ISSN 
2015 Vol. 44, No. 2
2015, 44(2): 163-171.
doi: 10.3969/j.issn.1001-0548.2015.02.001
Abstract:
The human mobility pattern in ordinary life is influenced by various factors. Two datasets of location information reported by mobile phones are utilized to analyze the spatial mobility pattern of check-in behavior. Our research focuses on the impacts of the numbers of visited locations, average jump distances, radiuses of gyration and most frequent visited locations on the predictability of check-in behavior. It is found that the check-in behavior shows certain memory effect. The numbers of visited locations and the visiting patterns to the most frequent visited locations have more significant influence on the predictability and regularity of check-in behavior, meanwhile the impacts of radiuses of gyration and the average jump distances are obviously unremarkable.
The human mobility pattern in ordinary life is influenced by various factors. Two datasets of location information reported by mobile phones are utilized to analyze the spatial mobility pattern of check-in behavior. Our research focuses on the impacts of the numbers of visited locations, average jump distances, radiuses of gyration and most frequent visited locations on the predictability of check-in behavior. It is found that the check-in behavior shows certain memory effect. The numbers of visited locations and the visiting patterns to the most frequent visited locations have more significant influence on the predictability and regularity of check-in behavior, meanwhile the impacts of radiuses of gyration and the average jump distances are obviously unremarkable.
2015, 44(2): 172-177.
doi: 10.3969/j.issn.1001-0548.2015.02.002
Abstract:
In this paper, a novel behavior spreading model is proposed, which incorporates the reciprocal interaction and collaborative spreading of information and behavior on social networks. Various network parameters and transmission factors, describing the interaction of information and behavior, the memory and timeliness of information, the social reinforcement and so on, are incorporated into the spreading process. The simulation results indicate that the involved parameters have important influence on the transmission range and speed, but their functions are varied according to different circumstances. Our research shows that the behavior spreading on social networks not only possesses some common laws similar to the transmission of information and diseases, but also exhibits its unique characteristics.
In this paper, a novel behavior spreading model is proposed, which incorporates the reciprocal interaction and collaborative spreading of information and behavior on social networks. Various network parameters and transmission factors, describing the interaction of information and behavior, the memory and timeliness of information, the social reinforcement and so on, are incorporated into the spreading process. The simulation results indicate that the involved parameters have important influence on the transmission range and speed, but their functions are varied according to different circumstances. Our research shows that the behavior spreading on social networks not only possesses some common laws similar to the transmission of information and diseases, but also exhibits its unique characteristics.
2015, 44(2): 178-182.
doi: 10.3969/j.issn.1001-0548.2015.02.003
Abstract:
Based on the features of international trade network, this paper proposes a basic model of bootstrap percolation to simulate cascading process of crisis propagation. Two general models, proportion model and GDP model, are derived from the basic model. According to different transmission modes, the affected countries are classified into four types and specific propagation processes are explored. Simulation results show that the vast majority of affected countries occur phase transition phenomenon with the threshold Ω. Comparing the threshold Ω in phase transition point of different sources countries, we can obtain their relative crisis consequences. For a given parameter Ω, we discuss the specific transmission and their scope. Results reflect the rationality of this model.
Based on the features of international trade network, this paper proposes a basic model of bootstrap percolation to simulate cascading process of crisis propagation. Two general models, proportion model and GDP model, are derived from the basic model. According to different transmission modes, the affected countries are classified into four types and specific propagation processes are explored. Simulation results show that the vast majority of affected countries occur phase transition phenomenon with the threshold Ω. Comparing the threshold Ω in phase transition point of different sources countries, we can obtain their relative crisis consequences. For a given parameter Ω, we discuss the specific transmission and their scope. Results reflect the rationality of this model.
2015, 44(2): 183-189.
doi: 10.3969/j.issn.1001-0548.2015.02.004
Abstract:
The current cognitive radio (CR) engine based on genetic algorithm usually adopts a weight-method to change multi-objective into a single objective, which may miss optimal solutions and reduce the efficiency of engine. This paper proposes an adaptive multi-objective immune genetic algorithm with vaccine injection (VAMIGA) to resolve this problem. The vaccine injection could optimize the decision result and convergence speed by saving and recycling the excellent genes. Compared with the strength Pareto evolutionary algorithm (SPEA2) on CR problems, the simulation results show that the VAMIGA reduces 2%~15% of the transmitted power and 6%~36% of the bit error rate (BER), and improves 6%~8% of modulation index. Thus, the VAMIGA can work more efficiently to solve multi-objective optimization and CR waveform design in different environment.
