Showing papers presented at "International Conference on Intelligent Computation Technology and Automation in 2012"

Key Technologies and Applications of Internet of Things

Abstract: Internet of Things (IoT) has been paid more and more attention by the government, academe and industry all over the world because of its great prospect. This article introduced the concept of IoT, the basic properties and characteristics. Then it discussed the key technologies of IoT in detail taking the wire-less sensor network as an example, and the application of IoT was illustrated by its typical use in environment monitoring.

An Approach of Household Power Appliance Monitoring Based on Machine Learning

Abstract: Monitoring household electrical consumption by employing appropriate techniques is of great significance to sustainable development of human society. This paper proposes one approach of nonintrusive appliance load monitoring (NIALM) for electrical consumption managing. This approach can automatically monitor the house power consumption of individual devices. It employs multiple-class support vector machine (M-SVM) to recognize different appliances. The approach is consisted of two stages. In stage one, harmonic feature analysis is applied on current signal. In stage two, a trained classifier based on M-SVM is applied to identify different appliances. This paper presents the principle of this approach, the experiment results on real data, and discussions on performance comparison with other study of supervised classification for household power appliance monitoring.

An Unsupervised Anomaly Detection Approach for Spacecraft Based on Normal Behavior Clustering

Abstract: This paper presents a new unsupervised anomaly detection approach for spacecraft based on normal behavior clustering. This method takes as input a set of unlabelled historical telemetry data and automatically detects anomalies within the data. After these abnormal data are removed, the method constructs system normal behavior model based on normal data clustering. Then at run-time, it monitors the status of the spacecraft and detects any anomalies appearing in the real-time telemetry data by checking deviations from the normal behavior model. The experimental results show that the method is efficient and practical for anomaly detection of spacecraft system.

WSN Node Localization Algorithm Design Based on RSSI Technology

Abstract: Constrained by extraneous factors and power consumption of nodes, the traditional localization algorithms have some disadvantages in localization process, such as low localization accuracy, large amount of computation, etc. For this reason, the WSN node localization system based on RSSI technology is proposed. This algorithm improves node localization accuracy by conducting iteration refinement to the solutions, through quantification of the measurement of target parameters and geometric approach computation. This method can effectively reduce the communication overhead among network nodes, save network power and extend the life cycle of the network. The experimental results show that this method can complete complex computation process with fewer parametric variables, thus strengthening the robustness and stability of the entire network.

Analysis and Research of System Security Based on Android

Abstract: Android is a smart mobile terminal operating platform core on Linux. But due to its open-source software and programmable framework character, it leads the Android system vulnerable to get virus attacks. This paper has deeply researched from the Linux system security mechanism, Android-specific security mechanisms and other protection mechanisms. And on this basis, Android devices have achieved closely guarded on normal state. So that attackers can not use the kernel module or core library to get highest access permission and be attacked. Meanwhile, to further strengthen the security of Android devices, it enables them to properly handle the high-risk threat. This paper also strengthened intrusion detection system (HIDS) based on the host in order to detect malicious software and strengthen the Android system-level access control.

Storage Capacity of the Hopfield Network Associative Memory

Abstract: The Hop field model is a well-known dynamic associative-memory model. In this paper, we investigate various aspects of the Hop field model for associative memory. We conduct a systematic simulation investigation of several storage algorithms for Hop field networks, and conclude that the perceptron learning based storage algorithms can achieve much better storage capacity than the Hebbian learning based algorithms.

The Navigation of Mobile Robot Based on Stereo Vision

Abstract: In this paper, an overview of stereo vision, stereo image processing is introduced as well as a method of navigation based on stereo vision for mobile robot is provided. Making use of 3-d stereo reconstruction, the recognition of visible terrain in front of the mobile robot is solved for successive obstacle avoidance. The area-based stereo reconstruction algorithm which combines the pyramidal data structure and dynamic programming technique has been used for the recognition of the local environment. The vision system can thus be used to identify, locate and approach mechanical objects autonomously.

Question Answering System Based on Web

Abstract: This paper summarizes the classification, implementation and evaluation of question answering system (QA). QA is divided into four categories: chat robot, QA based knowledge base, QA retrieval system and QA based on free text. Web QA system is composed with question analysis module, Web information retrieval module and answer extraction module. In the last, TREC QA evaluation is outlined.

PID Controller Parameters Optimization Based on Artificial Fish Swarm Algorithm

Abstract: Parameter optimization is the core of PID controller design, for which, an approach based on artificial fish swarm algorithm(AFSA) is proposed. Experimental results verify the validity and feasibility of the proposed approach, and the results show that the approach is better than PSO algorithm, it can obtain the parameters with better control performances faster response speed, shorter adjustment time and smaller overshoot.

