hematoma is irregular or the signal to noise ratio(SNR) of input signal is low, the accuracy and robustness of this method both decrease. In order to alleviate this situation and improve the reconstruction effect of cerebral hematoma images, a multi wavelength optical image feature fusion method was proposed in this paper. Experimental results show that the mean relative volume error (VRE) of the hematoma reconstruction model based on the multi-wavelength optical image feature fusion method is only 0.66 %, and the mean value of the average Manhattan distance (AMD) between the reconstructed absorption coefficient and the true absorption coefficient is 0.0052 mm−1. Compared with the model using single-wavelength optical image mapping feature extraction method, the multi-wavelength optical image feature fusion method reduces the average VRE of the model by 16.5 %, and the average AMD by 16.1 %. And compared with the model directly inputting single-wavelength optical information, its average VRE and average AMD decreased by 80.0 % and 27.8 %. This method has clinical application significance for the monitoring of subdural hematomas. [ABSTRACT FROM AUTHOR]

Application-Specific ICs (ASIC) are manufactured in bulk for a long time. In this paper, an approach to High-Speed Application-Independent IC (HS-AIIC) design is discussed. A resolution-selective (RS) and resolution-adaptive (RA) 8-bit Flash ADC are designed for use in various high-speed applications. With the choice of resolution, one can work with the trade-off between speed, power consumption, and resolution for a particular application. The proposed resolution selection algorithm can be implemented for any set of resolutions for a flash ADC design. Further, an adaptive block is added to make the ADC design adaptive in nature so that we do not have to select a particular resolution manually. The proposed design entrusts on saving manufacturing cost and increases the functionality of ADC on a single chip. Proposed resolution adaptive 8-bit flash ADC design dissipates 512 mW of power with an ENOB of 7.56 bits and SNDR of 46.27 dB for 1 GHz sampling clock pulse. [ABSTRACT FROM AUTHOR]

Although deep learning algorithms have achieved significant progress in a variety of domains, they require costly annotations on huge datasets. Self-supervised learning (SSL) using unlabeled data has emerged as an alternative, as it eliminates manual annotation. To do this, SSL constructs feature representations using pretext tasks that operate without manual annotation, which allows models trained in these tasks to extract useful latent representations that later improve downstream tasks such as object classification and detection. The early methods of SSL are based on auxiliary pretext tasks as a way to learn representations using pseudo-labels, or labels that were created automatically based on the dataset's attributes. Furthermore, contrastive learning has also performed well in learning representations via SSL. To succeed, it pushes positive samples closer together, and negative ones further apart, in the latent space. This paper provides a comprehensive literature review of the top-performing SSL methods using auxiliary pretext and contrastive learning techniques. It details the motivation for this research, a general pipeline of SSL, the terminologies of the field, and provides an examination of pretext tasks and self-supervised methods. It also examines how self-supervised methods compare to supervised ones, and then discusses both further considerations and ongoing challenges faced by SSL. [ABSTRACT FROM AUTHOR]

Image-to-image translation translates an image from one domain to another. The goal is to learn the translation relationship between different image domains. Compared with the translation models to be trained using paired training data, CycleGAN has the advantage of learning to translate between domains without paired input–output training examples. However, when using CycleGAN to translate images among multiple domains, the complexity of the model increases nonlinearly with the number of domains. To reduce the model complexity of CycleGAN-based translation models, we assume that there is a hidden space shared by different domains, and this space stores the common features of images. Then, we design a common encoder to learn image features in the hidden space. Based on the hidden space, we propose a translation model that scales linearly with the number of domains. To further improve the common feature representation accuracy, we introduce the adversarial component in the hidden space to learn the common features. We test the proposed models on different datasets, including painting style and season transfer datasets and achieve good results. [ABSTRACT FROM AUTHOR]

