Search

With ever more complicated functionalities being integrated in modern autonomous systems, traditional design methods may not remain sufficient to deliver trusted and high-performance systems with stringent temporal, safety and cost efficiency requirements. In this paper, we discuss the limitations of the traditional design methods with the above requirements enforced, in which hardware and software design are often considered separately. To tackle these limitations, this paper presents a novel design solution that synthesizes both software-level and hardware-level design. First, we highlight and analyze the interconnections between software-level methods (e.g. priority assignment and task allocation) and hardware design (e.g. cache and memory management), in terms of the resulting system performance, e.g. latency. Second, by applying the identified interconnections, we propose an optimization framework to produce high-quality synthesized solutions of both software and hardware design based on a set of candidate design methods. In addition, we describe potential research directions derived from the work and major challenges that can be investigated jointly by engineers and researchers from embedded systems, system safety and programming languages communities.

Further study of the active RX-TX microstrip patch antenna is discussed in the paper. In previous studies it was used field effect transistor with positive feedback, which implied the regenerative resonant amplification. The presence of finite size of the patch resulted in reducing of amplification with frequency decreasing! The problem of such behavior is discussed in paper. In this paper it is proposed to concentrate the regenerative amplification in a single point, which will result in stable receiving/transmitting signal amplification without any degradation with operating frequency changing. The single point amplification can be realized with the use of two-pole devices such as tunnel diode, Gunn diode etc. Small sizes of diodes let us assume the any additional transmitting microwave lines are absent and the reasons of any feedback level changing are absent as well. The way of bias supplying of two-pole devices as well the microwave signal inputting/ outputting are widely discussed in paper. It is shown the active antenna with two-pole devices can operate in receiving and transmitting modes. No switching between receiving and transmitting modes does need to be implemented. The active antenna can operate in both modes simultaneously.

This paper analyzes the behavior of transient switching silicon carbide devices influenced by the parasitics of printed circuit board and the component itself. The parasitics in power and gate driver loops are described and investigated. Double pulse test simulations for silicon carbide devices are conducted in this report and verified through the circuit simulation. The parasitics in the printed circuit board are added based on the real experience printed circuit board design and the parasitics in silicon carbide devices are added based on the manufacturer modelling. The parasitic inductances in the simulation are varied +/- 30% to understand the effect of the parasitics. The drain and source parasitic inductances and the gate parasitic resistance are simulated separately to understand the individual effect on the ringing of silicon carbide devices. The ringing impact on the reliability of devices is also analyzed in this paper.

In the era of information society and 5G networks, cars are extremely important mobile information carriers. In order to meet the needs of multi-scenario business requirements such as vehicle assisted driving and in-vehicle entertainment, cars need to interact with the outside world. This interconnection and data transmission process is usually called vehicular communication (V2X, Vehicle-to-Everything). Device-to-device (D2D) communication not only has partial nature of communication, but also alleviate the current problem of spectrum scarcity of resources. The application of D2D communication in V2X can meet the requirements of high reliability and low latency, but resource reuse also brings interference. Software-defined networking (SDN) provides an optimal solution for interoperability and flexibility between the V2X and D2D communication. This paper reviews the integration of D2D and V2X communication from the perspective of SDN. The state-of-the-art and architectures of D2D-V2X were discussed. The similarity, characteristics, routing control, location management, patch scheduling and recovery is described. The integrated architecture reviewed in this paper can solve the problems of routing management, interference management and mobile management. It also overcome the disconnection problem between the D2D-V2X in terms of SDN and provides some effective solutions. - Qatar National Research Fund (QNRF) - [UREP27-020-1-003]. - Ministry of Higher Education, Malaysia (MOHE) - [FRGS/1/2018/ICT02/UKM/02/6]. - National Research Foundation of Korea (NRF) - [2019R1C1C1007277]. - Taif University (TU)- [TURSP-2020/260].

End users have become active participants in local electricity market transactions because of the growth of the smart grid concept and energy storage systems (ESS). This participation is optimized in this article using a stochastic two-stage model considering the day-ahead and real-time electricity market data. This model optimally schedules the operation of a Smart Home (SH) to meet its energy demand. In addition, the uncertainty of wind and photovoltaic (PV) generation is considered along with different appliances. In this paper, the participation of an EV (electric vehicle), together with the battery energy storage systems, which allow for the increase in bidirectional energy transactions are considered. Demand Response (DR) programs are also incorporated which consider market prices in real-time and impact the scheduling process. A comparative analysis of the performance of a smart home participating in the electricity market is carried out to determine an optimal DR schedule for the smart homeowner. The results show that the SH’s participation in the real-time pricing, scheme not only reduces the operating costs but also leads to better than expected profits. Moreover, total, day-ahead, and real-time expected profits are better in comparison with existing literature. The objective of this paper is to analyze the SH performance within the electrical market context so as to increase the system’s flexibility whilst optimizing DR schedules that can mitigate the variability of end-users generation and load demand.

