arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Mu Song

33 papers · 1994–2022

1994: 11995: 01996: 01997: 01998: 01999: 12000: 02001: 02002: 52003: 22004: 12005: 02006: 22007: 02008: 02009: 02010: 12011: 42012: 42013: 22014: 12015: 22016: 32017: 12018: 02019: 02020: 02021: 22022: 1
1994peak 5 in 20022022

Frequent co-authors

Papers

showing 1–25 of 33 RSS

Unmixing grain-size distributions in lake sediments: a new method of endmember modeling using hierarchical clustering

Xiaonan Zhang, Aifeng Zhou, Xin Wang +5 more

Artificial Intelligence · Quaternary Research · · cited 50× · DOI ↗ · more like this ·

Abstract The grain-size distribution (GSD) of sediments provides information on sediment provenance, transport processes, and the sedimentary environment. Although a wide range of statistical parameters have been applied to summarize GSDs, most are directed at only parts of the distribution, which limits the amount of environmental information that can be retrieved. Endmember modeling provides a flexible method for unmixing GSDs; however, the calculation of the exact number of endmembers and geologically meaningful endmember spectra remain unresolved using existing modeling methods. Here we present the methodology hierarchical clustering endmember modeling analysis (CEMMA) for unmixing the GSDs of sediments. Within the CEMMA framework, the number of endmembers can be inferred from agglomeration coefficients, and the grain-size spectra of endmembers are defined on the basis of the…

A comparative study of learning methods in tuning parameters of fuzzy membership functions

Mu-Song Chen

Artificial Intelligence · · cited 29× · DOI ↗ · more like this ·

We compare several popular training algorithms for tuning parameters of fuzzy membership functions (MFs). The algorithms compared are gradient descent (GD), resilient propagation (RPROP), Quickprop (QP), and Levenberg-Marquardt (LM) algorithms. These algorithms are combined with RLSE (recursive least squares estimate) to improve the efficiency of an ANFIS (adaptive network-based fuzzy inference system). The results, on average, show that the relative performance of these algorithms depends on the given task, but that RPROP produces better performance in terms of convergence speed, stability, and generalization properties.

Neural networks training using genetic algorithms

Mu-Song Chen, Fong Hang Liao

Artificial Intelligence · · cited 24× · DOI ↗ · more like this ·

Presents a genetic algorithm based system for evolving neural networks. New genetic operators, which combine a heuristic approach and pseudo gradient information, are designed to enhance the performance of genetic algorithms. In this way, the extension or contraction of search region can be more adaptive to the characteristics of the neural network's output error surface. The proposed methods are tested on the n-bit parity problem. By applying these methods, we have been able to find single layer networks in solving 2-, 3-, and 4-bit parity problems. Moreover, we attempt to incorporate GAs into the cascade correlation algorithm in optimizing the network architecture. Because of the complementary properties of exploration capability of genetic algorithms and local search of the derivative-base approach, the hybrid method is expected to outperform either method alone. Experimental results…

Image Denoising Using Adaptive Neuro-Fuzzy system

Nguyễn Minh Thạnh, Mu-Song Chen

Artificial Intelligence · International MultiConference of Engineers and Computer Scientists · · cited 16× · OpenAlex ↗ · more like this ·

Abstract — In this paper, we propose a generalized fuzzy inference system (GFIS) in noise image processing. The GFIS is a multi-layer neuro-fuzzy structure which combines both Mamdani model and TS fuzzy model to form a hybrid fuzzy system. The GFIS can not only preserve the interpretability property of the Mamdani model but also keep the robust local stability criteria of the TS model. Simulation results indicate that the proposed model shows a high-quality restoration of filtered images for the noise model than those using median filters or wiener filters, in terms of peak signal-to-noise ratio (PSNR). Index Terms —Generalized Fuzzy Inference System, Mamdani model, TS model, PSNR I. I NTRODUCTION The image corrupted by different kinds of noises is a frequently encountered problem in image acquisition and transmission [1]. The noise comes from noisy sensors or channel transmission…

An efficient learning method of fuzzy inference system

Mu-Song Chen, Ren-Jean Liou

Artificial Intelligence · · cited 11× · DOI ↗ · more like this ·

One of the important problems to be solved for fuzzy inference systems is to tune the free parameters for solving the given task. In this paper, we propose to combine the RPROP adaptive learning algorithm, which is much faster than the gradient descent type, with the recursive-least-squares-error technique for tuning parameters of fuzzy membership functions. The work is based on the previous study of the adaptive network based fuzzy inference system (Jang (1993)). The proposed method is tested on several function approximation and dynamic system identification problems, and the results are compared with that of the gradient descent technique. The simulation results show improvements in learning speed and error convergence over the gradient-descant method.