The properties of the drug may be altered by the combination, which may cause unexpected drug–drug interactions (DDIs). Prediction of DDIs provides combination strategies of drugs for systematic and effective treatment. In most of deep learning-based methods for predicting DDI, encoded information about the drugs is insufficient in some extent, which limits the performances of DDIs prediction. In this work, we propose a novel attention-mechanism-based multidimensional feature encoder for DDIs prediction, namely attention-based multidimensional feature encoder (AMDE). Specifically, in AMDE, we encode drug features from multiple dimensions, including information from both Simplified Molecular-Input Line-Entry System sequence and atomic graph of the drug. Data experiments are conducted on DDI data set selected from Drugbank, involving a total of 34 282 DDI relationships with 17 141 positive DDI samples and 17 141 negative samples. Experimental results show that our AMDE performs better than some state-of-the-art baseline methods, including Random Forest, One-Dimension Convolutional Neural Networks, DeepDrug, Long Short-Term Memory, Seq2seq, Deepconv, DeepDDI, Graph Attention Networks and Knowledge Graph Neural Networks. In practice, we select a set of 150 drugs with 3723 DDIs, which are never appeared in training, validation and test sets. AMDE performs well in DDIs prediction task, with AUROC and AUPRC 0.981 and 0.975. As well, we use Torasemide (DB00214) as an example and predict the most likely drug to interact with it. The top 15 scores all have been reported with clear interactions in literatures. [ABSTRACT FROM AUTHOR]

For any odd positive integer n, we express cyclic codes over Z 4 of length 4n in a new way. Based on the expression of each cyclic code C , we provide an efficient encoder and determine the type of C . In particular, we give an explicit representation and enumeration for all distinct self-dual cyclic codes over Z 4 of length 4n and correct a mistake in the paper "Concatenated structure of cyclic codes over Z 4 of length 4n" (Cao et al. in Appl Algebra Eng Commun Comput 10:279–302, 2016). In addition, we obtain 50 new self-dual cyclic codes over Z 4 of length 28. [ABSTRACT FROM AUTHOR]

Performing accurate predictions on photovoltaic power generation is a crucial factor in complementary power generation scheduling. How to resolve the problem of extracting hidden features and correcting abnormal data is the key factor of improving mid-term prediction accuracy. This study proposes a mid-term PV forecasting system using the De-Trend First, Attend Next strategy. The prediction system employs the detrending before attending strategy. It first adopts and corrects abnormal time series data, and then reconstructs the corrected time series data into trend and seasonal components. After data is constructed, different models are applied to trend and seasonal data for separately predictions, and such predictions then be reconstructed into a realistic prediction result. To find hidden features and seasonal trends of seasonal components, we propose a new model. This model is constructed with an encoder–decoder structure temporal convolution, and attention mechanism. This study evaluates the prediction system using data from a photovoltaic power station in Australia. The experimental results show that the proposed model has a Coefficient of Determination value of 0.992, which represents a 73% improvement in the mean squared error index compared to the baseline model. In summary, the experiment result demonstrates that the system has a good capability of providing and predicting accurate data, which plays a significant role in power grid dispatch. • Proposed a forecasting system with a "De-Trend First, Attend Next" strategy. • Independent modeling of trend and seasonal terms is effective. • The isolated forest algorithm detects photovoltaic anomalous data points. [ABSTRACT FROM AUTHOR]

A wireless instrumentation system was successfully developed and tested to overcome the constraints associated in collection of torque data of PTO-driven agricultural machinery with benefits such as no disturbance in telescopic action or vertical inclination of Cardan shaft. The system ensured no alterations in standard length of PTO driveline shaft. Further, telemetry system enabled remote acquisition of signals without inducing any risk of damaging the associated electronic components. It offered sensitivity of 1.6 mV/V per kN-m of applied torque with good linearity during static calibration. The performance was assessed based on accuracy, non-linearity, hysteresis, and non-repeatability and their values were 97.06%, 2.57%, 3.44%, and 0.26%, respectively. The data obtained for torque were validated with the torque computed from axial and radial load data of PTO ball bearing acquired simultaneously during rotary tillage. The low values of mean absolute percentage error (7.78% and 6.58%), maximum absolute variation (15.17% and 15.39%), and root mean square error (21.24 and 22.16 N-m), indicated good accuracy of the system. The experiments were designed with active tillage tools having two different numbers of blades, transmission gear ratios, engine speeds, tyre sizes, and four working depths as treatments. Results revealed that considered variables other than tire size, significantly affected the PTO torque and power. The decrease in PTO torque and power was found to be levelled off beyond velocity ratio 7.92. The developed instrumentation technique is simple, reliable and could be useful for database generation, implement design, and matching for effective utilization of tractor engine power. • Novel instrumentation technique was used to measure PTO (power take-off) torque. • It does not disturb the telescopic action or vertical inclination of Cardan shaft. • Sensitivity 1.6 mV/V per kN-m of applied torque with good linearity in calibration. • Data validated with torque computed from axial and radial load of PTO ball bearing. • The low values of MAPE, MAV, and RMSE indicated good accuracy of the system. [ABSTRACT FROM AUTHOR]