Since its inception in 2015, Style Transfer has focused on texturing a content image using an art exemplar. Recently, the geometric changes that artists make have been acknowledged as an important component of style[42], [55], [62], [63]. Our contribution is to propose a neural network that, uniquely, learns a mapping from a 4D array of inter-feature distances to a non-parametric 2D warp field. The system is generic in not being limited by semantic class, a single learned model will suffice; all examples in this paper are output from one model.Our approach combines the benefits of the high speed of Liu et al. [42] with the non-parametric warping of Kim et al. [55]. Furthermore, our system extends the normal NST paradigm: although it can be used with a single exemplar, we also allow two style exemplars: one for texture and another geometry. This supports far greater flexibility in use cases than single exemplars can provide.

Crowdsourcing plays an important role in modern IT landscape, enabling the use of human information processing abilities to solve problems that are still hard for machines. One of the specific (and most demanded) applications of crowdsourcing is collecting item annotations, i.e., describing the contents of complex items with a help of labels (tags). Input received from crowdsourcing participants is typically unreliable, therefore, to increase the quality of annotations, each item is typically processed by several participants and the obtained annotations have to be aggregated. The paper considers a special case of annotating, where a set of possible labels, as well as the set of relationships between the labeled items and the labels are defined by an OWL 2 ontology (OWL QL). Such semantic item annotations turn out to be very useful in organizing large collections of items and enabling semantic search in them. In order to increase annotations quality, the paper proposes two aggregation methods - Onto Voting and OntoSB, differing in that the first one is agnostic with respect to participants' reliability and the second one accounts for variations in reliability. Simulation experiments with ontology-based annotations of varying quality show that the proposed aggregation methods increase the quality of collected ontology-based item annotations.

Automotive radar sensors offer a promising and effective modality for perception and assessment of the surrounding environment. A key element of environment sensing in automotive radars is the reliable detection, classification and tracking of vulnerable road users such as pedestrians and cyclists. In this paper, we propose an integrated Bayesian framework dubbed BayesRadar, which improves the overall classification accuracy by tracking the embedding vector of a neural network and its prediction uncertainty via recursive Kalman filtering over time. Apart from the classification accuracy of a model, a critical measure includes the analysis of statistical confidence over the target class score. Such measures for predicting the true correctness likelihood of the classification estimates are essential in safety-critical automotive applications. Therefore, in this paper, we present and evaluate the classification, embedding cluster score and statistical confidence performance of the proposed framework in the context of classifying vulnerable road users compared to state-of-art deep learning approaches. Furthermore, we demonstrate superior performance of the BayesRadar for unseen classes compared to long short-term memory based temporal tracking of the embedding vectors.

Electricity network modelling and grid simulations form a key enabling element for the integration of newer and cleaner technologies such as renewable energy generation and electric vehicles into the existing grid and energy system infrastructure. This paper reviews the models of the African electricity systems and highlights the gaps in the open model landscape. Using PyPSA (an open Power System Analysis package), the paper outlines the pathway to a fully open model and data to increase the transparency in the African electricity system planning. Optimisation and modelling can reveal viable pathways to a sustainable energy system, aiding strategic planning for upgrades and policy-making for accelerated integration of renewable energy generation and smart grid technologies such as battery storage in Africa.

Machine learning is presently acknowledged as a significant ingredient of research in many fields, including robotics. The use of robots to perform assorted tasks is evident in difficult, uncompromising, and hazardous spaces and sectors such as manufacturing, transportation, healthcare, landmines, mining, patrolling, disaster relief etc. For a robot to carry out its assigned task, it normally has to navigate safely without collisions to different locations, which also means understanding its working environment, collectively known as the robot navigation problem. This paper considers finding a solution using neural networks to the robot navigation problem, particularly the path planning problem that includes fixed obstacles. The objective of the path planning problem is to find a route to the final destination that is optimal and also collision-free. Different training algorithms and network structures are used to construct models that can predict a turning angle for the point-mass robot which will be used to avoid obstacles in the robot's path to the destination. This paper will present a comparative analysis of the performance of different feedforward neural network models. The results suggest that the feedforward neural network model with 10 neurons and Bayesian regularization performed the best. The model has been used to avoid obstacles in two different environments. The trajectories show that the robot has safely avoided obstacles in its path and reached the destination.

