Loading...
ISSN 1000-1239 CN 11-1777/TP

Table of Content

15 January 2007, Volume 44 Issue 1
Paper
A Time Self-Decay Trust Management Algorithm for P2P Computing Security
Shi Zhiguo, He Yeping, and Zhang Hong,
2007, 44(1):  1-10. 
Asbtract ( 497 )   HTML ( 1)   PDF (566KB) ( 658 )  
Related Articles | Metrics
A novel quantitative network trust evaluation algorithm for peer-to-peer computing security system is proposed. The main features of the proposed model include time decay function and entity union function. First of all, the current representative algorithms of network trust evaluation are summerized and classified systematically. And the significant research fields of the related algorithms are categarized and the definition of trust related terminology is given. In this paper, an algorithm with trust time correct function, domain trust correct function and the definition of accuracy is constructed. Four features of the algorithm: time decay feature, history experience related feature, new entity award feature and union feature, are proved. At the same time, a natural trust decay curve is described and eight typical domain features are given. The experiments are also designed to evaluate the correctness and performances of the proposed algorithm, the results of the experiments are compared with Azzedin algorithm, and the results prove the sound performance and correctness of the algorithm. Finally, some related future research fields of the paper are pointed out.
Research on Privacy Preservation Mechanism for Credentials and Policies in Grid Computing Environment
Qiang Weizhong, Zou Deqing, and Jin Hai
2007, 44(1):  11-19. 
Asbtract ( 400 )   HTML ( 0)   PDF (458KB) ( 650 )  
Related Articles | Metrics
Privacy preservation of access control policy and credential for grid entities is an important aspect of grid security, whose importance has been increasingly highlighted with the wide application of grid technology. Using the theory of secure function evaluation, a circuit composition mechanism is proposed which adapts to the expression of composite policy, and a policy evaluation protocol is proposed based on oblivious transfer mechanism and scrambled circuit evaluation protocol; Using the theory of homomorphic encryption, an equility test protocol for attributes is proposed; Based on the two protocols, a privacy preserving protocol is proposed for the access control policies and credentials. Analysis shows that the proposed scheme can provide complete privacy preservation for the attributes of policy and credential, and also can avoid circle dependence problem caused by the related method.
Flexible Resource Capacity Reservation Mechanism for Service Grid Using Slack Time
Hu Chunming, Huai Jinpeng, and Wo Tianyu
2007, 44(1):  20-28. 
Asbtract ( 442 )   HTML ( 2)   PDF (571KB) ( 490 )  
Related Articles | Metrics
Resource reservation and allocation is commonly used to provide QoS guarantee for grid services. However, the reservation request with fixed parameters may introduce resource capacity fragment problem caused by the peaks of resource utilization. In this paper, a flexible capacity reservation mechanism is proposed, in which slack time is added into the reservation scenario to get more flexibility of resource reservation. Beside this, an algorithm is introduced to support the reservation and admission control with slack time. Four strategies are proposed according to the existing scheduling algorithms. The algorithm is implemented in CROWN service grid middleware. Evaluation result shows that better performance can be achieved by using slack-enabled algorithm, and the min-min-based strategy performs better.
Research on a Grid Resource Allocation Algorithm Based on MAS Non-Cooperative Bidding Game
Jiang Weijin and Wang Pu
2007, 44(1):  29-36. 
Asbtract ( 394 )   HTML ( 0)   PDF (381KB) ( 590 )  
Related Articles | Metrics
With MAS (multi agent system) coordinated technology and market bidding game rules, a grid resource allocation model based on market economy is introduced, which could show the relation between supply and demand. The model could make use of the computing and negotiating power of the consumer well and could consider consumer behavior fully. It makes the resource application and assignment of the consumer have the higher rationality and effectiveness. In the meantime, the utility function of consumer is given; the entity and the uniqueness of Nash equilibrium in the resource allocation game and the Nash equilibrium solution are discussed. A grid resource allocation algorithm is designed based on the distilled resource allocation model. The results of emulation test show that the resource allocation algorithm could be a reference to resources of consumers and a standardization of consumers' bidding, which makes the allocation of the whole network resource tend to be more reasonable.
An Effective Approach to Automatic Semantic Annotation for Service Resources
Yang Yanping and Tan Qingping
2007, 44(1):  37-43. 