The current cognitive radio (CR) engine based on genetic algorithm usually adopts a weight-method to change multi-objective into a single objective, which may miss optimal solutions and reduce the efficiency of engine. This paper proposes an adaptive multi-objective immune genetic algorithm with vaccine injection (VAMIGA) to resolve this problem. The vaccine injection could optimize the decision result and convergence speed by saving and recycling the excellent genes. Compared with the strength Pareto evolutionary algorithm (SPEA2) on CR problems, the simulation results show that the VAMIGA reduces 2%~15% of the transmitted power and 6%~36% of the bit error rate (BER), and improves 6%~8% of modulation index. Thus, the VAMIGA can work more efficiently to solve multi-objective optimization and CR waveform design in different environment.
2015, 44(2): 190-194.
doi: 10.3969/j.issn.1001-0548.2015.02.005
Abstract:
This paper utilizes intermodulation distortion (IMD) sweet spot effect to solve the linear problem of high-efficient power amplifier (PA), so that PA can maintain high levels of both efficiency and linearity. IMD sweet spot is the result of interactions between circuit's small-signal nonlinear distortion (weak nonlinearity) determined by the PA quiescent operating point and large-signal distortion effects determined by the device current turn-on and current saturation (strong nonlinearity). The weak nonlinearity can be controlled by changing the gate bias. By seeking appropriate bias voltage for IMD sweet spot generated at the beginning of the transistor gain compression, PA can achieve high efficiency and good linearity. Based on the effect of sweet spot, a high-efficiency linear PA has been designed. Test results show that at center frequency of 2.2 GHz, 37.1 dBm output powers, 53.4% drain efficiency can be obtained, while the third-order intermodulation distortion (IMD3) and the fifth-order intermodulation distortion (IMD5) are less than -30 dBc.
This paper utilizes intermodulation distortion (IMD) sweet spot effect to solve the linear problem of high-efficient power amplifier (PA), so that PA can maintain high levels of both efficiency and linearity. IMD sweet spot is the result of interactions between circuit's small-signal nonlinear distortion (weak nonlinearity) determined by the PA quiescent operating point and large-signal distortion effects determined by the device current turn-on and current saturation (strong nonlinearity). The weak nonlinearity can be controlled by changing the gate bias. By seeking appropriate bias voltage for IMD sweet spot generated at the beginning of the transistor gain compression, PA can achieve high efficiency and good linearity. Based on the effect of sweet spot, a high-efficiency linear PA has been designed. Test results show that at center frequency of 2.2 GHz, 37.1 dBm output powers, 53.4% drain efficiency can be obtained, while the third-order intermodulation distortion (IMD3) and the fifth-order intermodulation distortion (IMD5) are less than -30 dBc.
2015, 44(2): 195-200.
doi: 10.3969/j.issn.1001-0548.2015.02.006
Abstract:
In this paper a rotation and PEGASIS-based energy-efficient chain routing protocol is proposed to reduce energy consumption in wireless sensor nodes and prolong the life span of wireless sensor networks (WSNs). The proposed protocol is an improved protocol of geographical adaptive fidelity (GAF) protocol based on (power-efficient gathering in sensor information system, PEGASIS) chain protocol. Simulation results demonstrate that the proposed protocol has better performance in energy conservation than traditional routing protocols, such as PEGASIS, LEACH and EEPB. Moreover, more uniform distribution of the wireless sensor nodes could be realized when more than a half of the sensor nodes died.
In this paper a rotation and PEGASIS-based energy-efficient chain routing protocol is proposed to reduce energy consumption in wireless sensor nodes and prolong the life span of wireless sensor networks (WSNs). The proposed protocol is an improved protocol of geographical adaptive fidelity (GAF) protocol based on (power-efficient gathering in sensor information system, PEGASIS) chain protocol. Simulation results demonstrate that the proposed protocol has better performance in energy conservation than traditional routing protocols, such as PEGASIS, LEACH and EEPB. Moreover, more uniform distribution of the wireless sensor nodes could be realized when more than a half of the sensor nodes died.
2015, 44(2): 201-204.
doi: 10.3969/j.issn.1001-0548.2015.02.007
Abstract:
Estimation of signal parameters via rotational invariance techniques (ESPRIT) algorithm does not need spectral peak searching, but the estimated variance is greater than the multiple signal classification (MUSIC) algorithm. This paper proposes a generalized eigenvector-based ESPRIT algorithm, which makes full use of generalized eigenvectors of rotation invariant relationship matrix and obtains a better performance when compared with traditional ESPRIT algorithm. The experimental results show that the proposed algorithm can achieve the similar performance of the MUSIC algorithm while keeping a low computational complexity.