Sign languages is a critical requirement that helps deaf people to express their needs, feelings and emotions using a variety of hand gestures throughout their daily life. This language had evolved in parallel with spoken languages, however, it do not resemble its counterparts in the same way. Moreover, it is as complex as any other spoken language, as each sign language embodies hundreds of signs, that differs from the next by slight changes in hand shape, position, motion direction, face and body parts contributing to each sign. Unfortunately, sign languages are not globally standardized, where the language differs between countries and has its own vocabulary and varies although they might look similar. Furthermore, publicly available datasets are limited in quality and most of the available translation services are expensive, due to the required skilled human personnel. This paper proposes a deep learning approach for sign language detection that is finely tailored for the Egyptian sign language (special case of the generic sign language). The model is built to harnesses the power of convolutional and recurrent networks by integrating them together to better recognize the sign language spatio-temporal data-feed. In addition, the paper proposes the first Egyptian sign language dataset for emotion words and pronouns. The experimental results demonstrated the proposed approach promising results on the introduced dataset using combined CNN with RNN models.

Great Britain has reached high penetrations of intermittent Distributed Generation (DG), which has influenced many aspects of power system planning and operation. There has been a lack of technical requirements for DG to provide grid support and for network operators to monitor, control and gather detailed information of DG installations. The work reported in this paper has explored available data sources to provide better visibility of the amount, size and type of DG installed at various locations in the distribution networks and evaluated it alongside whole electricity system trends. Several of the issues associated with a high volume of generation being unobservable and uncontrollable are surveyed and discussed, such as active network management and coordination of transmission and distribution (T & D) networks. Finally, the paper discusses system resilience, emphasising that DG loss-of-mains protection continues to present a threat to system security and DG adds uncertainty to the effectiveness of automatic defence mechanisms. The latter issues were brought into focus following a system disturbance in GB on August 9th, 2019 and are reflected on in this study.

As computers continue to play a more critical role in everyday life, it becomes ever more important to secure the devices to which we entrust our personal and private data. This paper serves to detail the security issues present in out-of-date machinery in both the organizational and personal settings. This paper tests a Windows 10 machine running version 1511, and its ability to withstand a targeted virus attack. First, we examine the tools used to create a Reverse TCP virus. Second, we discuss how to generate the virus using those tools and deliver the payload to our Windows machine. Finally, we showcase what can be done to the targeted machine by creating our own user in the database and accessing the Windows machine remotely from our Kali Linux machine. This study has implications for researchers and practitioners by illustrating the destructive capabilities of a simple virus and what can be done to secure a network against such threats.

Distributed algorithms are now present in many computing activities. The price decline of computer hardware, increase in speed of communication, the versatility of the Internet, and the advancement of network technology all have contributed immensely to the incredible growth and importance of distributed computing. Therefore, even an undergraduate is now expected to know at least the basics of distributed algorithms. This paper describes the experience and performance of students who are introduced to distributed algorithms in an undergraduate Algorithms course. The paper shows that for teaching new concepts, providing in-class examples should be emphasized more for effective student learning. It is more important compared to prior knowledge of similar concepts.

This paper briefly discusses the modern music industry setting and blockchain technology and how blockchain can be used to create a better music industry for the benefit of artist. The music industry is big and complex which generated a lot of problems for both the artist side and the industry as corporations and also put the artists and music labels and management at odds which has participated in a making the music industry a very bad environment and full of mistrust. Blockchain is classified by Gartner as one of the top ten strategic technologies in 2019, blockchain is based in distributed set or records linked together forming a chain, these records are held by different entities in a peer-to-peer network these entities work together to secure the information in the chain. by analyzing how the industry works and also present some use cases of the use of blockchain within the music industry. the paper gives a proposal to a new royalty payments system based on the environment of the blockchain technology. The paper also gives a brief about the challenges and road blocks that may face the implementation of the technology within the music industry.

In this paper, it is aimed to examine the behavior of Thyristor Controlled Braking Resistor (TCBR), which is one of the (Flexible Alternating Current Transmission System) FACTS types, in wind energy conversion systems containing synchronous generator. TCBRs often involve the use of shunt coupled braking resistors to correct the temporary imbalance between the mechanical power input and the electrical power output of the generators. In principle, small frequency oscillations can be damped by applying a controllable resistor insertion method to the power system. Within the scope of this paper, the performance of the system is evaluated by using TCBR in a short circuit fault in a three-phase wind energy based power system. According to the results obtained from the experimental studies, the excessive oscillations in the speed of the generator together with the failure reached their nominal values in a short time with the activation of the TCBR.

© 2021 IEEE.Coronavirus usually transmits from the animal to the human, but now, the virus transmission is between persons. Therefore, scientists and researchers are trying to develop several types of machine learning methods to defend against COVID-19. Medical images play a significant role in this time due to they can be used to recognize COVID-19 accurately. However, in this paper, we used X-Ray images, the images undergone to sharpening techniques to increase the results further. The texture techniques named local binary pattern (LBP) have been used in order to extract features. The features obtained were applied to the support vector machine (SVM). The results we achieved were 100% for all performance measurements. Our results were conspicuously superior compared to the state-of-the-art papers published.