Asbtract ( 365 )   HTML ( 0)   PDF (475KB) ( 621 )  
Related Articles | Metrics
In service grid, the implementation of distributed computing lies in the solution to the problem of how services interchange information. It is imperative that services interfaces be described in a machine understandable way, which can provide a semantic foundation for the dynamic discovery and composition of services. The technique of service semantic annotation satisfies this requirement, which refers to annotate service description using machine understandable metadata from shared domain ontologies. In this paper, an effective approach to automatic semantic annotation of services is presented. The process of service annotation is divided into two phases of ontology annotation phase and concept annotation phase. Emphasis is put on the solution to the problem of ontology annotation and a corresponding algorithm is presented based on machine learning. The experiments on the service resources in real world show that this algorithm is highly practical.
Research and Application of Multi-Objective Aircraft Optimization System Based on Grid
Chang Yan, Liu Xu, Cheng Wenyuan, Xie Xianghui, and Cui Degang
2007, 44(1):  44-60. 
Asbtract ( 422 )   HTML ( 0)   PDF (423KB) ( 436 )  
Related Articles | Metrics
A common computing model of genetic algorithm is proposed for aircraft multi-objective topology optimization. According to this model based on P2P computing, the distributed computational resources are pooled into a high performance-computing environment and providing uniform resource service and visible user using interface. The multi-objective optimization grid (MOOG) is put forward to resolve the composite material multi-objective topology optimization in aircraft design and discussed in detail are the grid system architecture and the improved multi-objective genetic algorithm. Finally, the solid experimental results of optimizing one type of high aspect ratio wing show that the MOOG has reduced the computation time and has good parallel acceleration.
Hierarchical Access Authentication Method in Mobile IPv6 Networks
Zhang Hanwen, Zhang Yujun, Tian Ye, Xiao Wenshu, and Li Zhongcheng
2007, 44(1):  51-57. 
Asbtract ( 565 )   HTML ( 0)   PDF (452KB) ( 471 )  
Related Articles | Metrics
Access authentication is very important for mobile IPv6 networks. In this paper, hierarchical access authentication method in mobile IPv6 (HAMIPv6) is proposed. The method can accomplish access authentication and mobile registration while reducing exchanges between mobile node and home domain by utilizing authentication vector and message piggyback. The method simplifies handoff authentication process by reducing handoff delay and cost. Also the method supports mutual authentication between terminal and network. Performance evaluation shows that the proposed method is better than the others, especially in the conditions that MN is far from home domain and move frequently within a domain.
Self-Similarity Parameter Estimation and Scaling Properties Analyses of Aggregated Wireless Traffic
Yu Qin and Mao Yuming
2007, 44(1):  58-64. 
Asbtract ( 290 )   HTML ( 0)   PDF (422KB) ( 419 )  
Related Articles | Metrics
Hurst parameter is important to measure the self-similarity degree and bustiness of network traffic. Based on R/S statistic, variance-time plots and periodogram-based analysis methods, an optimal linear regression wavelet model is proposed to perform the accurate, quick and effective estimation of the Hurst parameter in wavelet field. The aggregating process and statistic characteristics of multiple input traffic sources in WLAN are studied. Simulation results compare the Hurst parameter estimation values of self-similar WLAN traffic by using the above statistical approaches. Furthermore, theoretical analyses and simulation results demonstrate that the aggregated traffic at WLAN also exhibits self-similarity, which actually intensifies rather than diminishes burstiness. These results can be very useful for accurate modeling, traffic control, resource allocation optimization and performance improvement of WLAN.
Enforcing Admission Control Using Admission-Time-Ratio and PI Controller
Jiang Ying, and Meng Dan
2007, 44(1):  65-70. 
Asbtract ( 427 )   HTML ( 0)   PDF (378KB) ( 1044 )  
Related Articles | Metrics
In this paper a session-based admission time-ratio feedback control mechanism is presented for improving the Web QoS of commercial applications. The mechanism combines a novel session-based admission control (AC) policy with a PI controller. The admission control strategy manages the amount of requests that the server can accept, by dynamically changing the admitting time interval in the control period. During the admitting time interval all requests are admitted. The admission time ratio is naturally defined as the ratio of the admitting time interval in the control period. The novel admission control mechanism is agile and more easily implemented than adjusting the queue length. Moreover, PI controller designed using system identification and control theory makes this scheme robust and auto-tunable. The simulation results of Tomcat server provided in the paper demonstrate that it can improve the performance effectively and ease the resource control and QoS control.
