Search

Automatic modulation recognition (AMR) is considered one of most important techniques in the non-cooperative wireless communication systems. Traditional algorithms, e.g., support vector machine (SVM) based on high order cumulants (HOC), are hard to achieve the reliable performance. In this paper, we propose an effective AMR algorithm based on deep learning (DL) with capabilities of automatically extracting representative and effective features. Our proposed method resorts to in-phase and quadrature (IQ) samples which are IQ components of received baseband signal, respectively. We adopt convolutional neural networks (CNN) and recurrent neural networks (RNN) to classify six types of signal modulations over additive white Gaussian noise (AWGN) channel and Rayleigh fading channel, respectively. Simulation results show that DL-AMR is much better than traditional algorithms under two fading channels.

With the maturity of robotics and serious aging situation, home service robots have been prevailing around the world, which requires the robotic arm of home service robots to be flexible for obstacle avoidance to work in complex and changing home environment. Based on the actual requirements of home service robots, this paper proposes an obstacle avoidance planning algorithm based on improved artificial potential field method for dynamic obstacle avoidance of redundant manipulators. The velocity of the target object and obstacle were introduced into the end-planned field structure to plan a collision-free path for the end of the arm and track the dynamic target point. Meanwhile, the deviation speed of the arm member in the obstacle repulsive field was added to ensure obstacle avoidance. Finally, the effectiveness of the algorithm was verified in the simulation environment.

Automatic modulation recognition (AMR) is considered one of most important topics in non-cooperative communication systems. Traditional algorithms, e.g., high order cumulants (HOC) and support vector machine (SVM), are hard to achieve reliable performance. In this paper, we propose an effective AMR algorithm based on deep learning (DL) with capabilities of automatically extracting representative and effective features. The proposed method resorts to in-phase and quadrature (IQ) samples that are in-phase and quadrature components of received baseband signal, respectively. We adopt convolutional neural networks (CNN) and recurrent neural networks (RNN) to classify six types of signal modulations in additive white Gaussian noise (AWGN) channel and Rayleigh channel, respectively. Simulation results show that the features automatically extracted by DL algorithm is more suitable and effective for AMR and its performance is far beyond that of traditional algorithms in two channels.

In the process of daily care for the elderly, the bathing of the elderly in bed is a problem that cannot be ignored. The multi-functional bathing robot is a smart assistant aid that helps solve the problem of bathing in the elderly. In this paper, based on the actual needs of bathing for the elderly in bed, the ergonomic design of the multi-functional bathing robot was carried out. Firstly, the human body size based on anthropometry is studied. The body size of the elderly and the relevant dimensions of the nursing staff are analyzed and studied to determine the relevant dimensions of the mechanical structure of each part of the assisting bath robot. Secondly, through the questionnaire survey, the psychological needs of the elderly were clarified, and the appearance color of the multifunctional shower robot was determined. The material of the multifunctional assisted bath robot was selected based on the hygiene requirements. Finally, the refinement and rendering of the three-dimensional model of the bathing robot was completed, and the ergonomic design of the multi-functional bathing robot with safety, comfort, convenient operation was realized.

This paper studies positioning issue in navigation using InSAR concept. A novel method based on wrapped interferometric phase registration, which differs from common InSAR based navigation method, is proposed. This method may be useful when navigating throw hills or areas with low SAR image characteristics but rich terrain elevation features. Accurate DEM data is needed to generate ideal interferometric phase template for registration. Three dimensional positioning is achieved by estimating vertical bias and horizontal bias in order. The proposed method works well under different squint angle. Simulations are carried out to verify the method. Precision of the method under different squint angle is also exhibited in simulations.

How to correct test papers efficiently is an important problem that perplexes teachers in many colleges and universities. For the low efficiency of the total score calculation, this paper proposed an intelligent method based on image processing techniques and Convolutional Neural Network (CNN) to calculate the total score of each test paper automatically. Teachers can use the proposed system to calculate the total score of the test paper, which largely reduces teachers’ workload. The proposed model can quickly recognize and calculate the total score of test papers. The average time of the total score calculation of each test paper was 0.752 seconds in the experiment. Experimental result shows satisfying performance of the proposed method.