Analysis and optimization of neural networks for remote sensing

Xian-Ping Jiang, Mu‐Song Chen, M.T. Manry +2 more

Artificial Intelligence · Remote Sensing Reviews · · cited 11× · DOI ↗ · more like this ·

Abstract A technique for improving the topology of a trained neural network, used for an inversion or classification problem, is presented. The technique models the multilayer perceptron as a power series, which allows us to (1) remove units from the network which are well‐approximated by zero‐degree or first‐degree polynomials, (2) measure the effect of removing a hidden layer, and (3) determine the degree of the overall polynomial discriminant which approximates the network. The smaller, pruned networks can process data faster than can the larger original networks. The network degree is a direct measure of the nonlinearity inherent in the particular inversion or classification problem of interest. Neural networks for inversion of surface scattering parameters and classification of sea ice are analyzed to illustrate the technique.

Constructive proof of efficient pattern storage in the multi-layer perceptron

Aishwarya Gopalakrishnan, Xiangping Jiang, Mu-Song Chen +1 more

Artificial Intelligence · · cited 10× · DOI ↗ · more like this ·

We show that the pattern storage capability of the Gabor polynomial is much higher than the commonly used lower bound on multi-layer perceptron (MLP) pattern storage. We also show that multi-layer perceptron networks having second and third degree polynomial activations can be constructed which efficiently implement Gabor polynomials and therefore have the same high pattern storage capability. The polynomial networks can be mapped to conventional sigmoidal MLPs having the same efficiency. It is shown that training techniques like output weight optimization and conjugate gradient attain only the lower bound of pattern storage. Certainly they are not the final solutions to the MLP training problem.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Adaptive equalization using complex-valued multilayered neural network based on the extended Kalman filter

Ren-Cheng Huang, Mu‐Song Chen

Artificial Intelligence · · cited 8× · DOI ↗ · more like this ·

In this paper, a complex-valued neural network based on the Kalman filter is presented for channel equalization in a communication system. The complex-valued decoupled extended Kalman filter algorithm is derived which provides a faster convergence rate and better performance than those of complex back-propagation algorithms. Computer simulation results showed that the proposed scheme has a faster convergence behavior and smaller signal constellation than the gradient descent based learning algorithm via the linear and nonlinear channel equalization problem.

Routability-Enhanced Scheduling for Application Mapping on CGRAs

Mu Song, Yi Zeng, Bo Wang

Hardware and Architecture · IEEE Access · · cited 6× · DOI ↗ · more like this ·

Coarse-Grained Reconfigurable Architectures (CGRAs) are a promising solution to domain-specific applications for their energy efficiency and flexibility. To improve performance on CGRA, modulo scheduling is commonly adopted on Data Dependence Graph (DDG) of loops by minimizing the Initiation Interval (II) between adjacent loop iterations. The mapping process usually consists of scheduling and placement-and-routing (P&R). As existing approaches don’t fully and globally explore the routing strategies of the long dependencies in a DDG at the scheduling stage, the following P&R is prone to failure leading to performance loss. To this end, this paper proposes a routability-enhanced scheduling for CGRA mapping using Integer Linear Programming (ILP) formulation, where a global optimized scheduling could be found to improve the success rate of P&R. Experimental results show that our approach…

RETRACTED: Assisting Smart Construction With Reliable Edge Computing Technology

Qiang Yue, Mu Song, Longguan Zhang +5 more

Computer Networks and Communications · Frontiers in Energy Research · · cited 5× · DOI ↗ · more like this ·