A Cluster-Based Two Level TDMA Scheduling Protocol for Large Scale Wireless Sensor Network
Gong Haigang, Liu Ming, and Wang Xiaomin
2007, 44(1):  71-77. 
Asbtract ( 538 )   HTML ( 0)   PDF (394KB) ( 470 )  
Related Articles | Metrics
One of the key problems for wireless sensor network (WSN) is the design of medium access control (MAC) protocol, as MAC is the major consumer of sensor energy. TDMA-based MAC protocol is inherently collision free but it has poor scalability. cluster-based TDMA protocol proves to be more scalable than traditional TDMA protocol while it introduces inter-cluster interference problem. A two level TDMA scheduling (TLTS) protocol is presented to solve the problem, in which the first scheduling ensures that neighboring clusters collecting their member's data during different frame in order to avoid inter-cluster interference and the second scheduling schedules members of cluster sending their data to their respective cluster head during different slots to avoid intra-cluster interference. Simulation results show that TLTS performs better than HEED when node density turns higher. TLTS is more practical than HEED which uses CDMA code to avoid inter-cluster interference for deploying large scale WSN.
P2P-VCG: A Game Theory Proposal for Bandwidth Allocation
Huang Guanyao, Hong Peilin, and Li Jinsheng
2007, 44(1):  78-84. 
Asbtract ( 583 )   HTML ( 2)   PDF (382KB) ( 634 )  
Related Articles | Metrics
Existing P2P systems can NOT achieve both individual rationality and system efficiency for the lack of efficient BA (bandwidth allocation) and accounting proposal. In this paper E-currency and game theory are brought to P2P and the difficulty of BA in it is discussed. The utilities of all customers are used as a benchmark and two main novelties are brought forward: the first is that the VCG algorithm is brought in to P2P, and the second is that emphasis is put on how to choose bandwidth by clients. This kind of client-based bandwidth allocation proposal is much more suitable for P2P. The detailed BA & accounting algorithm is presented and its practical application is discussed.
End-to-End Available Bandwidth Measurement Based on Queueing Analysis
Huang Guowei, Wu Gongyi, and Xu Jingdong
2007, 44(1):  85-91. 
Asbtract ( 342 )   HTML ( 1)   PDF (373KB) ( 579 )  
Related Articles | Metrics
End-to-end available bandwidth is an important factor of the performance of network, most research works focus on the self-congestion methods by far. But the self-congestion methods may induce congestion on the end-to-end path and then influence the traffic on the path seriously. In this paper, a model-based method is proposed to avoid the drawback of the self-congestion methods. Firstly, it treats Internet as a queueing system which has a single server and serves two types of traffic: the probing traffic and the background traffic, and then builds a relation between the load of the end-to-end path and the coefficient of variation of the interdeparture time of the probing traffic by analyzing the departure process of the probing traffic. Finally, the end-to-end available bandwidth is obtained from this relation. Compared with the self-congestion methods, the proposed method will not induce congestion and is more accurate.
Multi-Step Prediction of VBR Video Traffic Based on Mutifractal Analysis
Wang Shenghui and Qiu Zhengding
2007, 44(1):  92-98. 
Asbtract ( 496 )   HTML ( 0)   PDF (404KB) ( 527 )  
Related Articles | Metrics
Real-time prediction of video source traffic is an important step in network resource management and end-to-end quality-of-service (QoS) strategies. However, together with the long-range-dependence (LRD) and the traffic non-stationarity, it suggests that conventional prediction tools, which only use short-range dependence (SRD), are not appropriate for VBR video traffic prediction. In this paper, by analyzing the correlation structure of multifractal coefficients, the original LRD trace can be converted to a series of SRD sequence in multifractal domain. Because the LRD feature of trace is used, the multi-step performance of proposed multifractal model is much better than traditional methods.
A Fuzzy Classification of Web Pages Based on the Transposition-Learning Rule
Zhang Maoyuan, Zou Chunyan, and Lu Zhengding
2007, 44(1):  99-104. 