Sensor fusion is a process that indirectly finds its way into various aspects of engineering without being a relevant observatory mark. This same feature of sensor fusion also applies to the GPS navigation sector. The pivotal aim of this research paper is to analyze and demonstrate various ingenious advantages of incorporating sensor fusion with various existing systems regardless of their respective field of operations. Further, paper focuses on the importance of giving more priority to the algorithms and software that ensure smooth working of the sensors rather than heavily investing on more expensive hardware resources. Algorithms like the Kalman filter play a major role in the field of sensor fusion and can be easily integrated with the existing algorithm’s used in modern day GPS navigation systems. Mainly, the explanations provided in this paper are revolving around the aircraft industry with important focus given to Military-based Aircrafts. Finally, the paper includes various simulations that focuses on replicating the movement of aircrafts in real-world scenarios and study various advantages and side effects of using a system that is the resultant of the fusion between an IMU and GPS sensor thereby overcoming the limitations of the existing GPS-based Navigation.

Renewable energy sources are the sources of energy collected from renewable sources, which do not deplete or will not be replenished in the lifetime of humans. The conventional sources of energy like coal, oil and gas are not only moving towards their depletion but are also becoming a great concern for environmentalists as it is causing adverse consequences on the environment. Therefore, the mission to create clean and renewable energy sources is suitable and essential. Examining at the national level, there are approximately 30 countries all over the world which are making use of renewable energy sources. Its use of renewable energy is contributing 20% of the energy supply. Solar energy is the energy that is harnessed by making use of ever changing technologies such as solar heating.This paper is going to discuss the off-grid Solar System for the house. Other than this, the paper also discussed the scope of renewable sources of energy in Northern Cyprus. In North Cyprus has a huge potential for solar energy and the market opportunities for the investors are higher for building grid connected PV projects in comparison to the wind farm projects. Moreover, the paper also talks about different types of PV systems while also discussing the design of the off grid PV system for House in northern Cyprus. One of the most important topics that this paper also discusses is the cost performance analysis of the off grid solar system for house and the cost. This is also useful for future research in the similar area of work.

This paper presents a smart software solution, designed to enhance the medical system of Romanian hospitals, by introducing technologies within the medical system, that will bring significant advantages to hospitals, the medical staff and patients, by automizing the existent, consistent of mostly manually performed processes. Compare to the current systems used in the Romanian hospitals, the solution presented in this paper revolves around a web platform, using a NoSQL MongoDB database, as well as a mobile application, that can be used to manage the medical records of a hospital and preview records from a mobile device, based on the readings performed on a Near Field Communication (NFC) chip and the placement of beacons in hospital rooms. This paper aims to emphasize the benefits brought by technology in the medical field, from multiple parties involved: the medical staff (hospital administrators, doctors, nurses), possible adjacent parties (pharmacies, suppliers) and patients.

This paper presents the characterization of an antenna array containing two radiant elements, each element having three quarters of wavelength and posted on the same ground plane. Over a metallic and wide enough mesh ground plane, the three-quarter wavelength element outperforms the widely used quarter wavelength antenna configuration at medium elevation angles and at low elevations has equal, including a better gain. The antenna array has been designed and developed for the satellite frequency band between 420 and 440 MHz, allocated to multiple applications. The comparison between the two configurations, the results of the simulations and experimental measurements are presented in this paper emphasizing the advantages of the developed array.

The exchange of information between vehicles, which is based on the reliability of wireless communication systems, has an important role in reducing traffic accidents and improving traffic flow. One of the most-known functionalities based on vehicle-to-vehicle (V2V) communication is the cooperative adaptive cruise control (CACC) system, in which the vehicles travel in a platoon and communicate with each other at the same time. To obtain an improved stability of the traffic flow using CACC systems, in this paper, the adjustment of the distances between vehicles is achieved with the help of a distributed generalized predictive control (DGPC) strategy. Therefore, this paper proposes an analysis for the above-mentioned architecture using the TrueTime toolbox, which can simulate wireless communication networks and network nodes, providing a complete perspective of testing the control architecture.

The advancements that have taken place in the field of smart meters, such as the Advanced Metering Infrastructure (AMI), and the Internet of Things (IoT), have made it possible for smart cities to be equipped with a variety of electronic devices. This way, smart cities have become more accessible because of the availability of technologies and equipment, with numerous elements becoming applicable. The goal of the present paper is to offer a comprehensive overview of the smart city concept, together with usage possibilities, advantages, and opportunities. Furthermore, the paper presents some of the IoT technologies with their combination possibilities and applications throughout various components of the smart city. Another aspect covered by the paper is the possibility of smart cities applications in terms of future technological advancement. An in-depth analysis of practical implementations around the world and their main drawbacks is presented.

