ISSN 
2013 Vol. 42, No. 6
2013, 42(6): 802-805.
doi: 10.3969/j.issn.1001-0548.2013.06.001
Abstract:
The era of big data has brought big opportunities to network science, yet the study of network science is also facing big challenges of processing huge amount of data. Meanwhile, a new project outsourcing model, the crowd-sourcing model, has been widely applied recently to solving data related problems. This paper presents a brief introduction to the new driving forces of network science: big data and crowd-sourcing model. We also review the 2013 Alibaba data platform innovation competition as well as our award-winning work. We expect that driven by the crowd-sourcing model, data science, which deals with big data processing techniques, shall evolve together with network science into a prosperous future.
The era of big data has brought big opportunities to network science, yet the study of network science is also facing big challenges of processing huge amount of data. Meanwhile, a new project outsourcing model, the crowd-sourcing model, has been widely applied recently to solving data related problems. This paper presents a brief introduction to the new driving forces of network science: big data and crowd-sourcing model. We also review the 2013 Alibaba data platform innovation competition as well as our award-winning work. We expect that driven by the crowd-sourcing model, data science, which deals with big data processing techniques, shall evolve together with network science into a prosperous future.
2013, 42(6): 806-816.
doi: 10.3969/j.issn.1001-0548.2013.06.002
Abstract:
Faced with severe traffic congestions, the level of traffic planning and organization has been paid much attention. As a crucial and fundamental data for traffic planning and traffic organization, travel demand has been an important research topic for a long time. Stepping into the big data era, the fast developments in the research fields of human mobility and complex network offer new methodologies for predicting travel demand and improving transportation networks. In this paper, we review the important works in the areas of human mobility and transportation network, discuss the potential opportunities which bring to transportation engineering, and finally introduce a new interdisciplinary study of the two research areas.
Faced with severe traffic congestions, the level of traffic planning and organization has been paid much attention. As a crucial and fundamental data for traffic planning and traffic organization, travel demand has been an important research topic for a long time. Stepping into the big data era, the fast developments in the research fields of human mobility and complex network offer new methodologies for predicting travel demand and improving transportation networks. In this paper, we review the important works in the areas of human mobility and transportation network, discuss the potential opportunities which bring to transportation engineering, and finally introduce a new interdisciplinary study of the two research areas.
2013, 42(6): 817-830.
Abstract:
We present a review on some local immunization algorithm, including community networks, adaptive networks, and coupled networks. For community networks, through the empirical analysis of five empirical networks we find that the distribution of weak ties is heterogeneous, which indicates that some bridge-nodes with more weak links play a more important role in information diffusion. We propose an efficient local algorithm to identify bridge-nodes. For adaptive networks, we study the effects of community-based control strategies on disease spreading and find that it is not ‘the earlier, the better' to control diseases. For coupled networks, we present a facilitate-restrain asymmetric interacting spreading model, and investigate the impact of local disperse awareness on disease spreading. The presented results contribute to improving the understanding of key nodes in complex networks and offer beneficial reference and enlightenment for disease control and prevention.
We present a review on some local immunization algorithm, including community networks, adaptive networks, and coupled networks. For community networks, through the empirical analysis of five empirical networks we find that the distribution of weak ties is heterogeneous, which indicates that some bridge-nodes with more weak links play a more important role in information diffusion. We propose an efficient local algorithm to identify bridge-nodes. For adaptive networks, we study the effects of community-based control strategies on disease spreading and find that it is not ‘the earlier, the better' to control diseases. For coupled networks, we present a facilitate-restrain asymmetric interacting spreading model, and investigate the impact of local disperse awareness on disease spreading. The presented results contribute to improving the understanding of key nodes in complex networks and offer beneficial reference and enlightenment for disease control and prevention.
2013, 42(6): 831-835.
doi: 10.3969/j.issn.1001-0548.2013.06.004
Abstract:
In order to achieve low power consumption, the supply voltage is reduced gradually but this increases the probabilistic characteristic of digital circuits. In this paper, we propose the design method which combines of the reduced precision redundancy (RPR) algorithm with triple-modular redundancy (TMR) algorithm and Redundant residue number systems (RRNS) algorithm, referred as RPR-TMR and RPR-RRNS architectures respectively. The performance of these two kinds of structures is designed and analyzed. The results of case study show that the proposed of architectures can archive low power consumption and high output SNR.
In order to achieve low power consumption, the supply voltage is reduced gradually but this increases the probabilistic characteristic of digital circuits. In this paper, we propose the design method which combines of the reduced precision redundancy (RPR) algorithm with triple-modular redundancy (TMR) algorithm and Redundant residue number systems (RRNS) algorithm, referred as RPR-TMR and RPR-RRNS architectures respectively. The performance of these two kinds of structures is designed and analyzed. The results of case study show that the proposed of architectures can archive low power consumption and high output SNR.