Energy Aware Resource Scheduling Algorithm for Data Center Using Reinforcement Learning

Abstract: More and more attention is paid for energy consumption aware and power control for data center with the emergency of energy crisis. The use of virtualization technology makes it possible for dynamic configuration of data center resources. The N:1 mapping visualization technology is employed to integrate many physical machines into an virtual resource pool to control resources centralized, and then reinforcement learning is applied to resource management and decision making for an uncertain task flow data center. Finally, an automatic resource control algorithm with energy consumption aware is proposed. This algorithm is implemented in the Cloud Sim platform to improve the energy consumption of the data center. The experimental results show that our algorithm can reduce about 40% of the energy consumption of the non-power-aware data center and reduce 1.7% of that of the greedy scheduling algorithm in data center.

Study on the Image Segmentation Based on ICA and Watershed Algorithm

Abstract: An image segmentation approach based on independent component analysis and watershed algorithm is proposed to complement medical image segmentation efficiently. Considering that the noise, uniform gray intensity and many low contrast adjacent areas in the medical images may cause the images over-segmented when using the watershed algorithm, so that the independent component algorithm is applied to reduce the noise' affection in image-segmentation and keep the images clear texture simultaneously, then watershed algorithm is used for image-segmentation. Compared to the method based on the wavelet filter and the classic watershed algorithm, the experiments display that this approach is efficient for images segmentation.

An Effective Automated Essay Scoring System Using Support Vector Regression

Abstract: In this paper, we introduce an effective automated essay scoring system. To implement the system, we extract several features, including the surface features such as the number of words in the essay, number of words longer than 5, and complex features such as grammar checking, sentences, whether the essay is off-topic, the similarity to full-score essays. We get the result of 86% precision given the two scores deviation and average deviation of 0.88 compared to human score on real CET4 data.

Automatic Video Text Detection and Localization Based on Coarseness Texture

Abstract: Video text recognition is crucial to the research in all video indexing and summarization and has been used in video semantics analysis. Video text detection and localization is important for video text recognition. In this paper we present a new approach to implement video text detection and localization. In the text detection, we perform motion detection in 30 frames to get the motion mask. Then we get the synthesized frame based multi-frame integration. Finally, we retrieve an edge map of the synthesized frame using wavelet, and we detect text rows based on statistic coarseness of the edge map. In text localization, we locate the accurate boundary of text row based on edge map. Then we use the multi-frame verification (MFV) to remove the candidate text regions with motion occurrence on 30 consecutive frames. Experimental results show that our method is able to obtain a robust performance in the variation of language, font, size, color.

Robust Tracking of the Flight Path Angle of Aircraft Based on UDE and IID

Abstract: Nonlinear robust tracking of the flight-path angle in the longitudinal channel of aircraft is a challenging control problem. The main difficulty results from its non-minimum phase characteristics and the unavoidable uncertainty and disturbances existing in the internal dynamics. To solve the problem, this paper presents a nonlinear control scheme based on an Uncertainty and Disturbance Estimator (UDE) and the causal Ideal Internal Dynamics (IID), taking an F-16 aircraft model as the illustrative example. A significant feature of the presented scheme is that it does not need any information about the uncertainties, and that it can be implemented in real time due to the causality of the inversion of the internal dynamics. A numerical simulation example is given to show the effectiveness of the scheme.

Research on 3-Phase 4-Wire VIENNA Rectifier Based on One Cycle Control

Abstract: In this paper, a 3-phase 4-wire star-connected three phase three-switch three-level rectifier (VIENNA) with unity power-factor-correction (PFC) based on one cycle control (OCC) is presented. The control technology does not require multipliers to scale the current reference and only need integrators. The advantage of this rectifier is that the switch voltage stress is one half of the total output voltage and it doesn't require multipliers and three-phase voltage sensors. The rectifier can be decoupled in topology to three single phase three-level PFC with output parallel-connected. Even if three phase lacks input, the rectifier still has excellent PFC characteristics. The proposed control concept is verified by simulation and experimental results.

Frequent Itemsets Mining in Network Traffic Data

Abstract: Many projects have tried to analyze the structure and dynamics of application overlay networks on the Internet using packet analysis and network flow data. While such analysis is essential for a variety of network management and security tasks, it is difficult on many networks: either the volume of data is so large as to make packet inspection intractable, or privacy concerns forbid packet capture and require the dissociation of network flows from users' actual IP addresses. In this paper, an algorithm for mining privacy preserving item sets is proposed. On the one hand, only maximal item set is considered, which reduces the number of item sets greatly. On the other hand, the intermediate mining results are encrypted for the security concern. Experimental results show that the proposed algorithm is both accurate and efficient.