Asbtract ( 372 )   HTML ( 0)   PDF (364KB) ( 424 )  
Related Articles | Metrics
When the overlap of categories is excessive, the accuracy of Web page classification decreases. In order to classify the Web pages accurately, a framework of fuzzy classification of Web pages is presented, to give a mechanism of combining the human knowledge on Web page classification by a member function. Then a general learning rule of the coefficients is proposed. The Lyapunov function is used to analyze the convergence of the general learning rule, and it is proved in theory that the general learning rule has the inherent factor which adjusts the coefficient values to gain the minimum error. On the basis of theoretic convergence analyses of a single-coefficient learning algorithm, a transposition rule is proposed, which is applied to the single-coefficient learning algorithm to gain quick convergence speed in the phase of coefficient learning. It is shown that both from the theoretic deduction of the learning convergence and from the experiment result, the fuzzy classification of Web pages is an efficient method.
An Efficient Join Algorithm for Data on Tertiary Storage
Liu Baoliang, Li Jianzhong, and Gao Hong
2007, 44(1):  105-110. 
Asbtract ( 382 )   HTML ( 0)   PDF (365KB) ( 491 )  
Related Articles | Metrics
The online use of tertiary storage system provides a costly and feasible scheme for massive data management. In order to extend database system to manage data on tertiary storage, the relational operators, especially the join operation, are one of the key problems that must be resolved. An efficient tertiary join algorithm is presented. Experimental results show that the join algorithm is better than previous ones in performance and scalability. The join algorithm can greatly reduce the I/O cost compared with previous ones. When the data amount is huge, the performance of the join algorithm is even better than that of disk join. The result of the paper shows that tertiary storage can be used to manage massive data as well as disks, solving the key problem of storing and querying massive data.
An Index for Supporting XML Structural Join Efficiently and Effectively—CATI
Yu Yaxin, Wang Guoren, Zhang Haining, and Li Jianxin
2007, 44(1):  111-118. 
Asbtract ( 401 )   HTML ( 0)   PDF (507KB) ( 385 )  
Related Articles | Metrics
Evaluating structural joins is a crucial operation for performance of XML queries. Proposed in this paper is a novel index CATI, namely, compact ancestor tree index, which can support XML structural join effectively. The basic idea of CATI is to extract ancestor node occurrences from a given ancestor-descendant query (A-D query) pattern or a twig query pattern to establish a compact ancestor tree, and all descendant occurrences of an ancestor node will be linked behind the direct ancestor node in the ancestor tree. Since the stack-tree based join algorithm is a widely used technique due to its excellent join efficiency, the performance of CATI is compared with stack-tree for A-D query and twig query via extensive experiments on a real XML test database. The experimental results show that the performance of the CATI based structural join has significant improvement gains over the stack-tree based join.
DETECTOR: A Universal On-Line Keyword Search System over Relational Database
Cai Hongyan, Yao Jiali, and Wang Shan
2007, 44(1):  119-125. 
Asbtract ( 416 )   HTML ( 0)   PDF (450KB) ( 543 )  
Related Articles | Metrics
Now keyword search over relational database is becoming a hot research topic in both database and IR area. There are two kinds of popular keyword search systems, while neither of them can give accurate results with high speed. The on-line system can obtain the latest search results while the search speed is low; The off-line system speeds the query processing by means of preprocessing, but the disadvantage lies in expensive cost needed to preprocess again once the database is updated. An incremental update solution is designed based on an off-line system DIRECTOR to synchronize the updating of data graph with the database. It improves the real-time characteristic of the querying without decreasing its effectiveness. Another contribution of this research is that a universal pre-processing module is used to facilitate pre-processing even with limited knowledge of database schema.
Sequence Pattern Mining Without Duplicate Project Database Scan
Zhang Kun and Zhu Yangyong
2007, 44(1):  126-132. 
Asbtract ( 556 )   HTML ( 0)   PDF (413KB) ( 770 )  
Related Articles | Metrics
Sequence pattern mining has broad applications in the analysis of Web click streams, the prediction of disasters and the pattern discovery of DNA and protein sequences. PrefixSpan, which is based on frequent pattern growth approach, is currently one of the fastest algorithms towards this target. However, PrefixSpan will produce huge amount of duplicated project databases in mining dense data sets and long sequence patterns. In order to overcome this drawback, a random algorithm named SPMDS is proposed. The algorithm avoids scanning duplicated project databases by checking evidences computed by exercising one way hash function such as MD5 to pseudo projections of project databases, and also improves its performance by simplifying the search in the project tree using some necessary conditions. Both experiments and analyses show that SPMDS is better than PrefixSpan.