Although smart construction tools are already assisting the construction sector, still major reforms are considerably seen in the coming years with Artificial Intelligence of Things (AIoT). The purpose of this work was to develop a down-to-earth reliable edge-computing design and plan, which can be utilized to bolster savvy development situations with high quality of service. The number of devices connected to industrial processes is growing in tandem with the advancement of sophisticated technologies and the use of the AIoT. Big data are created as gadgets grow more integrated and need more computing power. However, as the amount of big data generated grows, processing and analytical issues arise. Cloud computing technologies are now being used to solve processing and analytical challenges. However, by using edge computing technologies, greater emphasis is placed on executing…

Online transductive support vector machines for classification

Mu-Song Chen, Tze-Yee Ho, Deng-Yuan Huang

Computer Vision and Pattern Recognition · · cited 5× · DOI ↗ · more like this ·

Transductive support vector machine (TSVM) is one kind of transductive inference process, which combines labeled samples with unlabelled samples to derive the decision rules for classification tasks. Compared with the classical SVM, the transductive SVM is more robust and can achieve better performance. However, there are some disadvantages still being explored. One of the vital drawbacks is its computational costs. Usually, the SVM and TSVM models need to be retrained from scratch for parameter variations whenever any new samples become available. This problem has hindered the use of transduction learning in many real world applications. To resolve these problems, the online transductive support vector machine (OTSVM) is present to improve the generalization performance of the TSVM model. The OTSVM integrates the incremental learning/decremental unlearning to learn new incoming…

Face recognition based on collaborative image similarity assessment

Deng-Yuan Huang, Ta-Wei Lin, Wu-Chih Hu +1 more

Computer Vision and Pattern Recognition · · cited 5× · DOI ↗ · more like this ·

In this paper, we propose an innovative approach for face recognition based on collaborative image similarity assessment (CISA). In the proposed method, the test sample is first represented by a linear combination of all the training samples for each face class. The classification task is then conducted using the similarity measures including structure similarity index measure (SSIM), root mean square (RMS), and similarity assessment value (SAV). Since CISA is only one phase, it is computationally efficient when comparing with the method of two-phase test sample sparse representation (TPTSR). To verify the performance of face classification, two popular face databases of the ORL and FERET are used. Results show that CISA is comparable with TPTSR on the classification rates for ORL database. However, CISA greatly outperforms TPTSR on the evaluation of the FERET database. Moreover, only…

An Educational Testbed for the Computational Analysis of Collaboration in Early Stages of Software Development Processes

Thomas Kowark, Junjin Mu, Mu Song +1 more

Information Systems · · cited 4× · DOI ↗ · more like this ·

Agile software development processes are widely adopted in software engineering projects. Their low organizational overhead and iterative nature make them ideal choices for small development teams. The application of those methods in software projects that require collaboration between multiple sub-teams is a challenging task that remains subject to intensive research. Especially the initial phases of such projects are crucial for project success since a problem-free inception period generates a basis for efficient development later in the process. We introduce a testbed that allows analyzing collaboration processes during those early stages of software development within a low-risk, educational setup. Participants of a software engineering lecture form development teams of considerable size and develop real-life applications in a realistic, yet controlled, environment. By combining…

A Neural Network Approach for Controller Area Network Message Scheduling Control

Chuan Ku Lin, Hao-Wei Yen, Mu-Song Chen +2 more

Hardware and Architecture · International MultiConference of Engineers and Computer Scientists · · cited 3× · DOI ↗ · more like this ·

The Scholars perceived the inaugural year of the MACC Scholars Program as extremely helpful and productive. Ongoing efforts should be made to continue to promote the development of diverse junior scientists in HIV research.

Development of Car Cloud Sensor Network Application Platform

Chi-Pan Hwang, Mu-Song Chen, Hsuan-Fu Wang

Computer Networks and Communications · · cited 2× · DOI ↗ · more like this ·

The development of car cloud sensor network isbased on the open sources of Hadoop that support themultidimensional information service studies in this paper. Theoverall system is divided into the layers of physical sensorinformation, communications platform, real-world information, aswell as applications to develop. The function of physical sensorinformation layer builds the connection between in-vehiclenetwork and 3G/4G mobile phone being bridged by the Bluetooth. The physical sensor information is collected into the distributivefile systems on the cloud computing platform by socket providedby communication platform. Finally, a car cloud sensor networkcontained privacy cars and environment regions sensor networksare generated by real-world information layer, and in turnsupport the development of applications based on the car cloudsensor network. The functions of application layer have…