2013, 42(6): 836-841.
doi: 10.3969/j.issn.1001-0548.2013.06.005
Abstract:
In relay-aided systems, the available data rate of a relay link user is determined by the minimum rate of the first hop link and the second hop link. Separate resource allocation in two-hop links may deteriorate system performance. Therefore, based on the throughput balance, a forward allocation (FA) algorithm and a backward allocation (BA) algorithm are proposed. In FA algorithm, resource allocation in the first hop greatly affects that of the second hop. In BA algorithm, resource allocation in the first hop is based on the throughput of the second hop. Simulation results show that the proposed algorithms are beneficial to solve the bottleneck problem of two-hop relay networks. In particular, the BA algorithm considers the optimal rates of users, and thus it can further improve system throughput and reduce power assumption.
In relay-aided systems, the available data rate of a relay link user is determined by the minimum rate of the first hop link and the second hop link. Separate resource allocation in two-hop links may deteriorate system performance. Therefore, based on the throughput balance, a forward allocation (FA) algorithm and a backward allocation (BA) algorithm are proposed. In FA algorithm, resource allocation in the first hop greatly affects that of the second hop. In BA algorithm, resource allocation in the first hop is based on the throughput of the second hop. Simulation results show that the proposed algorithms are beneficial to solve the bottleneck problem of two-hop relay networks. In particular, the BA algorithm considers the optimal rates of users, and thus it can further improve system throughput and reduce power assumption.
2013, 42(6): 842-847.
doi: 10.3969/j.issn.1001-0548.2013.06.006
Abstract:
In this paper, the end-to-end signal noise ratio (SNR) expression is derived in the multiple antenna two-way amplify-and-forward (AF) relay system using beamforming over non-identically distributed independent Nakagami-m channels. The cumulative distribution function and the probability density function of end-to-end SNR are deduced. Utilizing the accumulative distribution function, the closed expressions of outage probability and average symbol error ratio are derived. The simulation results demonstrate the validity of the theoretical analysis.
In this paper, the end-to-end signal noise ratio (SNR) expression is derived in the multiple antenna two-way amplify-and-forward (AF) relay system using beamforming over non-identically distributed independent Nakagami-m channels. The cumulative distribution function and the probability density function of end-to-end SNR are deduced. Utilizing the accumulative distribution function, the closed expressions of outage probability and average symbol error ratio are derived. The simulation results demonstrate the validity of the theoretical analysis.
2013, 42(6): 848-851.
doi: 10.3969/j.issn.1001-0548.2013.06.007
Abstract:
Multilateration is a new technology in air traffic control. Anti-jamming technology is important for multilateration systems. Signals for different targets will generate garbles which are main jamming in multilateration. This paper analyzes the features and laws of A/C garbles in time domain, defines pulse aliasing degree and pulse description words, and proposes a new algorithm to separate the aliasing pulses in time domain according to pulse width, amplitude and phase. The A/C transponders are separated totally after sub-pulses are cleared. At last, simulations show the efficiency of proposed algorithm.
Multilateration is a new technology in air traffic control. Anti-jamming technology is important for multilateration systems. Signals for different targets will generate garbles which are main jamming in multilateration. This paper analyzes the features and laws of A/C garbles in time domain, defines pulse aliasing degree and pulse description words, and proposes a new algorithm to separate the aliasing pulses in time domain according to pulse width, amplitude and phase. The A/C transponders are separated totally after sub-pulses are cleared. At last, simulations show the efficiency of proposed algorithm.
2013, 42(6): 852-856.
doi: 10.3969/j.issn.1001-0548.2013.06.008
Abstract:
An electromagnetic topology (EMT) model is presented to solve the problem of electromagnetic pulse (EMP) coupling through an aperture to multi-conductor transmission lines inside cavity. The topological diagram and the interaction sequence diagram of the proposed model are constructed by using EMT theory. The aperture is replaced by the equivalent electric and magnetic dipoles sources which can be exciting sources inside cavity, the equivalent distribution sources on multi-conductor transmission lines are computed by the supermatrix integral transform. The simulation results of incident voltage and electric field shielding effectiveness (FSE) indicate the validity of the proposed model.