Estimation of signal parameters via rotational invariance techniques (ESPRIT) algorithm does not need spectral peak searching, but the estimated variance is greater than the multiple signal classification (MUSIC) algorithm. This paper proposes a generalized eigenvector-based ESPRIT algorithm, which makes full use of generalized eigenvectors of rotation invariant relationship matrix and obtains a better performance when compared with traditional ESPRIT algorithm. The experimental results show that the proposed algorithm can achieve the similar performance of the MUSIC algorithm while keeping a low computational complexity.
2015, 44(2): 205-210.
doi: 10.3969/j.issn.1001-0548.2015.02.008
Abstract:
The base extension operation for residue number systems (RNS) plays an important role in RNS-based digital signal processing (DSP) systems. In this paper, a new base extension algorithm is proposed which can extend the dynamic range from moduli set {2n-1,2n,2n+1} to moduli set {22n-1,22n,22n+1}. In this paper, the very large scale integrated (VLSI) circuits implement of the proposed algorithm is also presented, with the properties of moduli set {2n-1,2n,2n+1}, the implement is composed of binary adders and modular adders; The analysis result based on unit-gate model and ASIC (application specific integrated circuit) implementation shows that the VLSI implementation of the proposed base extension algorithm exhibits better performances for the same dynamic range extension.
The base extension operation for residue number systems (RNS) plays an important role in RNS-based digital signal processing (DSP) systems. In this paper, a new base extension algorithm is proposed which can extend the dynamic range from moduli set {2n-1,2n,2n+1} to moduli set {22n-1,22n,22n+1}. In this paper, the very large scale integrated (VLSI) circuits implement of the proposed algorithm is also presented, with the properties of moduli set {2n-1,2n,2n+1}, the implement is composed of binary adders and modular adders; The analysis result based on unit-gate model and ASIC (application specific integrated circuit) implementation shows that the VLSI implementation of the proposed base extension algorithm exhibits better performances for the same dynamic range extension.
2015, 44(2): 211-214.
doi: 10.3969/j.issn.1001-0548.2015.02.009
Abstract:
In the traditional study of independent component analysis (ICA), the outlier data had not been considered. This paper proposes a method based on influence function to find the outliers from the observed data in ICA. General, outliers have a significant influence on the separation performance of ICA. Using the influence functions to project the observed data, the impulsive noisy components which mixed in the observed data can be eliminated from the normal data. The experimental results demonstrate the effectiveness of proposed method.
In the traditional study of independent component analysis (ICA), the outlier data had not been considered. This paper proposes a method based on influence function to find the outliers from the observed data in ICA. General, outliers have a significant influence on the separation performance of ICA. Using the influence functions to project the observed data, the impulsive noisy components which mixed in the observed data can be eliminated from the normal data. The experimental results demonstrate the effectiveness of proposed method.
2015, 44(2): 215-220.
doi: 10.3969/j.issn.1001-0548.2015.02.010
Abstract:
It has been proved that network coding, which allows network intermediate nodes to perform processing operations on the incoming packets instead of simply forwarding them, can approach the max-flow min-cut limit of the network graph. But such coding operations in network nodes incur additional computational overhead and consume public resources. Under condition of achieving the desired throughput in multicast scenario, this paper presents an improved quantum-inspired evolutionary algorithm (IQEA-NC) to minimize network coding resources. Compared with normal quantum-inspired evolutionary algorithm, IQEA-NC can achieve some effective improvements, such as decreasing the search space, increasing global search capacity, and jumping out of local optimum. The simulation experiment results show that IQEA-NC runs faster and more efficiently, improves the optimization performance compared with the existing algorithm.
It has been proved that network coding, which allows network intermediate nodes to perform processing operations on the incoming packets instead of simply forwarding them, can approach the max-flow min-cut limit of the network graph. But such coding operations in network nodes incur additional computational overhead and consume public resources. Under condition of achieving the desired throughput in multicast scenario, this paper presents an improved quantum-inspired evolutionary algorithm (IQEA-NC) to minimize network coding resources. Compared with normal quantum-inspired evolutionary algorithm, IQEA-NC can achieve some effective improvements, such as decreasing the search space, increasing global search capacity, and jumping out of local optimum. The simulation experiment results show that IQEA-NC runs faster and more efficiently, improves the optimization performance compared with the existing algorithm.
2015, 44(2): 221-226.
doi: 10.3969/j.issn.1001-0548.2015.02.011
Abstract:
Connectivity patterns in delay tolerant networks can be modeled as Edge-Markovian graph and this model is better than traditional negative exponential model. In this paper, the optimal control problem of two-hop routing algorithm is explored with energy constraint under the Edge-Markovian graph. Considering that the source forwards the message with certain probability and the probabilistic two-hop routing method is used to decrease the energy consumption, the problem turns into the finding of the optimal probability to maximize the delivery ratio with energy constraint. Thus, a theoretical model for the problem is proposed based on the discrete time Markov process. We further prove that the optimal forwarding probability conforms to the threshold form. Simulation and numerical results show the correctness of the theoretical model.