In this paper, the design and simulation of an ultra-fast 4 × 2 encoder have been optimized in a new and unique way using the wave interference technique based on two-dimensional photonic crystal structures. To create the given structure, nano-resonators (NR) were used in combination with curved 2-branch waveguides to take advantage of wave interference as well as reduce losses and light scattering in the structure to achieve the desired results. Due to its special design, the proposed structure has been symmetrical with a structural size of about 149 µm2. This symmetry caused the same results to be obtained in different states of the used encoder. The results from the structure simulation indicated a contrast ratio of 7.88 dB, a delay time of 0.21 ps, and a bit rate of 4.761 Tbit/s. All simulations were performed with a central wavelength of 1550 nm and an input power intensity of about 1 mW/µm2. To optimize the structure and evaluate the results obtained from it, the changes were made in the structural parameters such as the size of the radius of dielectric rods, the size of NRs, and the value of lattice constant, and the effects of wavelength, operating frequency, and input power intensity on the simulation results were investigated. The analysis of the results of simulations and optimizations indicated the design of a completely principled and logical structure with very suitable results in comparison with other structures proposed in this field so that it can be used appropriately in integrated optical circuits and can be considered as the basic structure for other designs. In the present research, the plane-wave expansion method was used to extract and analyze the photonic bandgap and the finite-difference time-domain method to obtain the results of the output spectrum of the proposed structure. [ABSTRACT FROM AUTHOR]

A goniometer is a device used to measure the angular range of motion of various human joints and muscle groups, such as the elbow, knee or waist. However, the majority of mechanical goniometers are heavy, provide inaccurate measurements and are affected by dust or stains. Electronic and optical fiber-based goniometers demonstrate better capability than their mechanical equivalents; however, due to their higher costs and complicated setup requirements, their usage is not viable. Optical-based encoder patterns were GC-ARE previously proposed and discussed in some studies. Such optical-based encoder patterns achieve the same angular resolution as traditional Gray code encoder patterns with many shortcomings, such as a high error rate and low tolerance, which arise as a result of the encoding method. This paper presents a design for a reliable and non-intrusive phase shift code absolute rotary encoder (PSC-ARE), which is based on an optical sensor, providing a low-cost, high-precision absolute rotary encoder system. An inexpensive optical mouse sensor is used to capture the code image and the phase shift code transforming time-domain data into frequency domain, which is used to decode the corresponding absolute angle. The PSC-ARE proposed in this study shows a performance improvement over conventional designs owing to the reduced interference from dust and other environmental particles and provides lower sensitivity to calibration errors. Furthermore, the proposed design is not only more compact and portable than the traditional goniometer, and it can also provide a longer lifespan as no stepper motor is included. Finally, the proposed design demonstrates high tolerance to "unintentional defocus" effects and noise effects. [ABSTRACT FROM AUTHOR]

Different logic circuits based on memristors have been extensively investigated. However, most of these circuits require accurate initialization. A self-learning logic circuit based on mermristors that can achieve various logic gates without initialization is proposed in this paper. Three functional blocks, including a sum block, a learning block, and a compare block, are elaborately designed in the proposed logic circuit. Programmable switches in the sum and compare blocks enable the circuit to perform various logic gates, such as Boolean, IMPLY, and random logical combinations. In these various logical operations, the learning block can automatically obtain different memristance states. The aforementioned logic operations can easily be extended to multi-fan-in logic and logical cascade operations. Circuit designs of an encoder and decoder are considered as application examples. Finally, PSpice simulation results of the logic circuits and extended applications are provided. Simulation results indicate that the proposed circuit can effectively perform different logic operations and exhibits excellent robustness to circuit device variations. [ABSTRACT FROM AUTHOR]

The transistorless technology, quantum dot cellular automata (QCA) becomes one of the potential candidates of nanotechnology for signal processing circuits because of its ultra‐low power requirement, high frequency of operation, extreme‐low area requirement, and short channel effect (SCEs) immune at nanoscale circuits. Quasi‐adiabatic switching of QCA makes its data transfer faster. This work reports faster, smaller, and low‐power, 4‐to‐2 and 8‐to‐3 encoder, and priority encoder circuits designed on QCA platform. Both of these encoders and priority encoders are widely used circuits for communication to encode the data. These circuits are realized using three‐input majority gates, five‐input majority gates, and co‐planar clock zone‐based crossover technique. The proposed circuits show the improvement in performance in terms of cell count, area requirement, delay, quantum cost, and energy dissipation as compared to the existing circuits. The cell count and the area requirement of the proposed circuits have been magnificently lowered as compared to the recent literature. Moreover, these circuits also require 75% lesser clock periods than the existing literature, thus demonstrate lower latency. This work shows the way for future exploration of QCA‐based encoder circuits. [ABSTRACT FROM AUTHOR]