An electromagnetic topology (EMT) model is presented to solve the problem of electromagnetic pulse (EMP) coupling through an aperture to multi-conductor transmission lines inside cavity. The topological diagram and the interaction sequence diagram of the proposed model are constructed by using EMT theory. The aperture is replaced by the equivalent electric and magnetic dipoles sources which can be exciting sources inside cavity, the equivalent distribution sources on multi-conductor transmission lines are computed by the supermatrix integral transform. The simulation results of incident voltage and electric field shielding effectiveness (FSE) indicate the validity of the proposed model.
2013, 42(6): 857-861.
doi: 10.3969/j.issn.1001-0548.2013.06.009
Abstract:
A low-power high-dynamic range digitally controlled variable gain amplifier (VGA) is designed in this paper. First, a novel method for defining the output common-mode voltages of VGA is proposed by adjusting the current of the load resistor according to the change of its resistance to keep the voltage drop across the load resistor unchanged, so that the common-mode feedback circuit is not required for the VGA, resulting in decreased power consumption. Second, coupling capacitors are used between the stages to remove the DC offset, and hence no extra DC offset cancellation circuit is necessary. The VGA is designed and simulated in TSMC 0.18μm CMOS process. Simulation results show that, the VGA dissipates an average current of 504.7μA while achieving 1.16MHz -3dB bandwidth and providing 81dB gain range with 3 dB resolution, and the gain error is less than ±0.65dB.
A low-power high-dynamic range digitally controlled variable gain amplifier (VGA) is designed in this paper. First, a novel method for defining the output common-mode voltages of VGA is proposed by adjusting the current of the load resistor according to the change of its resistance to keep the voltage drop across the load resistor unchanged, so that the common-mode feedback circuit is not required for the VGA, resulting in decreased power consumption. Second, coupling capacitors are used between the stages to remove the DC offset, and hence no extra DC offset cancellation circuit is necessary. The VGA is designed and simulated in TSMC 0.18μm CMOS process. Simulation results show that, the VGA dissipates an average current of 504.7μA while achieving 1.16MHz -3dB bandwidth and providing 81dB gain range with 3 dB resolution, and the gain error is less than ±0.65dB.
2013, 42(6): 862-868.
doi: 10.3969/j.issn.1001-0548.2013.06.010
Abstract:
Vulnerability to environmental interference makes RSSI multiple figures, which leads to the destruction of the one-to-one map between the "RSSI-Distance", and a larger error in the distance mapping which solely uses the point-to-point mapping between the RSSI and the distance. To solve this problem, a ranging method based on weighted distance interval probability is proposed: establishing the mapping relation between RSSI and distance interval, calculating the probability of the occurrence of RSSI in each point of the interval distance, and then weighted summing with probabilities of all intervals. The results is then used as the distance corresponding to the RSSI. The simulation results show that this method provides a higher ranging precision in different degrees compared with the method of curve fitting and interval ranging method.
Vulnerability to environmental interference makes RSSI multiple figures, which leads to the destruction of the one-to-one map between the "RSSI-Distance", and a larger error in the distance mapping which solely uses the point-to-point mapping between the RSSI and the distance. To solve this problem, a ranging method based on weighted distance interval probability is proposed: establishing the mapping relation between RSSI and distance interval, calculating the probability of the occurrence of RSSI in each point of the interval distance, and then weighted summing with probabilities of all intervals. The results is then used as the distance corresponding to the RSSI. The simulation results show that this method provides a higher ranging precision in different degrees compared with the method of curve fitting and interval ranging method.
2013, 42(6): 869-874.
doi: 10.3969/j.issn.1001-0548.2013.06.011
Abstract:
An indoor Time-Reversal (TR) Based RF locating method is proposed to locate a RFID tag in rich scattering environment. Unlike the conventional locating methods based on Received Signal Strength Indicator (RSSI), Time of Arrival (TOA), and Direction of Arrival (DOA) estimation which are sensitive to the multipath effects, TR method utilizes the multipath propagations as useful information. The proposed method first requests a RFID tag to emit a locating pulse, the scattered pulse is received by a sparse antenna array with 3 wavelengths spacing, and then the received signals are time-reversed and retransmitted in a simulation domain by using numerical electromagnetic calculation methods. Because the information of the environment (such as rooms, doors, et al) can be known in advance, the propagation paths of the pulse can be accurately reconstructed. In order to locate the RFID tag, a location estimator based on TR inversion is also given. By monitoring the output of this location estimator or the energy distribution in both time and spatial domains, the RFID tag can be located. This TR based RF locating method has great advantages over the traditional back-projection method because the multipath propagations in this rich scattering environment have been well utilized. Simulation results show the effectiveness of the proposed method. Furthermore, originating from the well utilization of multipath propagations, TR based method can still work very well for locating with this very sparse array.