Connectivity patterns in delay tolerant networks can be modeled as Edge-Markovian graph and this model is better than traditional negative exponential model. In this paper, the optimal control problem of two-hop routing algorithm is explored with energy constraint under the Edge-Markovian graph. Considering that the source forwards the message with certain probability and the probabilistic two-hop routing method is used to decrease the energy consumption, the problem turns into the finding of the optimal probability to maximize the delivery ratio with energy constraint. Thus, a theoretical model for the problem is proposed based on the discrete time Markov process. We further prove that the optimal forwarding probability conforms to the threshold form. Simulation and numerical results show the correctness of the theoretical model.
2015, 44(2): 227-232.
doi: 10.3969/j.issn.1001-0548.2015.02.012
Abstract:
This paper proposes a multi-channel switching method to consider the node density and service traffic. The nodes in the network are divided into three modes according to the node density and service information flow in nodes. The message is delivered in different patterns for different modes. The judging of the three modes and switching between them are studied. The switching method is simulated on OMNET++ platform and its performance is evaluated.
This paper proposes a multi-channel switching method to consider the node density and service traffic. The nodes in the network are divided into three modes according to the node density and service information flow in nodes. The message is delivered in different patterns for different modes. The judging of the three modes and switching between them are studied. The switching method is simulated on OMNET++ platform and its performance is evaluated.
2015, 44(2): 233-238.
doi: 10.3969/j.issn.1001-0548.2015.02.013
Abstract:
Rapid and accurate diagnosis of fault has a crucial role for speeding up the recovery process and ensuring the safe operation of the power system. This paper analyzes the data sources currently used in power system fault diagnosis, proposes the use of wide-area information for rapid diagnosis of power systems. First, wavelet transform is used to process the wide area measurement data, through correlation coefficient method to eliminate the noise and reconstruct the signal, then permutation entropy calculation use the of the reconstructed signal. Since the complexity of the arrangement of entropy reflects a one-dimensional time series, the signal changes with high sensitivity, therefore, it can be applied to power system fault diagnosis. Compared with the traditional wavelet analysis method, this method avoids the problem of selecting wavelet base and meets the conditions for online fault diagnosis. In addition, it is simple and easy to program. The approach is applied in IEEE10 machine 39-bus system fault diagnosis. The diagnostic results show the applicability and effectiveness of the method.
Rapid and accurate diagnosis of fault has a crucial role for speeding up the recovery process and ensuring the safe operation of the power system. This paper analyzes the data sources currently used in power system fault diagnosis, proposes the use of wide-area information for rapid diagnosis of power systems. First, wavelet transform is used to process the wide area measurement data, through correlation coefficient method to eliminate the noise and reconstruct the signal, then permutation entropy calculation use the of the reconstructed signal. Since the complexity of the arrangement of entropy reflects a one-dimensional time series, the signal changes with high sensitivity, therefore, it can be applied to power system fault diagnosis. Compared with the traditional wavelet analysis method, this method avoids the problem of selecting wavelet base and meets the conditions for online fault diagnosis. In addition, it is simple and easy to program. The approach is applied in IEEE10 machine 39-bus system fault diagnosis. The diagnostic results show the applicability and effectiveness of the method.
2015, 44(2): 239-244.
doi: 10.3969/j.issn.1001-0548.2015.02.014
Abstract:
The application of three-dimensional mapping technology has improved the waveform capture rate of digital storage oscilloscopes (DSO), however it does not eliminate the dead time of DSO. In this research, a new 3D-imaging method, fragmentation and centralized mapping, is proposed. It divides high-speed large capacity dynamic memory (DDR3 SDRAM) into several regions based on the time-base. In each region, the waveforms with the same triggering condition are stored, and the waveforms will be post-processed with three-dimensional mapping Technology. Dual-port RAM based fast waveform mapping technology and parallel waveform mapping technology decrease the mapping time greatly. This method guarantees high waveform capture rate and enable larger data storage capacity. It also enables seamless acquisition for the duration of each stored waveform. By mathematical modeling, it can be proven that our system can capture the sporadic transient event in shorter time. Through double pulse test method, it can be demonstrated that the highest capture rate of the system is measured to be 6 250 000 wfms/s. After analysis, it is obtained that the ratio of dead time is reduced to 43.86%.