Traffic lights may cause a lot of braking, acceleration or even sharp braking, which leads to fuel waste, safety risks and increasing emission during driving. In this paper, we proposed a car navigation map equipped with speed recommendation, which guides the vehicle to timely arrival at green light with minimal use of braking. According to the location of the vehicle and the traffic signal information, we proposed a speed recommendation algorithm for driver to reduce the waiting time at the intersection with traffic lights, i.e., making the vehicle arrive at the intersection when the traffic light is green. The proposed algorithm allows the driver pass the intersection with traffic light without stopping. Thus, it makes driving safer, more efficient and environment-friendly. The simulation is conducted on a traffic simulation software named Vissim. Simulation result shows the satisfying performance of the proposed algorithm.

Due to workload fluctuation, the cloud vendors expect to rent their idle resources with virtual machines (VMs) configured in advance when their workload is lower. By integrating volunteer computing into cloud architectures, we envision a gigantic self-organizing cloud (SOC) being formed to reap the huge potential of untapped commodity computing power via the Internet. However, due to heterogeneity of the resources, the performance difference exists generally. To reduce the risk of the single-point-of-failure, we proposed an election-based resource provision algorithm using those VMs as discrete as possible. We formulate the concept of discreteness for allocating VM for a request that needs k virtual machines. Each VM consists of several servers in multiple cabinets. The discreteness is measured to be the number of cabinets that k virtual machines span. We show that it is an NP-hard problem, and provide a heuristic election algorithm based maximum cost performance. The technologies of analytic hierarchy process (AHP) and simple additive weighting (SAW) are adopted to guarantee VM performance that users hope to obtain. The experiment results show more benefits of our algorithm.

Traditional machine learning techniques may suffer from evasion attack in which an attacker intends to have malicious samples to be misclassified as legitimate at test time by manipulating the samples. It is crucial to evaluate the security of a classifier during the development of a robust system against evasion attack. Current security evaluation for Support Vector Machine (SVM) is very time-consuming, which largely decreases its availability in applications with big data. In this paper, we propose a fast security evaluation of support vector machine against evasion attack. It calculates the security of an SVM by the average distance between a set of malicious samples and the hyperplane. Experimental results show strong correlation between the proposed security evaluation and the current one. Current security measure min-cost-mod runs 24,000 to 551,000 times longer than our proposed one on six datasets.

Machine learning has been broadly applied in different applications due to its satisfying performance. In security-related applications, e.g. facial recognition and fingerprint identification, an adversary misleads the pattern recognition system on purpose by manipulating training or test samples. Black-box attack is one of adversarial attacks which camouflages unseen samples to evade the detection with limited information on the system. Recently, there are a few studies on the robustness of machine learning, especially deep learning, under black-box attack. In this study, we investigate the security of the stacked autoencoder, which is one of the most famous models in deep learning against black-box attack. A training dataset generation method is proposed for a substitute model in black-box attack. By enquiring labels of samples from the target classifier, we train a model to approximate the target classifier. To have a better approximation to the decision boundary of the target classifier, a hybrid data generation method is adopted to increase the number of the training data. We generate new samples by combining both near and far away the decision boundary of the substitute model. The experiment results suggest that our proposed method downgrades the classifier in terms of accuracy more significantly than the existing one within less label query times. It shows the proposed method is more efficient to train a substitute model to approximate target classifier.

The fundamental issue in cross project defect prediction is selecting the most appropriate training data for creating quality defect predictors. Another concern is whether historical data of open-source projects can be used to create quality predictors for proprietary projects from a practical point-of-view. Current studies have proposed statistical approaches to finding these training data, however, thus far no apparent effort has been made to study their success on proprietary data. Also these methods apply brute force techniques which are computationally expensive. In this work we introduce a novel data selection procedure which takes into account the similarities between the distribution of the test and potential training data. Additionally we use feature subset selection to increase the similarity between the test and training sets. Our procedure provides a comparable and scalable means of solving the cross project defect prediction problem for creating quality defect predictors. To evaluate our procedure we conducted empirical studies with comparisons to the within company defect prediction and a relevancy filtering method. We found that our proposed method performs relatively better than the filtering method in terms of both computation cost and prediction performance.