Analysis of the Queue Service Probability for the EDF Scheduling Algorithm

Mu-Song Chen, Chi-Pan Hwang, Hsuan-Fu Wang

Hardware and Architecture · · cited 2× · DOI ↗ · more like this ·

This paper is concerned with the Earliest-DeadlineFirst (EDF) policy for scheduling a constrained multi-queuesystem with a single server in the overloaded environment. Thisstudy originates from the research of the controller area network(CAN). In the CAN system, variability and constraints, e.g. heavy(or overloaded) traffic, timing constraints, limited bandwidth, etc., have crucial impacts on its scheduling performance. Violating these constrictions may lead to a lack of the ability ofmaintaining the integrity of the system communication. In thissense, this paper discusses the applicability of Earliest DeadlineFirst (EDF) technique to the scheduling of CAN messages. TheEDF scheduling algorithm permits more general applications tothe real time system under various situations and timingconstraints. It assumes implicitly that a message's urgencyincreases with the imminence of its deadline.…

An Online RBF Network Approach for Adaptive Message Scheduling on Controller Area Networks

Mu-Song Chen, Hao-Wei Yen

Hardware and Architecture · Journal of information science and engineering · · cited 2× · DOI ↗ · more like this ·

The Controller Area Network (CAN) is a communication bus for message transaction in real-time environments. A real-time system typically consists of several classes of messages and a scheduler is responsible to allocate network resources to fulfill timing constraints. Given sufficient bandwidth, the static scheduling algorithms can meet the bounded time delay. However, due to the availability of network bandwidth, not all messages can gain enough bandwidth to achieve the best control performance. Therefore, adaptive message scheduling policies that optimize the bandwidth utilization while supporting timeliness guarantees are of special interest. In this paper, we devise an online adaptive Message Scheduling Controller (MSC), which is designed to dynamically respond to the network dynamics. The MSC is realized by the Radial Basis Function (RBF) network with supervised parameter tuning.…

Online Controller Area Network message scheduling: analysis, implementation and applications

Mu Song Chen, Hao Wei Yen

Hardware and Architecture · International Journal of Systems Control and Communications · · cited 2× · DOI ↗ · more like this ·

Message scheduling is a critical research issue in the Controller Area Network (CAN). Messages that fail to meet predefined deadlines may deteriorate system performance significantly. In this paper, we propose an online Message-Scheduling Controller (MSC), which is realised by the Radial Basis Function (RBF) network. The MSC exploits two novel learning algorithms to adaptively tune its internal parameters. The learning algorithms provide complementary effects to prevent possible causes of non-uniform bandwidth allocation among requesting messages and to reduce possibilities of Transmission Failures (TFs). Simulation results show that the proposed algorithm outperforms the existing scheduling methods.

Crossover operators with adaptive probability

Mu-Song Chen, Fong Hang Liao

Artificial Intelligence · · cited 2× · DOI ↗ · more like this ·

Genetic algorithms (GAs) are adaptive methods, which can be employed to solve search and optimization problems. The GA relies on genetic operators to exchange gene between individuals for generating better offspring. An important issue to execute GA efficiently is to maintain population diversity and to sustain local improvement in the search stage. However, both effects always hinder each other. We propose to apply different kinds of crossover operators, i.e. arithmetic and BLX-/spl alpha/ crossovers, to control the diversity and convergence of the GA in continuous-space framework. We also utilize self-adaptation method to control the probability of crossover such that the balance of exploitation and exploration can be kept. It is shown empirically that the proposed methods outperform the classical GA strategy on several benchmark functions.