An Improved Probabilistic Neural Network with GA Optimization

Abstract: Probabilistic Neural Network (PNN) was applied to prediction mainly. Over the traditional neural network, less time was cost by PNN in network architecture determining and training. But the smoothing parameter used in the estimation results was a user-defined constant. How to determine this parameter's value is a crucial problem in PNN. Combined with adaptive genetic algorithm (GA), a novel PNN was proposed. Crossover probability c p and mutation probability m p were employed to optimize the smoothing parameter. These two probability factors were adaptive, they would vary with the fitness of population. A reasonable search breadth and depth was finished by the GA with c p and m p. The prediction accuracy of PNN was increased because of the smoothing parameters optimization. The simulation experiments demonstrated that the improved PNN model was effective. The prediction precision was increased more.

Fast-ICA Based Blind Estimation of the Spreading Sequences for Down-link Multi-Rate DS/CDMA Signals

Abstract: This paper deals with the problem of blind estimation of the spreading sequences in multi-rate direct-sequence code division multiple access(DS-CDMA). The characteristic of the down-link Multi-Rate DS/CDMA signal is analyzed, and the down-link Multi-Rate DS/CDMA signal is modeled as the blind source separation pattern at first. Then an efficient iterative dimension-increased method based Fast-ICA algorithm, from small spreading period to bigger spreading period, is exploited to estimate the whole or part of period spreading sequences blindly, the 2-norm of vectors is employed to separate spreading sequence which with different spreading period, and obtain all users' spreading sequence blindly at lastly. Mathematical analysis and simulation results demonstrate the validity of the new method.

Analysis and Optimal Design of Relative Position Detection Sensor for High Speed Maglev Train

Abstract: The relative position detection sensor provides the information about position, speed, direction and magnetic phase for high-speed maglev train, by detecting the varying inductance of coils along the long stator with tooth-slot structure. Actual processing technique can't ensure the identity of two differential coils, which will make the direct current (DC) component of differential voltage drift and reduce the accuracy of sensor. In this paper, the effect of tooth-slot is equivalent to varying relative permeability. The wave distortion of differential voltage is analyzed, and algebraic expression of DC drift is derived. Based on that, some optimized methods are proposed. Experiment proves that they can effectively eliminate DC drift and improve the accuracy of sensor.

Fault Diagnosis Based on Wavelet Neural Network

Abstract: As wind power generation is a complicated nonlinear time-varying system, it's hard to extract effective fault feature. A novel arithmetic that combines modified LDB (Local Discriminant Basis) algorithm and SOM-BP network is proposed in this paper for fault diagnosis and location. First original fault features are extracted by improved LDB algorithm, then these fault features are mapped into a new feature space with high class separability via SOM (Self-Organizing Feature Map) nonlinearly transform, finally BP is used as a nonlinear classifier to implement fault diagnosis and location.

The Utility Frequent Pattern Mining Based on Slide Window in Data Stream

Abstract: In traditional study of mining data stream, each item in the data stream is of equal importance. However, in practice, each item has a different significance, which is known as utility. This paper combines frequent mining item sets with utility and proposes an efficient algorithm for utility frequent pattern mining (UFPM). It combines bitmap with tree structure that can store and update the pattern of data stream quickly and completely by scanning only once. The algorithm generated by lexicographic order, proposes a novel tree U-tree and makes convenience for pattern updating and user reading. With a pattern growth approach in mining, the algorithm can effectively avoid the problem of a mass candidacy generation by level-wise searching. The experiments results show that our algorithm which is in high efficiency and good scalability outperforms the existing analogous algorithm.

LED lighting control system in tunnel based on intelligent illumination curve

Abstract: The LED tunnel lighting system is presented based on field bus RS485. An expert system is used to calculate the demand illumination curve of the tunnel. It is based on the outside illumination of the tunnel, car flow in traffic and car speed. And the tunnel illumination is controlled by the given illumination curve. It makes the bright and dark adaptation process of driver?s eyes in tunnel more natural and smooth, and safeguards the driving. To convert the illumination curve to the control instructions to every LED light, a feedforward and feedback control strategy based on the relationship between light flux and brightness in tunnel is presented. As well as the power energy of tunnel lighting system is saved.