The permanent intensification of information required in embedded systems leads to migration from single core to multicore systems which offer increased performance, allowing different software and hardware architectures based on heterogeneous or homogeneous cores. In the automotive field, the main AUTOSAR (Automotive Open System Architecture) component in such multicore architectures is represented by an adequate real-time operating system which coordinates the cores and their tasks according to different strategies. This paper briefly analyzes the features of the AUTOSAR multicore operating system. Afterwards, the paper proposes the extension of the operating system with adaptive capabilities which allow each core to manage besides its own tasks, the critical tasks of other cores only if required. To achieve this, the replication strategy is considered for task migration between cores. The simulation results illustrate the expected system behavior considering the proposed task migration strategy. The presented concept wishes to enhance the safety perspective of the automotive software.

This paper presents a software solution for monitoring people in quarantine during a pandemic. The development of the software solution presented in this paper aims to streamline the measure of isolation at home or quarantine centers. Persons affected by this measure must not leave the quarantine area in any form within the time limit set by the authorities. This software solution proposes a solution by which these people can be monitored by requesting their current location at a specified time interval. This solution assists the competent authorities by eliminating the need to verify quarantined persons by authorized personnel. Thus, those resources can be used for other pandemic prevention purposes.

In this paper three different types of architectures have been trained and tested to determine which one of them is more efficient in detecting exudates from retinal images. The database that was used is SUSTech_SYSU from which 100 images were extracted for image testing and another 100 images that were used in obtaining the training dataset. The three architectures that were studied in this paper are SqueezeNet, ResNet50 and GoogLeNet. The results showed that from the three architectures the most efficient in exudates detection was ResNet50 with an average accuracy of 0.928, followed by SqueezeNet with 0.910 and finally, GoogLeNet for which the average accuracy was 0.907.

The software world is in the midst of a computing paradigm shift, transitioning on-premise software to the efficiency and almost endless scalability offered by cloud solutions. While some simulation models are by design better prepared than others for this transition, relatively little progress has been done on the architectural side of software to fully harness the new possibilities offered by cloud computing. This paper aims to review the various simulation models with emphasis on agent-based ones, highlighting their advantages and present a short review of state-of-art software implementing those models. Finally, this paper will focus on an innovative way of applying an agent-based model to a variety of emergency simulations, while showcasing a lightweight and flexible software architecture capable of distributing workload in a cloud-based environment.

This paper deals with the issue of designing a fault-tolerant routing method for two-dimensional (2D) torus-connected Network-on-Chips (NoCs). 2D torus network contains several rings inside the network; therefore, it is not an easy task to design a fault-tolerant routing method that guarantees deadlock-freeness and 100% packet reachability. In this paper, we focus on the network on which packet routing is performed, and propose a method for realizing fault-tolerant routing with a well-known dimension order routing method. We extend the network architecture to make capable of utilizing faulty nodes as bus components. Performance evaluation shows that the proposed method improves throughput and latency by about 20.6% and 78.6 %, respectively, compared with an existing method. In addition, the amount of additional hardware is about 11 % for the whole circuit.

In this paper, a pseudo-noise (PN) code tracking method of optical-wireless code-shift-keying systems using two-layered PN codes is considered. A new tracking system that uses the two delay lock loops (DLL) is proposed. The two-layered PN codes used in this paper consist of spreading each chip of the selected modified pseudo-orthogonal M-sequence sets (MPOM) code, called outer PN code, with other MPOM code, called inner PN code. Therefore, the two DLLs of the proposed method are used in a suitable form for outer PN codes and inner codes of two-layered PN codes. In this paper, the symbol error rate taking into account jitter is evaluated.

This paper presents a method for radar-based contactless vital sensing and ECG (ElectroCardioGram) signal reconstruction using deep learning. ECG signal is a typical representation of heartbeat signals. However, its measurement usually requires any contact device, which is not suitable due to discomfort of subjects. Radar system is effective for vital sensing because it can measure small displacement of body surface caused by breathing and heartbeat without contact. On the other hand, most of the methods using radar system are limited to evaluating simple indices such as heart rate and heartbeat interval while subjects or devices are stationary. In this paper, we propose a method for body surface displacement signals using FMCW (Frequency-Modulated Continuous Wave) radar and reconstructing ECG signals using CNN (Convolutional Neural Network). The result of experiments on six healthy males shows the ECG signals are successfully reconstructed. Furthermore, we confirmed that the proposed method can reconstruct signal waveforms even in an environment with low SNR (Signal-to-Noise Ratio).