The application of three-dimensional mapping technology has improved the waveform capture rate of digital storage oscilloscopes (DSO), however it does not eliminate the dead time of DSO. In this research, a new 3D-imaging method, fragmentation and centralized mapping, is proposed. It divides high-speed large capacity dynamic memory (DDR3 SDRAM) into several regions based on the time-base. In each region, the waveforms with the same triggering condition are stored, and the waveforms will be post-processed with three-dimensional mapping Technology. Dual-port RAM based fast waveform mapping technology and parallel waveform mapping technology decrease the mapping time greatly. This method guarantees high waveform capture rate and enable larger data storage capacity. It also enables seamless acquisition for the duration of each stored waveform. By mathematical modeling, it can be proven that our system can capture the sporadic transient event in shorter time. Through double pulse test method, it can be demonstrated that the highest capture rate of the system is measured to be 6 250 000 wfms/s. After analysis, it is obtained that the ratio of dead time is reduced to 43.86%.
2015, 44(2): 245-253.
doi: 10.3969/j.issn.1001-0548.2015.02.015
Abstract:
For the recent twenty years, the business application of Internet has not only impacted the traditional financial institutions significantly, but also promoted the internet thinking in the financial business. New payment forms, new investment and financing channels are derived, for instance, P2P lending, crowdfunding, and credit loan based on big data. This paper systematically reviews the internet finance concepts, definitions, growth processes, business models, research status, and development challenges. Moreover, this paper also analyzes the social and economic backgrounds of the rapidly developing internet finance in China and discusses the future researches in the area.
For the recent twenty years, the business application of Internet has not only impacted the traditional financial institutions significantly, but also promoted the internet thinking in the financial business. New payment forms, new investment and financing channels are derived, for instance, P2P lending, crowdfunding, and credit loan based on big data. This paper systematically reviews the internet finance concepts, definitions, growth processes, business models, research status, and development challenges. Moreover, this paper also analyzes the social and economic backgrounds of the rapidly developing internet finance in China and discusses the future researches in the area.
2015, 44(2): 254-259.
doi: 10.3969/j.issn.1001-0548.2015.02.016
Abstract:
Micro-blog user recommendation has great significance and value for improving the user experience and promoting the long-term development of the social network. In this paper, multiple features reflecting the correlation between micro-blog users are extracted. Combining the user features and ranks scores for potential users, top-n potential users are recommended for the target user. The experimental results based on the dataset gained from Sina Micro-Blog shows that the proposed method is feasible and effective, and it can provide personalized user recommendation with high performance for micro-blog users.
Micro-blog user recommendation has great significance and value for improving the user experience and promoting the long-term development of the social network. In this paper, multiple features reflecting the correlation between micro-blog users are extracted. Combining the user features and ranks scores for potential users, top-n potential users are recommended for the target user. The experimental results based on the dataset gained from Sina Micro-Blog shows that the proposed method is feasible and effective, and it can provide personalized user recommendation with high performance for micro-blog users.
2015, 44(2): 260-265.
doi: 10.3969/j.issn.1001-0548.2015.02.017
Abstract:
This paper presents an improved ant colony algorithm for mobile robot path planning under dynamic complex conditions based on prior knowledge of global static environment. On the basis of conventional ant colony algorithm, by adjusting the transition probability, limiting the bounds of pheromone, and introducing relevant strategy to solve the deadlock problem, the improved ant colony algorithm not only can avoid the blindness of early planning and increase the diversity of solutions, but also can improve global search capability of the algorithm, and further reduce the possibility of algorithm prematurity as well. During the planning process, according to the direction changes of the dynamic obstacles, corresponding collision avoidance strategies are put forward. The Follw_wall behavior is introduced for unexpected situations in the environment. Simulation results show that the proposed algorithm is superior to conventional ant colony algorithm. It can effectively guide the mobile robot to avoid dynamic obstacles. Thus obtains a collision free optimal or suboptimal path, which adapts to the changes of the environment more effectively.
This paper presents an improved ant colony algorithm for mobile robot path planning under dynamic complex conditions based on prior knowledge of global static environment. On the basis of conventional ant colony algorithm, by adjusting the transition probability, limiting the bounds of pheromone, and introducing relevant strategy to solve the deadlock problem, the improved ant colony algorithm not only can avoid the blindness of early planning and increase the diversity of solutions, but also can improve global search capability of the algorithm, and further reduce the possibility of algorithm prematurity as well. During the planning process, according to the direction changes of the dynamic obstacles, corresponding collision avoidance strategies are put forward. The Follw_wall behavior is introduced for unexpected situations in the environment. Simulation results show that the proposed algorithm is superior to conventional ant colony algorithm. It can effectively guide the mobile robot to avoid dynamic obstacles. Thus obtains a collision free optimal or suboptimal path, which adapts to the changes of the environment more effectively.