An indoor Time-Reversal (TR) Based RF locating method is proposed to locate a RFID tag in rich scattering environment. Unlike the conventional locating methods based on Received Signal Strength Indicator (RSSI), Time of Arrival (TOA), and Direction of Arrival (DOA) estimation which are sensitive to the multipath effects, TR method utilizes the multipath propagations as useful information. The proposed method first requests a RFID tag to emit a locating pulse, the scattered pulse is received by a sparse antenna array with 3 wavelengths spacing, and then the received signals are time-reversed and retransmitted in a simulation domain by using numerical electromagnetic calculation methods. Because the information of the environment (such as rooms, doors, et al) can be known in advance, the propagation paths of the pulse can be accurately reconstructed. In order to locate the RFID tag, a location estimator based on TR inversion is also given. By monitoring the output of this location estimator or the energy distribution in both time and spatial domains, the RFID tag can be located. This TR based RF locating method has great advantages over the traditional back-projection method because the multipath propagations in this rich scattering environment have been well utilized. Simulation results show the effectiveness of the proposed method. Furthermore, originating from the well utilization of multipath propagations, TR based method can still work very well for locating with this very sparse array.
2013, 42(6): 875-879.
doi: 10.3969/j.issn.1001-0548.2013.06.012
Abstract:
Free-electron laser (FEL) wavelength can continuously tuning (up to X-ray band) with high power and high efficiency. FEL has a great potential in femtosecond science, biological molecules, nano science research field and etc. In this paper, we established the self-consistent equations of self amplified spontaneous emission (SASE) FEL by using the classical particle theory and developed the steady-state Matlab-SASE-FEL numerical code. The LCLS facility parameters were optimized and the characteristic of the radiation laser light was presented. The main advantage of this code is that the data can be easily post-processed based on the convenience of Matlab tool.
Free-electron laser (FEL) wavelength can continuously tuning (up to X-ray band) with high power and high efficiency. FEL has a great potential in femtosecond science, biological molecules, nano science research field and etc. In this paper, we established the self-consistent equations of self amplified spontaneous emission (SASE) FEL by using the classical particle theory and developed the steady-state Matlab-SASE-FEL numerical code. The LCLS facility parameters were optimized and the characteristic of the radiation laser light was presented. The main advantage of this code is that the data can be easily post-processed based on the convenience of Matlab tool.
2013, 42(6): 880-884.
doi: 10.3969/j.issn.1001-0548.2013.06.013
Abstract:
Analog signal processing is a real-time, efficient signal processing technique. Chirped delay line is a critical device for analog signal processing. The research status and progress of the chirped delay lines have been reported in this paper. By introducing a variety of approaches for the design of a chirped delay line, this paper summarizes the advantages and disadvantages of each method, and provides suggestions for further research.
Analog signal processing is a real-time, efficient signal processing technique. Chirped delay line is a critical device for analog signal processing. The research status and progress of the chirped delay lines have been reported in this paper. By introducing a variety of approaches for the design of a chirped delay line, this paper summarizes the advantages and disadvantages of each method, and provides suggestions for further research.
2013, 42(6): 885-889.
doi: 10.3969/j.issn.1001-0548.2013.06.014
Abstract:
To the sudden voltage-loss problem of battery pack in pure electrical vehicles, this paper proposes a seamless switching algorithm used for dual DC power supply. The key of this algorithm is to place a spare battery on the pure electrical vehicles as additional energy for the battery pack. When detected the pack battery has a trend of loss of voltage, in no time, the control system start the boost circuit of the spare battery. At the same time, the control system makes the battery pack and the spare battery in parallel as the power supply for the motor driver system. Then the performance of the battery pack is tested further to put in use of the battery pack power supply or the spare battery power supply. This algorithm can achieve seamless switching, make sure the load voltage stability and the equipment operation safety and reliably. The simulation results demonstrate the validity and feasibility of the proposed algorithm
To the sudden voltage-loss problem of battery pack in pure electrical vehicles, this paper proposes a seamless switching algorithm used for dual DC power supply. The key of this algorithm is to place a spare battery on the pure electrical vehicles as additional energy for the battery pack. When detected the pack battery has a trend of loss of voltage, in no time, the control system start the boost circuit of the spare battery. At the same time, the control system makes the battery pack and the spare battery in parallel as the power supply for the motor driver system. Then the performance of the battery pack is tested further to put in use of the battery pack power supply or the spare battery power supply. This algorithm can achieve seamless switching, make sure the load voltage stability and the equipment operation safety and reliably. The simulation results demonstrate the validity and feasibility of the proposed algorithm
2013, 42(6): 890-894.