An efficient parallel HDMI 2. 0 encoder is designed to solve the problems of poor scalability and large circuit area in the design of multi - channel video display system. The compatibility of HDMI 2. 0 encoder has been improved through the compatibility design of HDMI version 1. 4 and 2. 0 standards. The scalability and portability of HDMI 2. 0 encoder are improved by simplifying functions and providing user - friendly data interfaces. Through parallel design, the ECC circuit and the scramble circuit are kept al the same data rate as the entire encoder, and a pipelined design method was used lo implement the parallel encoding of the video data encoding circuit. The result analysis, function simulation and on - board test are carried out and their results show that the HDMI 2. 0 encoder conforms to the HDMI standard and can stably perform encoding and output. The results also showed that the coding ability of ultra - HD video is possessed by HDMI 2. 0 encoder designed in parallel, which can reach the maximum coding bandwidth of HDMI 2. 0 standard. [ABSTRACT FROM AUTHOR]

Our Attentive Maps Encoder Network (AMENet) predicts the future trajectories by observing the past ones. It has Encoders for encoding the observed and the future information, and a Decoder for predicting trajectories by sampling from the latent space and conditioning on the encoded information. Trajectory prediction is critical for applications of planning safe future movements and remains challenging even for the next few seconds in urban mixed traffic. How an agent moves is affected by the various behaviors of its neighboring agents in different environments. To predict movements, we propose an end-to-end generative model named Attentive Maps Encoder Network (AMENet) that encodes the agent's motion and interaction information for accurate and realistic multi-path trajectory prediction. A conditional variational auto-encoder module is trained to learn the latent space of possible future paths based on attentive dynamic maps for interaction modeling and then is used to predict multiple plausible future trajectories conditioned on the observed past trajectories. The efficacy of AMENet is validated using two public trajectory prediction benchmarks Trajnet and InD. [ABSTRACT FROM AUTHOR]

The paper proposes a design for an acoustic 4 × 2 encoder that incorporates acoustic waveguides within a two-dimensional square lattice phononic crystal structure. The encoder's structure comprises of a nylon matrix embedded with a 33 × 30 molybdenum rod array, featuring four input ports and two output ports. The simulation results demonstrate that the proposed structure can effectively achieve the acoustic 4 × 2 encoder functionality. To evaluate the performance of the acoustic 4 × 2 encoder, we analyzed the transmission spectrum using the finite element method (FEM). [ABSTRACT FROM AUTHOR]

Experimental measurement of transmission error and vibration of a gear pair with crown modification are developed. With the help of high-precision optical encoder, effects of gear misalignment on unloaded and lightly loaded dynamic transmission error, which are relative to gear rattle, are investigated. The gear mesh misalignment is introduced by eccentric sleeve assembled on the output shaft. Effects of modification and misalignment on the dynamic transmission error, are studied at different load and driving velocity conditions. The experimental results show that, with the increase of the crown amplitude, the peak-to-peak values of dynamic transmission error are decreasing dramatically. Impact deformation or elastic deformation is a very important part of the dynamic transmission error although they are unloaded or lightly loaded. The components in harmonics of meshing frequency will change distinctly comparing cases at low input shaft velocity without and with misalignment, but different phenomena are detected while increasing the input shaft velocity. Finally, the relation between transmission error and gear box vibration is illustrated, and spectrum kurtosis is introduced to reveal gear rattle. [ABSTRACT FROM AUTHOR]

Semantic segmentation is one of the important technologies in autonomous driving, and ensuring its real‐time and high performance is of utmost importance for the safety of pedestrians and passengers. To improve its performance using deep neural networks that operate in real‐time, the authors propose a simple and efficient method called ADFNet using accumulated decoder features, ADFNet operates by only using the decoder information without skip connections between the encoder and decoder. They demonstrate that the performance of ADFNet is superior to that of the state‐of‐the‐art methods, including that of the baseline network on the cityscapes dataset. Further, they analyse the results obtained via ADFNet using class activation maps and RGB representations for image segmentation results. [ABSTRACT FROM AUTHOR]