2015, 44(2): 266-271.
doi: 10.3969/j.issn.1001-0548.2015.02.018
Abstract:
According to known and unknown database attacking, we propose an architecture of multi-dimensional attack-aware database activity monitor based on captured SQLs, in which the user database behavior schema set can be constructed in the beginning by monitoring their requests and detect potential attacks by analyzing SQL queries/statements during database running. Based on the SQL's syntactic structure and semantic feature, we present different user behavior models on SQL schematic and semantic level, session level, and structure for libraries of user behavior patterns. Malicious transactions are detected by means of calculating the structure distance of user database requests with SQLs or SQL sequences in schema matching set of the detection engine.
According to known and unknown database attacking, we propose an architecture of multi-dimensional attack-aware database activity monitor based on captured SQLs, in which the user database behavior schema set can be constructed in the beginning by monitoring their requests and detect potential attacks by analyzing SQL queries/statements during database running. Based on the SQL's syntactic structure and semantic feature, we present different user behavior models on SQL schematic and semantic level, session level, and structure for libraries of user behavior patterns. Malicious transactions are detected by means of calculating the structure distance of user database requests with SQLs or SQL sequences in schema matching set of the detection engine.
2015, 44(2): 272-277.
doi: 10.3969/j.issn.1001-0548.2015.02.019
Abstract:
An adaptive multi-focus image fusion algorithm based on algebraic multigrid (AMG) method is proposed for the capability to extract structural information of an image. The data of the coarse level is extracted to reconstruct the image block, appropriate source image block is selected into the fusion result according to the mean square error between the reconstructed image block and the original image block. For smoothing the fused result, an adaptive strategy is used. Experimental results show that most of the clear objects can be retained in the fused image with the proposed algorithm without loss of effective information.
An adaptive multi-focus image fusion algorithm based on algebraic multigrid (AMG) method is proposed for the capability to extract structural information of an image. The data of the coarse level is extracted to reconstruct the image block, appropriate source image block is selected into the fusion result according to the mean square error between the reconstructed image block and the original image block. For smoothing the fused result, an adaptive strategy is used. Experimental results show that most of the clear objects can be retained in the fused image with the proposed algorithm without loss of effective information.
2015, 44(2): 278-282.
doi: 10.3969/j.issn.1001-0548.2015.02.020
Abstract:
It is always a difficult problem in face recognition on how to process the multimodal information (e.g. variation in lighting or orientation). Traditional graph embedding algorithms neglect congener correlation between different multimodal clusters of the same class (i.e. subject) and do not properly incorporate discriminative information between classes. In this paper, a robust graph embedding face recognition algorithm is proposed. It properly captures multimodal structure within one class and also realizes a balance between local manifold structures and the global discriminative information. Experiments in several public databases demonstrate that the proposed algorithm can achieve better performance than the state-of-arts reported in recent literatures.
It is always a difficult problem in face recognition on how to process the multimodal information (e.g. variation in lighting or orientation). Traditional graph embedding algorithms neglect congener correlation between different multimodal clusters of the same class (i.e. subject) and do not properly incorporate discriminative information between classes. In this paper, a robust graph embedding face recognition algorithm is proposed. It properly captures multimodal structure within one class and also realizes a balance between local manifold structures and the global discriminative information. Experiments in several public databases demonstrate that the proposed algorithm can achieve better performance than the state-of-arts reported in recent literatures.
2015, 44(2): 283-288.
doi: 10.3969/j.issn.1001-0548.2015.02.021
Abstract:
In order to accurately extract edge information of magnetic tile surface defect, an edge detection algorithm based on image weighted information entropy and wavelet modulus maxima is proposed. Because the magnetic tile surface with low contrast and textured background has a negative influence on edge extraction, a new BHPF filter with adaptive changing cutoff frequency is designed. The clarity and complexity of textured background are quantitatively described by weighted information entropy of image gradient variance. The filter changes its parameter through matching the non-linear relationship between information entropy and cutoff frequency. To prevent the losing of edge information, the best decomposition scale is obtained by the level determination function. In order to ensure the edge continuity and veracity, wavelet modulus maxima is judged through a double threshold to get the edge point. Experimental results show that the algorithm outperforms the conventional canny and sobel algorithms in detection of magnetic tile crack edge. This edge detection algorithm can also detect other defects.