The method of fault location based on the traveling-wave detection device on the overhead transmission line has been proposed in this study. The high-frequency fault traveling-waves are collected by installing two traveling-wave detection devices based on Rogowski roil on the high-voltage transmission lines. Phase-mode transformation and the detection method of wavelet transform mode pole maximum is used to determine accurate time of line mode initial travelling wave and reflected waves reaching different devices respectively. All devices transmits the time information of the wave head via a wireless communication network back to master station; according to the time differences of different wave heads, the master station can preliminarily determine the fault section and the real-time traveling wave, ultimately realize the accurate fault location. Simulation results show that this method has a high range accuracy and reliability

For the purpose of WSN (Wireless Sensor Networks) performance improvement, mobile sinks were considered, which are a kind of special nodes in charge of data collection in WSN. Mobility of sinks can improve the performance of whole system, such as connectedness, sense coverage, and energy efficient. An effective structure regular hexagonal grid plot is chosen, which is suitable for improving connectedness and sense coverage. A route planning for mobile sinks is proposed by applying a well designed ant colony optimization algorithms. Finally, the advantages and effectiveness of mobility-based sinks in WSN are illustrated by an example.

The method of fault location based on the traveling-wave detection device on the overhead transmission line has been proposed in this study. The high-frequency fault traveling-waves are collected by installing two traveling-wave detection devices based on Rogowski roil on the high-voltage transmission lines. Phase-mode transformation and the detection method of wavelet transform mode pole maximum is used to determine accurate time of line mode initial travelling wave and reflected waves reaching different devices respectively. All devices transmits the time information of the wave head via a wireless communication network back to master station; according to the time differences of different wave heads, the master station can preliminarily determine the fault section and the real-time traveling wave, ultimately realize the accurate fault location. Simulation results show that this method has a high range accuracy and reliability

In sea sand by the determination of free chlorine ions to study the chloride ion to the strength of sand concrete. Experimental data show that between 0% to 0.1% chloride ion content of sea sand concrete strength comparison between the value of a group that is found in 20% of fly ash were mixed with 3 days flexural strength is increased with the concentration of chloride ion Rise, the other group had no obvious rule, dilute the sand and did not play down the comparison that does not dilute the sand sea sand either 3 days or 28 days strength values are higher than the desalination of sea sand, the presence of chloride ions in sea sand Early strength has a positive role in promoting.

Sea sand concrete is a kind of concrete in which mixed sea sand as fine aggregate, which is large-scale application in the coastal areas in recent years, especially in Ningbo area. The sea sand solves the problem of river sand shortage, coupled with cheaper price, most of the ready-mixed concrete companies are willing to use desalted sea sand instead of river sand. Many companies even are using sea sand without any treatments. The history of using sea sand in our country is very short, so the durability of the sea sand concrete need to be tested by many projects in the future, along with the development of science and technology and the invention of precision instruments, the durability of sea sand concrete will be adequate assurance. Ningbo area has a wide range usage of sea sand concrete. By measuring the strength of sea sand concrete with fly ash, finding out the law of strength for the sea sand concrete by the presence of fly ash. At the same time, we used electron scanner to analysis the sea sand concrete internal microstructure, researched the influence of chlorine salts on sea sand concrete micro-structural features.