A novel rhodamine‐phenalenedicarbonitrile based "on‐off" sensor (Rho‐Pdc) has been synthesized which is characterized by 1H NMR, 13C NMR and mass spectrometry. The photophysical studies of the dyad suggest its existence in spirolactam ring open form of rhodamine. ICT (Intramolecular Charge Transfer) active sensor displays an absorption band at 577 nm with a shoulder at 520 nm and fluorescence band at 593 nm. The treatment of dyad Rho‐Pdc with different anions shows its sensitivity and selectivity towards CN− ions in aqueous methanolic solution with lowest detection limit of 0.13 μM and the binding constant was determined to be 1.17 × 104 M−1 by applying Benesi‐Hildebrand equation. The presence of CN− ions hinder the ICT process of dyad Rho‐Pdc and shows a blue shift in absorption band of dyad Rho‐Pdc with formation of new band at 488 nm as well as decrease in the intensity of emission band. Rho‐Pdc has also been used for the colorimetric detection of CN− ions among different anions and concentrations. The reversibility of Rho‐Pdc.CN−complex has been achieved with Fe3+ ions which formed the basis for the construction of multiple molecular logic circuits. [ABSTRACT FROM AUTHOR]

The use of neural machine algorithms to translate English is a hot topic in current research. The ability to capture long-distance information in English translation using traditional sequence neural frameworks is too poor and has its own limitations. The current improved frameworks, such as recurrent neural network translation effect is not ideal. Aiming at the shortcomings of traditional machine translation algorithms, this paper establishes an attention coding and decoding model, combines the attention mechanism with a neural network framework, and uses TensorFlow to implement the entire English translation system. This method can improve the accuracy of translation. Experiments show that the BLUE value of the algorithm model constructed in the article has different degrees of improvement compared with traditional machine learning algorithms, which further proves the effectiveness of the improved attention mechanism model proposed in this article in English translation. [ABSTRACT FROM AUTHOR]

Cryptography is one of the paramount and most vital data treatment processes, it allows us to be secure in our electronic transactions. The process of cryptography protects our valuable data such as private account numbers and transaction amounts, electronic signatures replace handwritten signatures or credit card authorizations, and public-key encryption provides confidentiality. The objective of data encryption is to keep digital data confidentiality save as it is stored on computer systems and transferred using the internet or other computer networks. In this paper we will focus in enhancing security level of the encryption-decryption process by introducing a novel method, which uses any digital color image to encode-decode secret message, the using of a special key to encrypt-decrypt the encoded-decoded message, the color image will be known only by the transmitter and receiver to keep the process of data treatment confidential, the obtained experimental results by the proposed method will be analyzed to prove the enhancement in process efficiency and confidentiality. [ABSTRACT FROM AUTHOR]

At present, most intrusion detection algorithm cannot achieve a good balance between intrusion detection rate and false positive rate, in order to effectively avoid such problems, this paper proposed an intrusion detection method based on ADEN. The method first initialized the parameters of the encoder part in the ADEN model by training the deep belief network, and initialized the parameters of the decoder part by using the normal distribution. Then it tuned the parameters of the asymmetric deep belief network model by calculating the reconstruction error, so that the model can obtain the optimal low-dimensional representation of the original data. Finally, it used the data obtained by the encoder as input data of the classifier and detected. The ADEN model can extract features that are more conducive to classification and save more test time in the model initialization phase. The experimental results show that the method can achieve better detection performance and achieve better detection accuracy for small categories of samples. [ABSTRACT FROM AUTHOR]

In this study, the authors will show that coding the positions of the symbols, instead of their values, can be a good way to implement efficient Slepian–Wolf (SW) coding and can reduce the complexity of both the encoder and the decoder. The authors will also propose a practical distributed video coding (DVC) system that exploits this idea. This system will use binary maps to indicate the positions of the most probable symbols, instead of separating them into bitplanes. Simulations show that this position‐based SW coding allows a simple and more efficient DVC system with improved rate‐distortion performance, compared to the bitplane‐based DVC system that uses the same side information. The memory requirements at the encoder are reduced by about 50% and the number of channel coding operations is also reduced. This DVC system also allows to use an easy way to control quantisation, reduces the decoding latency, and allows fast parallel decoding. [ABSTRACT FROM AUTHOR]

In this paper, we propose a Point Encoder GAN for 3D point cloud inpainting. Different from other 3D object inpainting networks, our network can process point cloud data directly without any labeling and assumption. We use a max-pooling layer to solve the unordered of point cloud during the learning procedure. We add two T -Nets (from PointNet) to the encoderdecoder pipeline, which can yield better feature representation of the input point cloud and a more suitable rotation angle of the output point cloud. We then propose a hybrid reconstruction loss function to measure the difference between the two sets of unordered data. Using small sample models on ModelNet40 only, the proposed Point Encoder GAN yields end-to-end inpainting results surprisingly. Experiment results have shown a high success rate. Several technical measures are used to identify the good qualities of our generated models. [ABSTRACT FROM AUTHOR]