doi: 10.3969/j.issn.1001-0548.2013.06.015
Abstract:
How to calculate grid infrastructure investment to meet the social, economic development needs and enhance investment efficiency. Grid investment calculation has gradually become the urgent problem faced by the grid corporation. Grid infrastructure investment calculation is difficult, due to influences with many factors, complex relationships and difficulty of grid development demand forecasting. This paper studies a grid infrastructure investment calculation model based on gray prediction. Firstly, an indicator system for grid infrastructure investment calculation is built by analyzing the relationship between indicators and grid infrastructure investment. Secondly, a gray model based indicator prediction model is built to predict the development trend of indicators which are closely related to grid infrastructure; the analytic hierarchy process (AHP) is used to determine each index change impact on investment; the amount of grid infrastructure investment is calculated by using the indicators predictive value and determine indicators,. Finally, the validity and rationality of the proposed calculation model are verified by the example of a grid infrastructure investment calculation. The proposed calculation model is extremely important theoretical and practical significance for the coordination of grid infrastructure investment and planning the scale of investment.
How to calculate grid infrastructure investment to meet the social, economic development needs and enhance investment efficiency. Grid investment calculation has gradually become the urgent problem faced by the grid corporation. Grid infrastructure investment calculation is difficult, due to influences with many factors, complex relationships and difficulty of grid development demand forecasting. This paper studies a grid infrastructure investment calculation model based on gray prediction. Firstly, an indicator system for grid infrastructure investment calculation is built by analyzing the relationship between indicators and grid infrastructure investment. Secondly, a gray model based indicator prediction model is built to predict the development trend of indicators which are closely related to grid infrastructure; the analytic hierarchy process (AHP) is used to determine each index change impact on investment; the amount of grid infrastructure investment is calculated by using the indicators predictive value and determine indicators,. Finally, the validity and rationality of the proposed calculation model are verified by the example of a grid infrastructure investment calculation. The proposed calculation model is extremely important theoretical and practical significance for the coordination of grid infrastructure investment and planning the scale of investment.
2013, 42(6): 895-899.
doi: 10.3969/j.issn.1001-0548.2013.06.016
Abstract:
This paper establishes rice microwave scattering model according to the Monte Carlo method, and modifies the model to meet the rice growth characteristics. The model simulation results were compared with scatterometer measured data, the result show that the proposed rice backscatter model can well simulate the backscattering characteristics of rice. The variation ranges of input parameters of the model were set according to the actual growth parameters of rice. The rice backscattering coefficients with different input parameters, such as rice straw and rice leaf parameter, were simulated. Finally, the sensitivity analysis according to the rice backscattering coefficient and growth parameters is presented.
This paper establishes rice microwave scattering model according to the Monte Carlo method, and modifies the model to meet the rice growth characteristics. The model simulation results were compared with scatterometer measured data, the result show that the proposed rice backscatter model can well simulate the backscattering characteristics of rice. The variation ranges of input parameters of the model were set according to the actual growth parameters of rice. The rice backscattering coefficients with different input parameters, such as rice straw and rice leaf parameter, were simulated. Finally, the sensitivity analysis according to the rice backscattering coefficient and growth parameters is presented.
2013, 42(6): 900-904.
doi: 10.3969/j.issn.1001-0548.2013.06.017
Abstract:
Cloud computing is a hot spot in computer research field, in which the performance of task scheduling algorithm directly affects total performance of cloud computing systems. For satisfying users' different service requirements, a dual fitness genetic annealing algorithm (DFG2A) is presented. Task scheduling scheme based on this algorithm can balance different attribute requirements for each task, and improve customer satisfaction index of platform. The simulation results show that DFG2A can achieve a balance between task running time and user requirements.
Cloud computing is a hot spot in computer research field, in which the performance of task scheduling algorithm directly affects total performance of cloud computing systems. For satisfying users' different service requirements, a dual fitness genetic annealing algorithm (DFG2A) is presented. Task scheduling scheme based on this algorithm can balance different attribute requirements for each task, and improve customer satisfaction index of platform. The simulation results show that DFG2A can achieve a balance between task running time and user requirements.
2013, 42(6): 905-910.
doi: 10.3969/j.issn.1001-0548.2013.06.018
Abstract:
Delay-tolerant networks (DTNs) are a type of emerging networks characterized by extraordinarily long delay paths and frequent network partitions. For the distinct characteristics of DTNs, routing becomes one of the most challenging open problems. This paper presents a routing strategy based on capability model in DTN, called CM-RSD. The CM-RSD combines capability model with quota-based routing strategy, and sprays more message copies to stronger nodes. Furthermore, this strategy can select relay nodes based on the energy states of nodes, thus protecting the survival of network nodes effectively. Our simulation results show that the CM-RSD can improve the message delivery ratio significantly while keeping the overhead ratio low in distinct scenarios with different node densities, quota of copies, and mobility models.