Recursive MVDR for direction finding using circular arrays

Mu-Song Chen, Ren-Jean Liou

Signal Processing · · cited 1× · DOI ↗ · more like this ·

Digital beamforming is one of the important topics in wireless communication systems. The goal of beamforming is to detect the arriving angles of various sources such that the direction of arrivals (DOA) for different signals can be determined for further process. For direction finding, traditional minimum variance distortionless response (MVDR) beamformer provides a straightforward approach in solving this problem. The approach of MVDR is based on optimum filtering and spectrum analysis. However, the process requires extensive computation to inverse the correlation matrix, which creates major obstacle in real-world applications, especially when the dimensionality is high and matrix is close to singular. On this account, this paper proposes a fast approach to solve the inverse matrix recursively. The process utilizes the earlier part of the recursive least squares (RLS) algorithm, which…

Recognition of table-form documents using high order correlation method

Ren-Jean Liou, Mu-Song Chen

Computer Vision and Pattern Recognition · · cited 1× · DOI ↗ · more like this ·

Table-form document recognition has many applications in office automation. An algorithm is proposed in the paper for automatic form processing. A high order correlation method was originally developed for point target detection in three-dimensional space. It computes the spatio-temporal cross-correlations of consecutive data to extract track information in series of images. It was shown that the method provides very good target detection and noise rejection rates. The technique can be easily reformulated in two-dimension for curve detection in regular images. Form processing is one of the good application examples of this technique. In the paper, we apply the high order correlation method to table-form document recognition. We also show that this process is relatively efficient and accurate in detecting and identifying all the lines in a document. In addition, the facts that the…

Novel Absolute Rotary Position Sensor Based on Combination of Gray Scale Pattern Disc

Ji Myong Choe, Nam Chol Yu, Nam Jin Ri +2 more

Computer Networks and Communications · Research Square · · cited 0× · DOI ↗ · more like this ·

Abstract Absolute rotary angular /position sensors play an important role in various applications and rapid development of new technologies requires further accurate measurement and control. In this paper, a novel, very simple, low-cost and high accurate absolute rotary angular/ position sensor is presented. The sensor operation is based on the combination of circular gradient gray scale and gray code pattern. A simple experiment is done in order to demonstrate proof of concept of proposed sensor. The experimental results show that the proposed absolute rotary angular/position sensor has excellent linear characteristics with accuracy below ±1° and resolution of 0.1° within the full measurement range from 0° to 360°. The proposed idea and experimental results can be helpful to design absolute rotary angular/position sensor to improve performance of it.

Using GPU to Shorten the Match Time of Rule Reasoning Based on Rete Algorithm

Jinyu Guo, Chi-Pan Hwang, Mu-Song Chen

Artificial Intelligence · · cited 0× · DOI ↗ · more like this ·

Usually, most of execution time of match-resolve-act reasoning cycle is spent in the matching phase. This issue has prevented the applicability of rule base systems. In this paper, the parallelism of α- and β-networks constructions in Rete algorithm have been realized on Graphics Processing Unit (GPU). It is possible to speed up the reasoning time 20 times faster than the current high performance multi-core processors. Furthermore, the parallel realization of Rete algorithm with GPU should be helpful on developing intelligent agents or data mining, using rule base systems.

A NEW FAMILY OF PARAMETRIC ISOPERIMETRIC INEQUALITIES

Revista De La, Xiang Gao, Jin-mu Song +1 more

Artificial Intelligence · · cited 0× · OpenAlex ↗ · more like this ·

Abstract. In this paper, we deal with isoperimetric-type inequalities for the closed convex curve in the Euclidean plane R2. In fact we establish a family of parametric inequalities involving some geometric functionals associated to the given closed convex curve with a simple Fourier series proof. Furthermore, we investigate some stability properties of such inequalities. 1. Introduction and

Influence of the Meshing Phase Difference of Planetary Gear Transmission System in 2. 5MW Wind Turbine Gearbox on Its Inherent Frequency

Xue Dan, Wang Chunxiu, Mu Song +1 more

Sociology and Political Science · DOAJ (DOAJ: Directory of Open Access Journals) · · cited 0× · OpenAlex ↗ · more like this ·

A translational- torsional coupling dynamics model of planetary gear transmission on the low speed shaft in 2. 5 MW wind turbine gearbox is built by using lumped- parameter method. Through comparing the inherent frequency of the three planetary gears under the conditions of the mesh stiffness are equal and taking the same average mesh stiffness with the conditions of the different mesh stiffness which are influenced by meshing phase difference,the conclusion that the influence of different combination of mesh stiffness under the influence of meshing phase difference on the low inherent frequency is less,but have much more influence on the high inherent frequency is obained. This analysis of inherent frequencies laid a foundation for dynamics analysis of the planetary gear transmission system in 2. 5MW wind turbine.