Search

Background: Informative gene selection is an essential step in performing tumor classification. However, it is difficult to select informative genes related to tumors from large-scale gene expression profiles because of their characteristics, such as high dimensionality, relatively small samples, and class imbalance, and some genes are superfluous and irrelevant. Objective: Many researchers analyze and process gene expression data to obtain classified gene subsets by using machine learning methods. However, the gene expression profiles of tumors often have massive computational challenges. In addition, when improving feature importance and classification accuracy, cost estimation is often ignored in traditional feature selection algorithms, which makes tumor classification more difficult. Methods: In this study, a novel informative gene selection method based on cost-sensitive fast correlation- based filter (CS-FCBF) feature selection is proposed. Results: First, the symmetric uncertainty index is used to evaluate the correlation between informativeFirst, the symmetric uncertainty index is used to evaluate the correlation between informative genes and class labels, then a large number of irrelevant and redundant genes are quickly filtered according to importance. Thereby, a candidate gene subset is generated. Second, costsensitive learning, which introduces the misclassification cost matrix and support vector machine attribute evaluation, is used to obtain the top-ranked gene subset with minimum misclassification loss. Finally, the candidate gene subset is optimized. genes and class labels, then a large number of irrelevant and redundant genes are quickly filtered according to importance. Thereby, a candidate gene subset is generated. Second, costsensitive learning, which introduces the misclassification cost matrix and support vector machine attribute evaluation, is used to obtain the top-ranked gene subset with minimum misclassification loss. Finally, the candidate gene subset is optimized. Conclusion: This experiment was verified in eight independent tumor datasets. By comparing and analyzing CS-FCBF with another three hybrids of typical gene selection algorithms combined with cost-sensitive learning, we found that the method proposed in this study has a better classification performance with fewer selected genes, which might provide guidance in tumor diagnosis and research.

Ultra-wideband (UWB) technology is a mature technology that contested other wireless technologies in the advent of the IoT but did not achieve the same levels of widespread adoption. In recent years, however, with its potential as a wireless ranging and localization solution, it has regained momentum. Within the robotics field, UWB positioning systems are being increasingly adopted for localizing autonomous ground or aerial robots. In the Industrial IoT (IIoT) domain, its potential for ad-hoc networking and simultaneous positioning is also being explored. This survey overviews the state-of-the-art in UWB networking and localization for robotic and autonomous systems. We also cover novel techniques focusing on more scalable systems, collaborative approaches to localization, ad-hoc networking, and solutions involving machine learning to improve accuracy. This is, to the best of our knowledge, the first survey to put together the robotics and IIoT perspectives and to emphasize novel ranging and positioning modalities. We complete the survey with a discussion on current trends and open research problems., Comment: 2021 10th Mediterranean Conference on Embedded Computing, Track: CPS&IoT

Obtaining high-precision wind speed prediction results is very beneficial to the utilization of wind energy and the operation of the power system. The purpose of this study is to develop a novel model for wind speed causality processing and short-term wind speed forecasting. In this study, the hybrid model combining causality processing strategy called “decomposition- virtual nodes-pruning” and Long Short-Term Memory Network based on Neighborhood Gates is proposed to obtain high-precision wind speed predictions. First, Pearson Correlation Coefficient, Maximal Information Coefficient and Granger causality test are used to explore the correlation and causality between wind speed and meteorological factors. Then, the causality is divided into five categories: center, chained, ring, tree and network causality, according to the topological structure of causality. Next, all types of causality can be unified into an equivalent tree causality by the causality processing strategy. Afterward, Long Short-Term Memory Network based on Neighborhood Gates is proposed to dynamically adjust the network structure according to the specific equivalent tree causality. Finally, the performance of the proposed model is verified by eight models from three aspects with different features, different methods and different equivalent trees in the case in Fuyun meteorological station, Xinjiang province, China. The evaluation metrics of the prediction results obtained by the proposed model are optimal among the eight models. The experimental results show that the proposed model is very competitive and very suitable for processing complex causality in wind speed prediction.

Depression, a major mental illness, has widely affected lives all over the world. Being depressed not only affects patients’ mood, but also has a negative impact on patients’ physical and mental health. It may lead to the lack of enthusiasm for daily life, low mental state, anxiety, irritability, anger and even suicidal tendencies. As the need to automatically detect depression using machine learning algorithms increases, an automatic depression detection method based on audio files and convolutional neural network (CNN) is proposed in this paper. First of all, we delete long silent sections of each audio file and splice the rest into a brand-new one. After that, add the label which represents if the participant is healthy or not to each file. Then, Mel frequency cepstrum coefficients (MFCCs), the features of speech signal, are extracted into matrix vector feature to represent the particular characteristics of participants’ own voice. Eventually, the features are imported into the convolutional neural network model to complete the model training and evaluation. The results on Distress Analysis Interview Corpus-Wizard of Oz (DAIC-WOZ) dataset show that the overall prediction accuracy is 0.85, and the average probability of correct prediction of a single file is 0.82.