Multi-Template ASM and Its Application in Facial Feature Points Detection
Li Ying, Lai Jianhuang,, and Yuen Pongchi5
2007, 44(1):  133-140. 
Asbtract ( 481 )   HTML ( 0)   PDF (659KB) ( 547 )  
Related Articles | Metrics
Active shape model is one of the most popular methods for facial image alignment. To improve its accuracy in facial feature points detection in facial images with various expressions and under nonlinear illumination, a multi-template ASM method, which integrates Gabor features and combines local ASM and global ASM, is proposed in this paper. Human faces often have various kinds of expression, such as smiling, surprising, anger, being confused and so on. For eyes, it can be open or closed, while for mouths, it can be smiling, widely open, tightly closed and “O” shape with surprise. Such two different kinds of eye states and four various kinds of mouth states give great nonlinear transformation to their ordinary shapes. Therefore, they can’t be processed simply in a single model. In this paper, two local templates for eyes, four local templates for mouths, and some global templates for the whole face are created. Under the assumption that the locations of two inner corners of eyes and two outer corners of mouths are known, in the new method, the approximate area for the two eyes is found first, and then the state of eyes in this area is determined by using the eyes local templates and the Hausdorff distance. Similarly, the state of the mouth can also be known. Finally the whole face contour is searched by global template corresponding to the estimated eyes and mouth states. The experiment shows that the method can achieve much higher detection rate than the standard ASM method.
Salient Building Detection Based on SVM
Qu Yanyun, Zheng Nanning, Li Cuihua, Yuan Zejian, and Ye Congying
2007, 44(1):  141-147. 
Asbtract ( 494 )   HTML ( 0)   PDF (544KB) ( 546 )  
Related Articles | Metrics
This paper focuses on detecting salient buildings in a scenery image. A method based on bottom-up attention mechanism is proposed to detect salient buildings. Firstly, Haar wavelet decomposition is used to obtain the enhanced image which is the sum of the square of LH sub-image and HL sub-image. Secondly, the enhanced image is projected in the vertical direction to obtain the projection profile, and building candidates are separated from the background based on multi-level thresholding. Thirdly, the structure statistic features of buildings are extracted based on Sobel operator. The feature vector is formed by the number of long horizontal edges and that of vertical edges. Finally, linear support vector machines are used to classify buildings and the others. The proposed approach has been experimented on many real-world images with promising results.
Research on Degree of Fitting Implicit Polynomial Curves and Surfaces
Wu Gang
2007, 44(1):  148-153. 
Asbtract ( 589 )   HTML ( 0)   PDF (292KB) ( 594 )  
Related Articles | Metrics
How to choose suitable degree of fitting implicit polynomial curves and surfaces is a key problem when the objects are described and recognized by implicit polynomials, but the problem is not solved in theory. A theorem is presented to determine the degree of fitting implicit polynomials based on the object boundary features, and the formula for the degree of fitting implicit polynomial is given. From the theory, the algorithm for determination of the degree of implicit polynomials is followed: detecting the stationary points in the object boundary based on its features, and then calculating the minimum degree according to the formula derived from the theorem. In addition, the theory can be extended to three-dimensional object. Finally, experiments are given to illustrate the efficiency and operation of the algorithm.
Processor Fault-Tolerance Technology Research and Prospect
Fu Zhongchuan, Chen Hongsong, Cui Gang, and Yang Xiaozong
2007, 44(1):  154-160. 
Asbtract ( 566 )   HTML ( 0)   PDF (373KB) ( 716 )  
Related Articles | Metrics
As technology advances and silicon feature size shrinks, computer system is faced with inevitably increasing risks of transient-fault susceptibility. Accordingly, processor dependability and trustworthiness have become the major problems of the application systems. Recently, much work has been done at different levels to accomplish fault-tolerance in processor systems against transient-faults. In this paper, a novel and comprehensive taxonomy of the latest processor fault-tolerance researches is put forward. Based on this taxonomy, the techniques of incorporating fault-tolerance, especially transient fault-tolerance, in modern processor systems at different levels are reviewed. Some important processor fault-tolerance architectures and representative researches are also briefly introduced and analyzed. Finally, some valuable advice and possible trends in processor fault-tolerance researches are proposed, hoping they will benefit related researchers.