In order to accurately extract edge information of magnetic tile surface defect, an edge detection algorithm based on image weighted information entropy and wavelet modulus maxima is proposed. Because the magnetic tile surface with low contrast and textured background has a negative influence on edge extraction, a new BHPF filter with adaptive changing cutoff frequency is designed. The clarity and complexity of textured background are quantitatively described by weighted information entropy of image gradient variance. The filter changes its parameter through matching the non-linear relationship between information entropy and cutoff frequency. To prevent the losing of edge information, the best decomposition scale is obtained by the level determination function. In order to ensure the edge continuity and veracity, wavelet modulus maxima is judged through a double threshold to get the edge point. Experimental results show that the algorithm outperforms the conventional canny and sobel algorithms in detection of magnetic tile crack edge. This edge detection algorithm can also detect other defects.
2015, 44(2): 289-294.
doi: 10.3969/j.issn.1001-0548.2015.02.022
Abstract:
The Arnold mapping with chaotic has achieved good results in the image scrambling and secure communication, however, the Arnold transformation matrix is periodic so that finding the cycle of the transformation matrix is the important basis of scrambling transformation. In order to study the periodicity of the 3-Arnold transform matrix, the new concept of the twin Fibonacci sequence is introduced and four related periodicity theorems are given. And then we prove that the molding cycle of 3-Arnold transform matrix is half of the molding cycle of the twin Fibonacci sequence. Accordingly, a new method to determine the molding cycle of the transformation matrix is formed. At last, a new several-rounds double-scrambling encryption algorithm based on the 3-Arnold mapping is proposed. Simulation results show the proposed algorithm outperforms the 2-Arnold mapping algorithm.
The Arnold mapping with chaotic has achieved good results in the image scrambling and secure communication, however, the Arnold transformation matrix is periodic so that finding the cycle of the transformation matrix is the important basis of scrambling transformation. In order to study the periodicity of the 3-Arnold transform matrix, the new concept of the twin Fibonacci sequence is introduced and four related periodicity theorems are given. And then we prove that the molding cycle of 3-Arnold transform matrix is half of the molding cycle of the twin Fibonacci sequence. Accordingly, a new method to determine the molding cycle of the transformation matrix is formed. At last, a new several-rounds double-scrambling encryption algorithm based on the 3-Arnold mapping is proposed. Simulation results show the proposed algorithm outperforms the 2-Arnold mapping algorithm.
2015, 44(2): 295-298.
doi: 10.3969/j.issn.1001-0548.2015.02.023
Abstract:
To solve the problems of data security and distributed data management that existed in SaaS platform, the paper puts forward a system application framework based on open SaaS service platform architecture. On the basis of this architecture, the paper gives abstract modeling on data application, which uses the idea of optimizing cache management strategy to solve data access problem in network environment. Combining with traditional LRU arithmetic and SIZE arithmetic, the paper gives a LRU-SIZE arithmetic which is effective on the management of data access of the open SaaS service platform. The combination of open SaaS service architecture and LRU-SIZE arithmetic can solve the problems of data access and data management of enterprise effectively. Meanwhile, it can improve the application experience of users.
To solve the problems of data security and distributed data management that existed in SaaS platform, the paper puts forward a system application framework based on open SaaS service platform architecture. On the basis of this architecture, the paper gives abstract modeling on data application, which uses the idea of optimizing cache management strategy to solve data access problem in network environment. Combining with traditional LRU arithmetic and SIZE arithmetic, the paper gives a LRU-SIZE arithmetic which is effective on the management of data access of the open SaaS service platform. The combination of open SaaS service architecture and LRU-SIZE arithmetic can solve the problems of data access and data management of enterprise effectively. Meanwhile, it can improve the application experience of users.
2015, 44(2): 299-305.
doi: 10.3969/j.issn.1001-0548.2015.02.024
Abstract:
The third generation photovoltaic devices based on solution-processed colloidal quantum dot (CQD) have particular advantage for harvesting sun's broad spectrum, because of the size tunability of CQD's bandgap. In the review, firstly, the fundamentals of solar cell and four key parameters were introduced. Secondly, evident deviations of CQD films from the homogenous bulk semiconductor were pointed out and the effects of CQD's properties on solar cell efficiency were analyzed. Finally, some main device architectures of PbS CQD solar cells developed in recent years were summarized. Advanced CQD thin films, more accurate exciton transport physical model, and better device architectures would help to improve the efficiency of CQD solar cells.
The third generation photovoltaic devices based on solution-processed colloidal quantum dot (CQD) have particular advantage for harvesting sun's broad spectrum, because of the size tunability of CQD's bandgap. In the review, firstly, the fundamentals of solar cell and four key parameters were introduced. Secondly, evident deviations of CQD films from the homogenous bulk semiconductor were pointed out and the effects of CQD's properties on solar cell efficiency were analyzed. Finally, some main device architectures of PbS CQD solar cells developed in recent years were summarized. Advanced CQD thin films, more accurate exciton transport physical model, and better device architectures would help to improve the efficiency of CQD solar cells.