Power electronic devices such as Insulated Gate Bipolar Transistors (IGBT) are widely used in electronic systems. The failure of IGBT may bring fatal damage to the whole system. The ability to accurately predict the remaining useful life (RUL) of IGBT can provide a basis for intelligent equipment maintenance, thereby minimizing the risk of catastrophe failure at work. The research in this field aims to achieve accurate and timely RUL prediction of equipment. This paper exploited the IGBT accelerated aging data from NASA Ames Research Center, and the monitoring data of IGBT is collected under thermal overstress load condition with square signal bias at gate. This paper developed three machine learning models to predict the RUL of IGBT, including Back Propagation Neural Network(BPNN), Random Forest and Extreme Learning Machine. These models realize real-time prediction of RUL, and the performance of these models is observed and analyzed.

In order to meet the requirements of secure start-up of embedded devices, this paper designs a secure and trusted circuit to realize the secure and trusted start-up of the system. This paper analyzes the principle and method of the circuit design, and verifies the preset information of the embedded device before the start of the embedded device, so as to ensure that the start process of the embedded device is carried out according to the predetermined way, and then uses the security module to measure the integrity of the data in the start process, so as to realize a trusted embedded system. The experimental results show that the security module has stronger security features and low latency. The integrity measurement is implemented in the trusted embedded system to realize the safe startup of embedded devices.

Rapid and accurate detection of garbage has become a hot issue in the field of computer vision. In this paper, we propose an intelligent garbage detection algorithm to improve the accuracy and speed of garbage detection. For the self-made garbage dataset, to reduce the annotation workload and the problem of missing annotation and wrong annotation, YOLOv4 combined with self-learning strategy is used to realize the training of garbage detection model and improve the accuracy of detection tasks. At the same time, the fast NMS algorithm is introduced to accelerate the prediction process and improve the model prediction speed. The experimental results show that the detection effect of the algorithm in this paper is significant, and it can effectively identify six different kind of garbage, with mean Average Precision (mAP) of 95.36% and a detection time of 25ms for a single image, which is expected to be applied to the recognition and detection of waste targets in industrial conveying scenarios.

Consumer comments left on the Internet are an important basis for the change in product demand. This paper constructs a garbage comment recognition model based on product feature extraction and attribute emotion analysis. It predicts the product attribute attention and emotional change trends in the next stage by using the time series analysis model. Combined with the changing trend of historical data, the importance of product attributes and market satisfaction is analyzed. The research model proposed in this paper is verified using online automotive product review data. The research results can provide decision support for enterprises to develop marketing strategies and product improvement and innovation.

This paper briefly introduces the problems existing in production management of electric porcelain. In order to solve this problem, an integrated monitoring system of electric porcelain based on Web is designed. This design takes Microsoft SQL SERVER 2012 as the data storage platform, using. NET three- layer architecture, B/S system structure as the development mode, and through Web Service technology to achieve remote data transmission. Effectively improve the data integration degree of electric porcelain production. This paper mainly introduces the overall framework and the function of each module of the electric porcelain integrated monitoring system.

The basic guarantee factor for the construction of power Internet of Things is the comprehensive perception and interconnection of information. In the case of ensuring the normal operation of the power system, the efficiency and stability of the system should be improved as much as possible. With the digital transformation of the power grid, a large amount of data is generated by power terminals, which puts forward higher requirements on computing, storage, and network performance. Therefore, edge computing technology is introduced into the power Internet of Things. The edge computing framework proposed in this paper is based on the business needs of the power grid, and this paper describes the application of edge computing in typical scenarios such as transmission, substation, and distribution. edge computing has lower latency, higher efficiency, security protection and other performance. Therefore, the application of edge computing in the power Internet of Things is of great significance.

How to effectively identify the fraud in financial reports has been a hot issue for investors and capital market regulators. On the basis of analyzing the deficiency of the existing methods, this paper introduces the Sparse Group Lasso-Support Vector machine method into the field of financial report fraud Identification. According to the characteristics of this method, taking the data of listed companies in China from 2008 to 2018 as a sample, and according to the main factors affecting the fraud risk of financial reports, this paper puts forward 15 financial indicators and non-financial indicators to construct an explanatory variable set, then, this paper makes an empirical analysis by using this method. The results show that the method has higher recognition rate and interpretability than other methods, such as Logistic regression, lasso-Support vector machine and Group lasso-Support vector machine, when the explanatory variables have natural grouping structure and higher group correlation level, it has strong theoretical and practical value.

This paper proposes a weighted threshold calculation method based on the cluster routing protocol LEACH to improve the randomness of the cluster head election, which causes the node to die too quickly. There are two factors included in this method, namely: the current energy of the node and its distance to the base station. With the threshold calculation method proposed in this paper, the probability that the node closer to the base station with more residual energy which elected as the cluster head is increased. At the same time, base stations are introduced to act as cluster heads. Ordinary nodes near the base station can send data directly to the base station, shortening the transmission path and reducing network energy consumption, thus achieving the goal of energy balancing. The results of the simulation experiments show that the improved protocol outperforms the original one ,mainly in terms of network lifetime and the time of appearance of the first dead node.