Nonvolatile memory (NVM) technologies are promising candidates as the next-generation main memory due to high scalability and low energy consumption. However, the performance bottlenecks, such as high write latency and low cell endurance, still exist in NVMs. To address these problems, frequent pattern compression schemes have been widely used, which however suffer from the lack of flexibility and adaptability. In order to overcome these shortcomings, we propose a well-adaptive NVM write scheme, called dynamic frequent pattern compression (DFPC), to significantly reduce the amount of write units and extend the lifetime. Instead of only using static frequent patterns in existing FPC schemes, which are predefined and not always efficient for all applications, the idea behind DFPC is to exploit the characteristics of data distribution in execution to obtain dynamic patterns, which often appear in the real-world applications. To further improve the compression ratio, we exploit the value locality in a cache line to extend the granularity of dynamic patterns. Hence, DFPC can encode the contents of cache lines with more kinds of frequent data patterns. Moreover, to further support efficient write and read operations in the context of multilevel/triple-level cell NVMs, we need to extend the DFPC to improve performance in terms of the access latency and energy consumption. We hence propose a latency-optimized and energy-efficient compression write scheme to encode the compressed data with low energy and latency states, i.e., enhanced DFPC, thus reducing the latency and energy consumption. We implement DFPC in GEM5 with NVMain and execute the applications from SPEC CPU2006 to evaluate our scheme. Experimental results demonstrate the efficacy and efficiency of DFPC. We have released the source codes for public use at Github https://github.com/dfpcscheme/DFPCScheme. [ABSTRACT FROM AUTHOR]

This study addresses the asymptotic tracking problem subjected to linear quadratic (LQ) constraints for linear discrete‐time systems, where packet dropout occurs in actuating channels. To solve this objective control problem, the controller‐coding co‐design approach is adopted, i.e. the controller, encoder and decoder are designed for taking full advantage of the network resource collaboratively, thereby achieving better transmission of control signals. A stabilisability condition in the mean square sense that reveals the fundamental limitation among the H2 norm of the plant, data arrival rates and coding matrices is first derived. Then, a solvability condition is conducted to handle the additional stochastic LQ control objective by a modified discrete‐time algebraic Riccati equation, and an iterative algorithm is also given for designing the corresponding state feedback gain and coding matrices. Relied on such design, the asymptotic tracking constraint is further fulfilled through solving a Sylvester equation, and the feedforward gain related to tracking is parameterised. Finally, a simulation with the implementation of the design method on two cooperative robots is included to show the effectiveness of the current results. [ABSTRACT FROM AUTHOR]

Enabling deep neural networks for tight resource constraint environments like mobile phones and cameras is the current need. The existing availability in the form of optimized architectures like Squeeze Net, MobileNet etc., are devised to serve the purpose by utilizing the parameter friendly operations and architectures, such as point-wise convolution, bottleneck layer etc. This work focuses on optimizing the number of floating point operations involved in inference through an already compressed deep learning architecture. The optimization is performed by utilizing the advantage of residual connections in a macroscopic way. This paper proposes novel connection on top of the deep learning architecture whose idea is to locate the blocks of a pretrained network which have relatively lesser knowledge quotient and then bypassing those blocks by an intelligent skip connection, named here as Shunt connection. The proposed method helps in replacing the high computational blocks by computation friendly shunt connection. In a given architecture, up to two vulnerable locations are selected where 6 contiguous blocks are selected and skipped at the first location and 2 contiguous blocks are selected and skipped at the second location, leveraging 2 shunt connections. The proposed connection is used over state-of-the-art MobileNet-V2 architecture and manifests two cases, which lead from 33.5% reduction in flops (one connection) up to 43.6% reduction in flops (two connections) with minimal impact on accuracy. [ABSTRACT FROM AUTHOR]

A simplification of the mechanism of an absolute multi-turn gear angular position encoder is proposed. Movement on the input measuring shaft is transmitted in parallel to several driven shafts and single-turn sensors located on the shafts. In this case the reading of the encoder may be calculated using methods of modular arithmetic from the readings of single-turn sensors based on the values of the parameters that determine the gear ratios of the encoder mechanism. [ABSTRACT FROM AUTHOR]