2015, 44(2): 306-310.
doi: 10.3969/j.issn.1001-0548.2015.02.025
Abstract:
The investigation on the temperature shift owing to the leakage of fuel pipelines is the fundamental research in the leakage's monitoring technology based on optical fiber sensors. This paper presents a theoretical analysis of the temperature shift caused by fuel pipeline leakage. The simulation in combination of real data and physical model shows that the monitoring technology based on optical fiber sensors is universally applicable to fuel pipelines that lie in different surroundings, and the easier the fuel pipeline leakage occurs (owing to a larger pressure), the easier this leakage is monitored.
The investigation on the temperature shift owing to the leakage of fuel pipelines is the fundamental research in the leakage's monitoring technology based on optical fiber sensors. This paper presents a theoretical analysis of the temperature shift caused by fuel pipeline leakage. The simulation in combination of real data and physical model shows that the monitoring technology based on optical fiber sensors is universally applicable to fuel pipelines that lie in different surroundings, and the easier the fuel pipeline leakage occurs (owing to a larger pressure), the easier this leakage is monitored.
2015, 44(2): 311-315.
doi: 10.3969/j.issn.1001-0548.2015.02.026
Abstract:
Considerable functional MRI (fMRI) studies have shown differences of brain activity between lie-telling and truth-telling. However there are few studies aiming at brain network feature of lie-telling. In this study, we obtained fMRI data of 32 subjects while responding to questions in a truthful, inverse and deceitful manner, then constructed whole-brain functional connectivity networks for the lie-telling and truth-telling conditions based on a canonical template of 116 brain regions, and used a multivariate pattern analysis approach based on machine learning to classify the lie-telling from truth-telling. The results showed that the classifier achieved high classification accuracy (82.03%, 84.38% for lie-telling, 79.69% for truth-telling) and could extract informational functional connectivities that could be used to discriminate lie-telling from truth-telling. These informational functional connectivities were mainly located among networks. These results not only demonstrated good performance when classifying with functional connectivities, but also elucidated the neural mechanism of lie-telling from a functional integration viewpoint.
Considerable functional MRI (fMRI) studies have shown differences of brain activity between lie-telling and truth-telling. However there are few studies aiming at brain network feature of lie-telling. In this study, we obtained fMRI data of 32 subjects while responding to questions in a truthful, inverse and deceitful manner, then constructed whole-brain functional connectivity networks for the lie-telling and truth-telling conditions based on a canonical template of 116 brain regions, and used a multivariate pattern analysis approach based on machine learning to classify the lie-telling from truth-telling. The results showed that the classifier achieved high classification accuracy (82.03%, 84.38% for lie-telling, 79.69% for truth-telling) and could extract informational functional connectivities that could be used to discriminate lie-telling from truth-telling. These informational functional connectivities were mainly located among networks. These results not only demonstrated good performance when classifying with functional connectivities, but also elucidated the neural mechanism of lie-telling from a functional integration viewpoint.
2015, 44(2): 316-320.
doi: 10.3969/j.issn.1001-0548.2015.02.027
Abstract:
Due to the non-stationary characteristics of scalp electroencephalography (EEG), traditional analysis methods, such as coherent method, etc., can't well detect statistical dependencies between time series recorded. synchronization likelihood (SL) based on generalized synchronization has been introduced to overcome some limitations of coherent estimations. And it is applied to analyze real EEG signals. Simulation results of Henon mapping system and actual EEG data demonstrate that the SL method is suitable for measuring the relationship between non-stationary signals. The changes of brain synchronization of healthy subjects are studied from eye closed to eye open during rest. Results show that the synchronization of alpha rhythm is significantly reduced in almost all electrodes, and the brain activity has a certain inhibition. All the results show that the method is of great significance in the study of EEG. It provides certain reference for future EEG research.
Due to the non-stationary characteristics of scalp electroencephalography (EEG), traditional analysis methods, such as coherent method, etc., can't well detect statistical dependencies between time series recorded. synchronization likelihood (SL) based on generalized synchronization has been introduced to overcome some limitations of coherent estimations. And it is applied to analyze real EEG signals. Simulation results of Henon mapping system and actual EEG data demonstrate that the SL method is suitable for measuring the relationship between non-stationary signals. The changes of brain synchronization of healthy subjects are studied from eye closed to eye open during rest. Results show that the synchronization of alpha rhythm is significantly reduced in almost all electrodes, and the brain activity has a certain inhibition. All the results show that the method is of great significance in the study of EEG. It provides certain reference for future EEG research.