Dynamic Binary Translation and Optimization
Li Jianhui, Ma Xiangning, and Zhu Chuanqi
2007, 44(1):  161-168. 
Asbtract ( 1060 )   HTML ( 4)   PDF (414KB) ( 1036 )  
Related Articles | Metrics
Dynamic binary translator is a just-in-time compiler, which translates the instructions of source architecture to the instructions of target architecture when an application is running. The technology enables the application compiled for source architecture running on top of target architecture without recompilation. This paper begins with the basic framework of dynamic binary translator, and then gives an overview of several leading dynamic binary translators. After that, it has a deep discussion about key challenges of the dynamic binary translator, including supporting precise exception in optimized code, mapping source architectural context to target architectural context, translating self modifying code, reducing translation overhead, and dynamic optimization using profiling data. The paper ends with the hot research topics and possible usage models of the dynamic binary translation technology.
Adaptive Stack Cache with Fast Address Generation
Huan Dandan, Li Zusong, Wang Jian, Zhang Longbing, Hu Weiwu, and Liu Zhiyong
2007, 44(1):  169-176. 
Asbtract ( 389 )   HTML ( 0)   PDF (544KB) ( 481 )  
Related Articles | Metrics
With the processor-memory performance gap continuing to grow, the performance of memory access becomes the major bottleneck of the performance improvement for modern microprocessors. Adaptive stack cache with fast address generation policy is proposed by investigating memory access behavior of programs. Adaptive stack cache with fast address generation policy decouples stack references from other data references, improves instruction-level parallelism, reduces data cache pollution and decreases data cache miss rate. Stack access latency can be reduced by using fast address generation scheme proposed here. Adaptive stack cache with fast address generation policy can also avoid unnecessary memory traffic. Stack cache can be disabled adaptively, when it is overflown. It can also be applied to multithread scheme by adding thread identifier. The results obtained indicate that about 25.8% of all memory reference instructions in SPEC CPU2000 benchmarks are executed in parallel by adopting adaptive stack cache with fast address generation. On average 9.4% data cache miss is reduced. The performance is improved significantly. The average IPC speedup is 6.9%.
A Multi-Objective Differential Evolution Algorithm Based on Max-Min Distance Density
Zhang Libiao, Zhou Chunguang, Ma Ming, and Sun Caitang
2007, 44(1):  177-184. 
Asbtract ( 604 )   HTML ( 0)   PDF (484KB) ( 624 )  
Related Articles | Metrics
Differential evolution is a simple and powerful globally optimization new algorithm. It is a population-based, direct search algorithm, and has been successfully applied in various fields. A multi-objective differential evolution algorithm based on max-min distance density is proposed. The algorithm proposed defines max-min distance density and gives a Pareto candidate solution set maintenance method, ensuring the diversity of the Pareto solution set. Using Pareto dominance relationship among individuals and max-min distance density, the algorithm improves the selection operation of differential evolution, ensures the convergence of the algorithm, and realizes the solution of multi-objective optimization problems by differential evolution. The proposed algorithm is applied to five ZDT test functions and two high dimension test functions, and it is also compared with other multi-objective evolutionary algorithms. Experimental result and analysis show that the algorithm is feasible and efficient.
Independent Component Analysis Based on Sliding Window
Wu Xiaopei, Ye Zhongfu, Guo Xiaojing, Zhang Daoxin, and Hu Renjun
2007, 44(1):  185-191. 
Asbtract ( 559 )   HTML ( 1)   PDF (424KB) ( 578 )  
Related Articles | Metrics
Independent component analysis (ICA) has become a hotspot in signal processing area, and its computation algorithms and application are widely studied recently. In this paper, a sliding window ICA algorithm is studied to deal with time variant mixing model that the traditional ICA algorithms fail to work. For solving the problem caused by the indeterminacy of ICs in applying sliding window ICA algorithm, the value of kurtosis is employed as the index to sort the independent components at each window position. This idea is proved to be effective in most cases. In addition, a recursive learning rule of separation matrix based on sliding window is given, which can reduce the computation load of the algorithm obviously. The selection of window length is also discussed in this paper. Furthermore, the separation performance of the proposed algorithm are compared with the batch ICA algorithm. Experiment results show that the proposed algorithm can work well in the time variant mixing model and can be used for online blind source separation and dynamic independent component analysis.