Accurate land use information extraction plays an important role in land management, environmental monitoring, urban and rural planning and development and ecological assessment. High resolution remote sensing images have the characteristics of class similarity. Tradi¬tional methods of extracting ground object information from remote sensing images are limited to specific scenes and data sources, and have low segmentation accuracy and efficiency. In this paper, we shed the light to this research topic. To further improve the accuracy of land use information extraction, this paper takes HRNet(High-level Resolution Network) as the backbone network, which improves the performance of the network by fusing multi-scale feature map information through parallel work. Then, using the feature map information as input, the classification of each pixel is predicted through OCRNet network to achieve semantic segmentation of the image. The results show that: OCRNet-18 segmentation model is superior to other deep learning models in terms of PA, M IoU and Kappa coefficients, and the accuracy of PA (total), Kappa coefficient and M IoU are 0.913, 0.866, and 0.644 respectively.

Classification of soil is crucial for the agricultural domain as it is an essential task in geology and engineering domains. Various procedures are proposed to classify soil types in the literature, but many of them consumed much time or required specially designed equipments/applications. Classification of soil involves the accounting of various factors due to its diversified nature. It can be observed that several critical domain-oriented decisions often depend on the type of soil like farmers might be benefitted from knowing the kind of soil to choose crops accordingly for cultivation. We have employed different Convolution Neural Network (CNN) architectures to identify the soil type accurately in real-time. This paper describes the comparative evaluation in terms of performances of various CNN architectures, namely, ResNet50, VGG19, MobileNetV2, VGG16, NASNetMobile, and InceptionV3. These CNN models are used to classify four types of soils: Clay, Black, Alluvial, and Red. The performance of the ResNet50 model is the best with a training accuracy and training loss of 99.47% and 0.0252, respectively compared to other competing models considered in this paper.

Video surveillance is an important method to obtain the dynamic changes of green macroalgae along the coast. The paper proposes a coastal green macroalgae extraction method based on the SLIC superpixel segmentation, CNN and SVM to realize the automated recognition of green macroalgae from lots of high-resolution RGB video data collected by unmanned aerial vehicle (UAV) and handheld devices. Firstly, SLIC algorithm is used to generate the multi-scale patches on the original high-resolution image. Then, three classification CNN is used to divide the multi-scale patches into three types: green macroalgae, background and mixing. Finally, SVM algorithm is used to extract the green macroalgae to improve the accuracy at the pixel level in the mixed patches. In order to evaluate the performance of the proposed method, experiments are conducted on our coastal green macroalgae image dataset. Compared with the method of RGB vegetation indices (such as ExR, RGBVI, NGBDI), the overall accuracy (OA), F1 score, and Kappa of the green macroalgae extraction with the method proposed in this paper are up to 95.23%, 0.9612, 0.9436, respectively. The results show that our method is significantly better than that of RGB vegetation indices since it effectively reduces the influence of sea waves and light on the recognition results. The automated extraction method for coastal green macroalgae proposed in this paper can provide a reference for the automatic monitoring of coastal green macroalgae with high precision.

Pneumonia is one of the most common infectious diseases in clinic. X-ray chest is an important basis for early diagnosis of pneumonia. With the development of computer vision technology, using convolutional neural network to train pneumonia image classification model has been gradually applied to the process of medical clinical diagnosis. However, there are many problems in the process of using convolutional neural network to train pneumonia image classification model, such as too long model training time, over fitting and low accuracy due to too small training dataset. To solve these problems, this paper proposes an optimization method of pneumonia image classification model based on transfer learning and feature fusion, which is called Transfer Fusion. The Transfer Fusion optimization method will transplant the trained source model parameters to the target model, and add a specific feature fusion classification layer, so as to significantly shorten the training time of the new model, improve the accuracy and prevent over fitting. In this paper, Transfer Fusion optimization method is applied to three common convolutional neural network models: Google InceptionNetV3, MobileNetV2 and ResNet50. Through a large number of experiments, the performance of the three models has been significantly improved and improved.

Reliability and real-time are extremely important indicators in the aerospace field. With the increasing complexity of nano satellite missions, research on the reliability and real-time of on-board computers has become very important. This paper uses AM5748 chip manufactured by Texas Instruments as the processor, which is equipped with RT-Linux for the software design of on-board computers. Aiming at the reliability of on-orbit data storage of on-board computers, this paper proposes a loading method combining U-BOOT and N-modular redundant (NMR), which can solve the static operating system’s single event upset (SEU) problem well; aiming at the concurrency problem of on-board computer OS drivers, a priority-based serialized driver framework is designed to solve the possible concurrency problems in the driver layer to a certain extent. Finally, the system's five-mode redundancy is verified by the fault injection, the driver framework is tested by Multi-process concurrency. The results shows that the system can resist the single event upset rate of up to 10-5/(bit∙day), and the driver framework can complete on-orbit multi-task scheduling, which solves the concurrency problems. Thus, the work is of great significance for the reliability and real-time research of nano satellites and on-board computers in orbit applications.