Mobile communication base stations are the basic facilities of telecommunication operation networks. When the communication base station is struck by lightning, a very high overvoltage will appear on the top of pole, which leads the power supply cable of RRU(Remote Radio Unit) directly break through, and a large lightning current will flow through the cable, damaging the communication equipment and the power supply. Therefore, the research on the lightning current distribution characteristics of the mobile communication base station has important theoretical significance and engineering application value for the lightning protection of the mobile communication base station. In this paper, a simulation model of the current distribution characteristics of the communication pole system with overvoltage protective varistor is established, and the influence of the impulse current, the length of the pole and the DC reference voltage on the current distribution characteristics of the pole system is studied. The results show that: (1) When $8 /20\ \mu \mathrm{s}$ impulse current is injected into the 10m long pole system, the dynamic resistance $r$ of varistor decreases with the increase of impulse current peak value, the current distribution ratio of pole and core of power cable gradually approaches to 9:1. (2) As the length of the pole increases, the influence of the varistor on the current waveform parameters of the core branch of the power gradually decreases, and the front time distortion rate through the power line core branch decreases. (3) When the peak value of the impulse current and the length of the pole are the same, the lower the DC reference voltage of the varistor, the smaller the current waveform distortion of the core branch of the power, and the greater the current distribution.

More widespread adoption requires swarms of robots to be more flexible for real-world applications. Multiple challenges remain in complex scenarios where a large amount of data needs to be processed in real-time and high degrees of situational awareness are required. The options in this direction are limited in existing robotic swarms, mostly homogeneous robots with limited operational and reconfiguration flexibility. We address this by bringing elastic computing techniques and dynamic resource management from the edge-cloud computing domain to the swarm robotics domain. This enables the dynamic provisioning of collective capabilities in the swarm for different applications. Therefore, we transform a swarm into a distributed sensing and computing platform capable of complex data processing tasks, which can then be offered as a service. In particular, we discuss how this can be applied to adaptive resource management in a heterogeneous swarm of drones, and how we are implementing the dynamic deployment of distributed data processing algorithms. With an elastic drone swarm built on reconfigurable hardware and containerized services, it will be possible to raise the self-awareness, degree of intelligence, and level of autonomy of heterogeneous swarms of robots. We describe novel directions for collaborative perception, and new ways of interacting with a robotic swarm.

This conceptual paper discusses how different aspects involving the autonomous operation of robots and vehicles will change when they have access to next-generation mobile networks. 5G and beyond connectivity is bringing together a myriad of technologies and industries under its umbrella. High-bandwidth, low-latency edge computing services through network slicing have the potential to support novel application scenarios in different domains including robotics, autonomous vehicles, and the Internet of Things. In particular, multi-tenant applications at the edge of the network will boost the development of autonomous robots and vehicles offering computational resources and intelligence through reliable offloading services. The integration of more distributed network architectures with distributed robotic systems can increase the degree of intelligence and level of autonomy of connected units. We argue that the last piece to put together a services framework with third-party integration will be next-generation low-latency blockchain networks. Blockchains will enable a transparent and secure way of providing services and managing resources at the Multi-Access Edge Computing (MEC) layer. We overview the state-of-the-art in MEC slicing, distributed robotic systems and blockchain technology to define a framework for services the MEC layer that will enhance the autonomous operations of connected robots and vehicles., Accepted to the Fifth International Conference on Fog and Mobile Edge Computing (FMEC 2020)

A recent trend in the IoT is to shift from traditional cloud-centric applications towards more distributed approaches embracing the fog and edge computing paradigms. In autonomous robots and vehicles, much research has been put into the potential of offloading computationally intensive tasks to cloud computing. Visual odometry is a common example, as realtime analysis of one or multiple video feeds requires significant on-board computation. If this operations are offloaded, then the on-board hardware can be simplified, and the battery life extended. In the case of self-driving cars, efficient offloading can significantly decrease the price of the hardware. Nonetheless, offloading to cloud computing compromises the system's latency and poses serious reliability issues. Visual odometry offloading requires streaming of video-feeds in real-time. In a multivehicle scenario, enabling efficient data compression without compromising performance can help save bandwidth and increase reliability.

The need of data compression at smart Edge/Fog-based gateways is undeniable as data compression can significantly reduce the amount of data that has to be transmitted over a network. This, in turn, has a direct impact on reducing transmission latency and increasing network bandwidth. In time-critical and data sensitive IoT applications such as healthcare, lossless data compression is preferable as compressed data can be recovered without losing any information. However, it is not an easy task to choose a proper lossless data compression algorithm for IoT applications as each lossless data compression method has its own advantages and disadvantages. This paper focuses on the analysis of lossless data compression algorithms run at smart Edge/Fog gateways. Widely used lossless data compression are run at different hardware which is often used as smart Fog/Edge gateways. The latency of data compression and compression rate in different cases of input data sizes are analyzed. The paper provides guidelines for choosing a proper lossless data compression algorithm for time-critical IoT applications.