Delay-tolerant networks (DTNs) are a type of emerging networks characterized by extraordinarily long delay paths and frequent network partitions. For the distinct characteristics of DTNs, routing becomes one of the most challenging open problems. This paper presents a routing strategy based on capability model in DTN, called CM-RSD. The CM-RSD combines capability model with quota-based routing strategy, and sprays more message copies to stronger nodes. Furthermore, this strategy can select relay nodes based on the energy states of nodes, thus protecting the survival of network nodes effectively. Our simulation results show that the CM-RSD can improve the message delivery ratio significantly while keeping the overhead ratio low in distinct scenarios with different node densities, quota of copies, and mobility models.
2013, 42(6): 911-915.
doi: 10.3969/j.issn.1001-0548.2013.06.019
Abstract:
A vehicular real-time information query mode based on vehicular Ad hoc network (VANET) is proposed in this paper. The real-time traffic information report, facilities service information report, user queries, etc., are disseminated via epidemic routing in the system. Only after a query and a report matched at a node, the report will be returned to the query node via infrastructure network to speed up the query efficiency. The system model is analyzed and the system delay is theoretically deduced. Furthermore, the derived conclusions are validated in STRAW/SWANS simulator, which provide a theoretical basis for parameters setting and optimization for the VANET system.
A vehicular real-time information query mode based on vehicular Ad hoc network (VANET) is proposed in this paper. The real-time traffic information report, facilities service information report, user queries, etc., are disseminated via epidemic routing in the system. Only after a query and a report matched at a node, the report will be returned to the query node via infrastructure network to speed up the query efficiency. The system model is analyzed and the system delay is theoretically deduced. Furthermore, the derived conclusions are validated in STRAW/SWANS simulator, which provide a theoretical basis for parameters setting and optimization for the VANET system.
2013, 42(6): 916-920.
doi: 10.3969/j.issn.1001-0548.2013.06.020
Abstract:
A unsupervised feature annotation criterion-information gain criterion (IGC)-based on feature matrix information gain is proposed to rank the feature variable. According to this rank, three new feature filtering methods：direct selection (DS), cumulate maximum entropy (CEM), and information gain maximum (IGM) are given to reduce clustering complexity. The clustering results of these three filtering methods with two existing variance selection (VS) and gene shaving (GS) methods were tested and compared by using classic QC or K-means algorithm and three biological datasets: rod-shaped viruses (RSV), mixed-lineage leukemia (MLL), and acute leukemia patients (ALP). The experiment results show our feature filtering method has obvious superiority in accelerating the clustering procedure and preserving the clustering structure of initial data.
A unsupervised feature annotation criterion-information gain criterion (IGC)-based on feature matrix information gain is proposed to rank the feature variable. According to this rank, three new feature filtering methods：direct selection (DS), cumulate maximum entropy (CEM), and information gain maximum (IGM) are given to reduce clustering complexity. The clustering results of these three filtering methods with two existing variance selection (VS) and gene shaving (GS) methods were tested and compared by using classic QC or K-means algorithm and three biological datasets: rod-shaped viruses (RSV), mixed-lineage leukemia (MLL), and acute leukemia patients (ALP). The experiment results show our feature filtering method has obvious superiority in accelerating the clustering procedure and preserving the clustering structure of initial data.
2013, 42(6): 921-925.
doi: 10.3969/j.issn.1001-0548.2013.06.021
Abstract:
For the flow shop scheduling problem which aims to minimize makespan, this paper gives a new derivation about its mathematical definition. A new heuristic method is proposed to shorten the waiting time of each job as much as possible on the basis of reducing the processing time of the first machine and last job. The result of simulation experiments shows that, the new heuristic algorithm has good performance, and the average quality and the stability of scheduling sequences generated by new method are significantly better than other heuristic algorithms with same complexity.
For the flow shop scheduling problem which aims to minimize makespan, this paper gives a new derivation about its mathematical definition. A new heuristic method is proposed to shorten the waiting time of each job as much as possible on the basis of reducing the processing time of the first machine and last job. The result of simulation experiments shows that, the new heuristic algorithm has good performance, and the average quality and the stability of scheduling sequences generated by new method are significantly better than other heuristic algorithms with same complexity.
2013, 42(6): 926-929.
doi: 10.3969/j.issn.1001-0548.2013.06.022
Abstract:
The security of the identity-based ring signature scheme of [18] is analyzed in this paper. It is found that this scheme did not possess the unforgeability property. It is also shown that the proposed scheme is insecure against a new kind of forgery attack, i.e. the identity assembly forgery attack. With such an attack, an adversary is able to forge valid ring signatures on any message based on the features of its identity, and the identity of the adversary can be even not included in the ring of the forged signature.