From the perspective of big data, internet public opinions in higher education institutions are informative, arbitrary, invisible and abrupt. Taking consideration of these features, this paper constructs a model to tracking network public opinion hot spots in higher education institutions. Adopting data processing ways including Chinese words segmentation and text clustering based on Single-Pass, this model consists three parts: data collection, data processing and data analysis. Through this method, this paper figures out important contents among a bulk of online public opinion topics in colleges and universities and determinate hot spot topics focused by campus students.

By taking the fire hazards of the power cabin in Utility Tunnel as the starting point, researching the fire detection and fire early warning technology, sorting out several key links in the cable laying, the paper designs a Fire Protection Monitoring System for power cabin in Utility Tunnel based on IOT technology. The Fire Protection Monitoring System is aimed at realizing the automatic fire alarm in the power cabin and reducing the fire risk. The system also helps to enhance the level of intelligence of Utility Tunnel management.model proposed in this paper is validated through case study.

This paper studies and proposes an energy data transaction method based on the combination of multi-layer hierarchical fusion fuzzy feature mapping method (MLHFFFM) and long short-term memory neural network (LSTM). First, analyze the big data collected to determine the type of missing data. On this basis, MLHFFFM and fuzzy C-means algorithm (FCM) are used to cluster electricity big data, and the LSTM neural network is trained based on the characteristic values of different users and historical data, and different pricing strategies are designed for different data types. Finally, the feasibility and reliability of the improved method proposed in this paper are verified through the data analysis and comparison experiment of the public data set.

The purpose of image enhancement is to improve the visual effect of the image. It is one of the hot spots of digital image enhancement technology to realize the enhancement effect by changing the contrast of gray scale. In this paper, a method of image contrast enhancement based on mean value segmentation is proposed, which is combined with pattern and image processing technology, to select the appropriate image gray membership function, and to select the threshold value of fuzzy theory and image processing by means of iteration and Otsu method, finally, the final image was decoded and the image was enhanced. In this paper, this technique is applied to the study of image enhancement of locomotive wheelset, and the image results of different processing methods are analyzed.

To solve the problems of texture blurring and structure inconsistency in large area image inpainting. This paper proposes an image inpainting method based on multi-scale feature fusion. We design a multi-scale feature fusion module to expand the receptive field. Besides, we devise attention module to capture information from distant areas in the feature map. In this paper, partial convolution and recursive structure are adopted to repair the boundary of the missing area progressively. This method can continuously strengthen the constraint on the center of the missing region and make the repair results more refined. Experimental results show that, compared with the existing image inpainting methods, this structure improves the performance of image inpainting, and the image quality generated by it is the best.

Modular Multilevel Converter (MMC) has the advantages of strong scalability, high output voltage level, and low output current harmonics. It is widely used in DC transmission, offshore wind power grid connection, and AC/DC hybrid power distribution network. However, the circulation causes bridge arm current distortion and converter station loss. This paper designs circulation suppression strategy based on quasi-PR controller, which can effectively suppress the double, quadruple, and six-frequency circulation component by setting four quasi-PR controllers. The PSCAD simulation model verified the circulation suppression strategy based on the quasi-PR controller studied in this paper can effectively improve the performance of MMC.

Artificial intelligence (AI) is the most looked up technology with a diverse range of applications across all the fields, whether it is intelligent transportation systems (ITS), medicine, healthcare, military operations, or others. One such application is autonomous vehicles (AVs), which comes under the category of AI in ITS. Vehicular Adhoc Networks (VANET) makes communication possible between AVs in the system. The performance of each vehicle depends upon the information exchanged between AVs. False or malicious information can perturb the whole system leading to severe consequences. Hence, the detection of malicious vehicles is of utmost importance. We use machine learning (ML) algorithms to predict the flaw in the data transmitted. Recent papers that used the stacking ML approach gave an accuracy of 98.44%. Decision tree-based random forest is used to solve the problem in this paper. We achieved accuracy and F1 score of 98.43% and 98.5% respectively on the VeRiMi dataset in this paper. Explainable AI (XAI) is the method and technique to make the complex black-box ML and deep learning (DL) models more interpretable and understandable. We use a particular model interface of the evaluation metrics to explain and measure the model’s performance. Applying XAI to these complex AI models can ensure a cautious use of AI for AVs.