Design the Adaptive Noise Canceller Based on an Improved LMS Algorithm and Realize It by DSP

Abstract: An adaptive noise canceller based on an improved LMS (least-mean-squares) algorithm has been designed by DSP. The algorithm is introduced and simulated in MATLAB, then realized on the digital signal processor TMS320C 6713. A 50HZ sine wave, used as the referenced input, traces the noise of signal until it is eliminated. Then the expected signal has been obtained. The experiment results indicate that the adaptive noise canceller can remove the 50HZ noise of signal conveniently and effectively, so it has the ideal filter characteristics of narrow-band.

Multi-channel Access Algorithm with Channel State Information Unknown

Abstract: The partially observed Markov decision process (POMDP) is very important in cognitive radio network. We can formulate the opportunistic spectrum access in the framework of POMDP. However, no attention was paid to the algorithm in the presence of unknown environment. For the rapidly changing cognitive network with unknown environment, we firstly establish the channel state model based on MLA and prove the MLA convergence. We consider a more practical scenario where the exact channel state transition probabilities are unknown and the sensing error exists. Secondly, based on the model has been established, we proposed the multi-channel access algorithm in one slot and put forward discount factor to achieve the reasonable tradeoff between the cost of sensing overhead and access. The simulation results showed that the performance of proposed multi-channel access algorithm outperform the single-channel optimal access.

Search the Optimal Preference of Affinity Propagation Algorithm

Abstract: In order to improve the clustering quality of the Affinity Propagation algorithm further and get more accurate number of clusters, this paper proposed a novel algorithm based on the Particles Swarm Optimization, which used In-Group Proportion index as fitness function to search the optimal preference of Affinity Propagation algorithm. Experimental results show that the predicted results had been tested with the novel proposed algorithm and the better results had been achieved.

IEMS: An Intelligent Environment Monitoring System of Server Room

Abstract: In order to achieve low cost, low complexity, more reliable and more stable intelligent room environment monitoring system, in this paper, the designed system is composed of the surveillance mainframe and several sensors based on C/S (Client/Server) architecture through the reliable RS-485 network, which will construct an effective and feasible hardware platform to complete the real-time monitoring of the current room environment. In addition, by adopting the B/S (Browser/Server) architecture, users are able to grasp the current monitoring information at any time and place through browsing the Web pages. In practice, the environment monitoring system based on RS-485 network and the development of the software by using the Microsoft Visual C++ 6.0 platform were discussed. The realization of the intelligent room environment monitoring system based on B/S architecture provides a well feasible scheme for the current computer intelligent control field. The successful application in several main server rooms of Nan tong University sufficiently proves the practical feasibility of the project.

A Nonovershooting Pitch-Angle Regulator for Aircraft with Conventional Aerodynamic Configuration

Abstract: The problem of designing a nonovershooting controller for general high-order linear time-invariant plants is still open. In this paper, we propose a new technique to design nonovershooting pitch-angle regulator for conventional aircraft based on the forth-order model of longitudinal motion. The regulator is composed of the error-feedback and the feed forward control terms. Two forms of the regulator are presented which apply to the state-feedback and the output feedback cases, respectively. Simulation examples are given to show the effectiveness of the regulator.

An experimental study of college teaching based on Matlab software

Abstract: Matlab is the Math Works company business mathematics software, mainly applied in engineering design, calculation, the control signal processing and communication, image processing and signal detection, modeling design and analysis, etc In the above application fields Matlab has powerful functions, in order to cultivate students' scientific spirit and practical ability and innovative consciousness, pay attention to the problems of engineering practice, and modeling ability training in colleges, so the different specialized has opened the Matlab foundation course The author in teaching practice and experiment, and based on the investigation of different professional open courses and thinking, the idea is: the numerical calculation method and statistics, the optimization module and software mat lab, the typical case study, combine to optimize the teaching effect, enhance students' learning interest, produce good effects on the subsequent courses and the work

The Application of Web Data Mining in the Electronic Commerce

Abstract: With the increasing of the information on Internet, more and more electronic data are appearing. Then, how should we immidiately discover useful knowledge and improve the utilization rate of information without being confused in the sea of information? Data mining come up with a new way of dealing with such problem. This paper sets force web data mining sources in e-commerce, the flow process and some techniques in dealing with web data mining. Finally, analyses the functions of web data mining used in e-commerce.

Network Energy Optimization for MPI Operations

Abstract: High-performance computing (HPC) systems consume huge amount of power, while bringing about higher operational costs, lower reliability. Power optimization has become an important target for HPC system design. Network energy optimization is significant for the energy optimization of the whole system. In this paper, the dynamic energy optimization method of network with MPI communication operation is presented, which can find MPI operation implementation with the least network energy by accurate dynamic energy estimation, and reduce the energy consumption of network. Taking barrier as an example, we validate the correctness and usefulness of the dynamic energy optimization method of network.