The security of the identity-based ring signature scheme of [18] is analyzed in this paper. It is found that this scheme did not possess the unforgeability property. It is also shown that the proposed scheme is insecure against a new kind of forgery attack, i.e. the identity assembly forgery attack. With such an attack, an adversary is able to forge valid ring signatures on any message based on the features of its identity, and the identity of the adversary can be even not included in the ring of the forged signature.
2013, 42(6): 930-934.
doi: 10.3969/j.issn.1001-0548.2013.06.023
Abstract:
In order to meet the demand of realizing the speaker-independent speech recognition on the embedded system, a embedded speech recognition system in discrete hidden Markov model (DHMM) is proposed. The software architecture of the recognition software part running on embedded system and the training software part running on PC system are designed. According to features of TMS320VC5509A processor, the recognition software on embedded system is implemented and optimized. Also it is presented that the Viterbi algorithm is optimized on the basis of the irreversible characteristics of the speech signal. Experiments show the real-time response speed of the system achieved within 200 ms at same the recognition rate, and show its high availability.
In order to meet the demand of realizing the speaker-independent speech recognition on the embedded system, a embedded speech recognition system in discrete hidden Markov model (DHMM) is proposed. The software architecture of the recognition software part running on embedded system and the training software part running on PC system are designed. According to features of TMS320VC5509A processor, the recognition software on embedded system is implemented and optimized. Also it is presented that the Viterbi algorithm is optimized on the basis of the irreversible characteristics of the speech signal. Experiments show the real-time response speed of the system achieved within 200 ms at same the recognition rate, and show its high availability.
2013, 42(6): 935-938.
doi: 10.3969/j.issn.1001-0548.2013.06.024
Abstract:
Inference leakage problem in high security database is an important subject to research. Analysis of inference channel is the basis of solving inference leakage problem. In this paper, a more sound inference rule about inference channels, namely weak functional dependency of equivalence (referred to as WFDoE), is proposed and its formal definition is also presented. Rough set theory is used to define WFDoE's the degree of credibility and support, a mining arithmetic of WFDoE inference rules from database is provided. Experimental result shows that the proposed arithmetic is able to enhance ability to prevent inference leakage.
Inference leakage problem in high security database is an important subject to research. Analysis of inference channel is the basis of solving inference leakage problem. In this paper, a more sound inference rule about inference channels, namely weak functional dependency of equivalence (referred to as WFDoE), is proposed and its formal definition is also presented. Rough set theory is used to define WFDoE's the degree of credibility and support, a mining arithmetic of WFDoE inference rules from database is provided. Experimental result shows that the proposed arithmetic is able to enhance ability to prevent inference leakage.
2013, 42(6): 939-943.
doi: 10.3969/j.issn.1001-0548.2013.06.025
Abstract:
This paper discusses the stabilization problem of a yaw-pitch gimbaled seeker system under the condition that system uncertainties and actuator saturation nonlinearity cannot be ignored. Based on gimbal dynamics, a dynamics model is formulated for the two-axis gimbaled seeker, and the perturbation effect of platform angular motion to system configuration is studied. The variation intervals of the perturbed parameters are furthermore estimated, and the uncertainty description of the gimbaled system is obtained. Two stable regions are conservatively estimated by judging the stability of interval matrix and solving the linear matrix inequalities (LMIs), respectively, and then the condition under which the two-axis gimbaled seeker can be stabilized and the method via which the system can be optimally stabilized are presented. The analysis method presented in this paper can provide guidelines for the optimal design of stable control loop for the two-axis gimbaled seeker.
This paper discusses the stabilization problem of a yaw-pitch gimbaled seeker system under the condition that system uncertainties and actuator saturation nonlinearity cannot be ignored. Based on gimbal dynamics, a dynamics model is formulated for the two-axis gimbaled seeker, and the perturbation effect of platform angular motion to system configuration is studied. The variation intervals of the perturbed parameters are furthermore estimated, and the uncertainty description of the gimbaled system is obtained. Two stable regions are conservatively estimated by judging the stability of interval matrix and solving the linear matrix inequalities (LMIs), respectively, and then the condition under which the two-axis gimbaled seeker can be stabilized and the method via which the system can be optimally stabilized are presented. The analysis method presented in this paper can provide guidelines for the optimal design of stable control loop for the two-axis gimbaled seeker.