Ag/AgBr/AgCl hollow microspheres were synthesised through template assisted synthesis method followed by calcination and light reduction method. Field emission scanning electron microscopy, UV–Vis absorption spectra and X-ray diffraction were utilized to characterize the morphology, optical absorption property and phase structure of the as-prepared photocatalyst, respectively. By degradation of Rhodamine B under visible light, the Ag/AgBr/AgCl hollow microspheres exhibited a significant enhancement in photocatalytic activity compared with Ag/AgCl and Ag/AgBr. Furthermore, the heterojunction formed between AgCl and AgBr, the remarkable visible light absorption attributed to Ag nanoparticles on the surface of AgBr are considered as the main reasons of the enhancement.

This paper presents an energy optimization based planning method for grasping bottles on an assembly line with a constant velocity using a seven-DOF robotic arm. The interception region is the intersection of the assembly line and the workspace of the end-effector of the robot. There will be a dedicated optimal final pose in the interception region when grasping occurs so that the robot consumes least energy. On the other hand, even with the same final pose of the end-effector, there will be an optimal trajectory from the start pose to the final pose that consumes least energy. We use Gaussian quadrature optimization method provided by GPOPS with minimum energy cost criteria to solve the two optimization problems at the same time. Simulation shows that optimization based planning approach consumes lower energy than proportional navigation guidance and polynomial planning methods.

In both the traditional diode uncontrollable rectifier circuit and thyristor phased rectifier circuit, the shortcomings of low power factor are difficult to overcome. PWM rectifier circuit is to use full control power switch devices and adopts PWM control mode of rectifier circuit, through the adoption of appropriate control strategies and PWM modulation technology, PWM rectifier circuit can realize real-time tracking ac voltage ac current to obtain higher power factor rectifier. Firstly, the basic principle of PWM rectifier circuit is analyzed in this paper. The main controlled factor can be obtained through the analysis of the principle of PWM rectifier circuit, and the target control. At the same time, to design circuit and control strategy by using Simulink in MATLAB software simulation. The simulation results show that the theoretical research and design method in this paper are reasonable and feasible.

Aiming at the complicated control problem of the existing inductively coupled power transfer system, a self-oscillating control method for the push-pull resonant converter with parallel-series compensation is proposed. The necessity of designing the start-up circuit is explained by illustrating the principle of self-oscillating control. Based on the analysis of the energy transfer mechanism of the start-up circuit, the method of injecting energy into the resonant tank to compensate the energy consumed by the resistance of the damped oscillation is obtained to solve the circuit start-up problem. A self-oscillating resonant converter experimental apparatus without the external controlling signal is designed and relevant parameters are tested, which shows the correctness of the start-up principle analysis and the simple reliability of the circuit designing.

Aiming at the problem of frequency stability of contactless power transmission system, a self-oscillating inductive coupled power transfer system suiting for parallel resonance is proposed. The switching frequency is automatically tracked through designing the start-up circuit to monitor the zero-crossing of the resonant network voltage and taking the zero-crossing as the condition to drive switches conducting alternatively to ensure that the system is in resonance state. The self-oscillating controlling principle is introduced, the energy transfer process of the self-oscillating push-pull system based on parallel-series compensated topological is analyzed, and the main circuit parameters method of the system is deduced. Experiment results show that the theoretical analysis and the proposed method are valid and feasible.

To provide the designer preview of the final color appearance of tufted carpet, this paper proposed a computer aided method to simulate the color appearance of tufted carpet which is made with different color yarns. The simulation method adopts a color block to represent a single loop pile in the carpet surface, and through adjusting color block size to simulate the extrusion between loop piles. The method starts from scanning the control pattern to analyze the color information for each pixel in control pattern, and convert color into the pile height level information. Base on the current and surrounded loop piles height level to infer the size and start point of color block for current loop pile. According the cam pattern, creel pattern, and color block information to generate the color appearance simulation for the tufted carpet. Test result shows that this method can effectively generate carpet color appearance simulation for tufted carpet comprising of multicolor yarns.

A new developed tufting machine adopt electronic cam to replace traditional transverse shifting mechanism. To facility the control operation of electronic cam for carpet tufting machine, this study proposed a pattern data prepare system. According to the actual control of electronic cam based needle bar transverse shifting, constructed the software frame structure, using visual basic develop the cam data preparation system. The software system realizes the functions of gathering point data, conversion of data, download and management of cam data etc. The developed data preparation system was tested in a sample carpet tufting machine. Test result shows that the system can fulfill the cam pattern design, conversion, download and file management, and the developed system greatly reduce the complex operations of electronic cam drive transverse shifting system on carpet tufting machine.

At present, most Ad Hoc grid task scheduling algorithms ignore the impact of trust mechanism, resulting in the potential dangerous in the real scheduling. In this paper, the trust model is defined, which satisfies the tasks' trust demands, in order to choose the suitable resource node to be executed in the scheduling. Moreover, a task scheduling algorithm extended from Min-Min algorithm (T-Min-Min) is proposed. The proposed algorithm not only satisfies the trust demand of the user but also considers the resource nodes' energy consumption and load balancing. The simulation suggests that it can effectively shorten the task completion time and the number of failed tasks in the Ad Hoc grid.