Fault location of overhead transmission line is one of the key technologies which guarantee the reliability of electric power system. In this study, the bandwidth of transient traveling wave was analyzed and the method of current transient traveling wave acquisition using Rogowski coil in the overhead transmission line is proposed. The corresponding relationships among transient traveling wave, refraction wave and reflection wave are discussed when detection devices are installed at different positions. The requirements for number and location of traveling wave detection devices are proposed. Furthermore, the location equation of distributed fault location system is established and the differentiation method of fault type and the false fault point is demonstrated based on the conclusion. The simulation of PSCAD indicates that the method of distributed fault location could eliminate the error brought by the uncertainty of traveling wave velocity, and the total error was small. It was expected to be widely applied in overhead transmission lines.

The prediction of software defect-fixing effort is important for strategic resource allocation and software quality management. Machine learning techniques have become very popular in addressing this problem and many related prediction models have been proposed. However, almost every model today faces a challenging issue of demonstrating satisfactory prediction accuracy and meaningful prediction results. In this paper, we investigate what makes high-precision prediction of defect-fixing effort so hard from the perspective of the characteristics of defect dataset. We develop a method using a metric to quantitatively analyze the unpredictability of a defect dataset and carry out case studies on two defect datasets. The results show that data unpredictability is a key factor for unsatisfactory prediction accuracy and our approach can explain why high-precision prediction for some defect datasets is hard to achieve inherently. We also provide some suggestions on how to collect highly predictable defect data.

Hard disk drive (HDD) performance and reliability at different environments are the interests and concerns of HDD researchers, engineers and consumers. This paper reports the experimental investigation of HDD vibration characteristics with different internal air pressures. Experiments are done by placing the HDD samples inside an altitude chamber and the vibrations are measured using a Laser Doppler vibrometer (LDV) and all-in-one (AIO) motor tester. The disk drive read/write test, vibration spectrum of actuator & suspension tip and disk flutter measurements reveal that the air pressure significantly affects the vibration behaviors of hard disk drive components and the system performances.

In this paper the authors propose a new, straightforward approach to derive model of the pivot nonlinearity in the hard disk drive. The proposed nonlinear model is not based on classical pivot nonlinearity models. It is derived using frequency domain measurements obtained for different amplitudes of the actuator excitation current. One of the advantages of the obtained model is that it is derived using actual experimental measurements conducted under the real operating conditions of the hard disk drive. Experimental results show good match of the derived model with both the frequency and time domain measurements. The derived model is subsequently used for feedforward compensation of the pivot nonlinearity and the experimental results are presented. Simulation studies are carried out for adaptive neural network compensation using the derived model.

Many wireless sensor network applications must resolve the inherent conflict between energy efficient communication and the need to achieve desired quality of service such as end-to-end communication delay. To address this challenge, we propose the Real-time Power-Aware Routing (RPAR) protocol, which achieves application-specified communication delays at low energy cost by dynamically adapting transmission power and routing decisions. RPAR features a power-aware forwarding policy and an efficient neighborhood manager that are optimized for resource-constrained wireless sensors. Moreover, RPAR addresses important practical issues in wireless sensor networks, including lossy links, scalability, and severe memory and bandwidth constraints. Simulations based on a realistic radio model of MICA2 motes show that RPAR significantly reduces the number of deadlines missed and energy consumption compared to existing real-time and energy-efficient routing protocols.

This paper deals with a new medical information system called Alarm Net designed for smart healthcare. Based on an advanced Wireless Sensor Network (WSN), it specifically targets assisted-living residents and others who may benefit from continuous and remote health monitoring. We present the advantages, objectives, and status of the system built at the Department of Computer Science at UVA. Early results of the prototype suggest a strong potential for WSNs to open new research perspectives for ad hoc deployment of multi-modal sensors and improved quality of medical care

Product development and design stage determines the large part of cost of whole product life-cycle, and bill of materials (BOM) is the important tool of this stage to control cost. On the BOM tree, every product is constituted with several parts, and every part is the most appropriate of several spare ones. This paper uses analysis network process (ANP) to confirm the weight of each spare part, and establishes quality-cost optimization mode based on BOM. Furthermore, the project of performance-cost compare and the first-rank project in the certain target cost are proposed by simulated annealing. Finally, the application of simulated annealing in BOM is studied.