2013, 42(6): 944-950.
doi: 10.3969/j.issn.1001-0548.2013.06.026
Abstract:
The paper presents a method to compute the robust stability region of fractional order interval plant with time delay by using fractional order PIλDμ controller. The edge theorem is adopted to decompose interval plant to several vertices sub-plants. The characteristic polynomials of vertices sub-plants and the value set of exposed edge for polytope are given. The D-decomposition technique is applied to solve the stability region of each vertices sub-pant. The values of λ and μ of PIλD and PIDμ controllers are obtained in the biggest stability region of all sub-plants. The fraction order PIλDμ controller is constructed by the values of λ and μ. The stability region of each sub-plant is plotted by using fractional order PIλDμ controller. Furthermore, the overlap of stability region of each sub-plant is the stability region of fractional order interval plant with time delay. The paper also proves that the overlap of stability region is the robust stability region of fraction order interval plant.
The paper presents a method to compute the robust stability region of fractional order interval plant with time delay by using fractional order PIλDμ controller. The edge theorem is adopted to decompose interval plant to several vertices sub-plants. The characteristic polynomials of vertices sub-plants and the value set of exposed edge for polytope are given. The D-decomposition technique is applied to solve the stability region of each vertices sub-pant. The values of λ and μ of PIλD and PIDμ controllers are obtained in the biggest stability region of all sub-plants. The fraction order PIλDμ controller is constructed by the values of λ and μ. The stability region of each sub-plant is plotted by using fractional order PIλDμ controller. Furthermore, the overlap of stability region of each sub-plant is the stability region of fractional order interval plant with time delay. The paper also proves that the overlap of stability region is the robust stability region of fraction order interval plant.
2013, 42(6): 951-954.
doi: 10.3969/j.issn.1001-0548.2013.06.027
Abstract:
On the basis of parallelizable real-time motion tracking algorithm, we propose the modified weighted phase separation (WPS) displacement estimation algorithm based on lateral guidance and 2D estimation window, and validate whether this algorithm can improve elastography performance. The displacement of the middle A-lines is estimated as the seed displacement, the other displacements are estimated from the middle to two sides, their initial displacements are selected from the lateral adjacent displacement estimate, and the final displacements are generated by using 2D weighted phase separation algorithm. The simulation shows that the modified WPS algorithm can achieve elastogram with higher SNRe than the original WPS at various strains or window-lengths. And the phantom experiment also shows that the modified WPS algorithm can generate the higher performance elastogram.
On the basis of parallelizable real-time motion tracking algorithm, we propose the modified weighted phase separation (WPS) displacement estimation algorithm based on lateral guidance and 2D estimation window, and validate whether this algorithm can improve elastography performance. The displacement of the middle A-lines is estimated as the seed displacement, the other displacements are estimated from the middle to two sides, their initial displacements are selected from the lateral adjacent displacement estimate, and the final displacements are generated by using 2D weighted phase separation algorithm. The simulation shows that the modified WPS algorithm can achieve elastogram with higher SNRe than the original WPS at various strains or window-lengths. And the phantom experiment also shows that the modified WPS algorithm can generate the higher performance elastogram.
2013, 42(6): 955-960.
doi: 10.3969/j.issn.1001-0548.2013.06.028
Abstract:
Focusing on the non-stationary and non-linear of heart sounds, a new method of feature extraction based on empirical mode decomposition (EMD) and Correlation dimension is proposed. The heart sound signals are decomposed into a finite number of intrinsic mode functions (IMFs). The IMF components are chosen by using the criteria of mutual correlation coefficient between IMF components and original signal and then the correlation dimension of the top four intrinsic mode functions (IMF1~IMF4) is calculated by using G-P algorithm. The eigenvectors are put into the artificial neural network for automatic discrimination between normal and abnormal signals. In the process of phase-space reconstruction, Cao theory and the mutual information function are used to determine the two important parameters: delay time and embedding dimension. The clinical data experimental diagnosis and contract test results show that the approach proposed could identify the pathological heart sound effectively.
Focusing on the non-stationary and non-linear of heart sounds, a new method of feature extraction based on empirical mode decomposition (EMD) and Correlation dimension is proposed. The heart sound signals are decomposed into a finite number of intrinsic mode functions (IMFs). The IMF components are chosen by using the criteria of mutual correlation coefficient between IMF components and original signal and then the correlation dimension of the top four intrinsic mode functions (IMF1~IMF4) is calculated by using G-P algorithm. The eigenvectors are put into the artificial neural network for automatic discrimination between normal and abnormal signals. In the process of phase-space reconstruction, Cao theory and the mutual information function are used to determine the two important parameters: delay time and embedding dimension. The clinical data experimental diagnosis and contract test results show that the approach proposed could identify the pathological heart sound effectively.
