ISSN 
2011 Vol. 40, No. 6
2011, 40(6): 802-814.
doi: 10.3969/j.issn.1001-0548.2011.06.001
Abstract:
Aging problem is a crucial issue nowadays. The population of old persons of age more than 65 will achieve 20% in Asia area in 2020. To care these aged persons, one of the important things is to provide early warning of the physical improper situations before it is too late to be warned by themselves. The intelligent high speed tele-care system could be a solution. It includes design, and implementation of low profile, short distance, high speed intelligent sensor network with high data rate communication, and modern signal processing techniques. The technologies behind is to promote high speed chip designs and to enhance the smart system command and control capability. A novel intelligent communication topology with smart-suit for body kinematics, blood pressure, heart beat rate and breath monitoring is presented. The prominent characteristic of the structure is that data source for every communicating unit can be determined by its current sensing state. Propagation delay is reduced and transmitting speed is improved. The topology fits to be adopted in a complex communication system, contributing to threshold control and integrated management for the early warning. Modern technologies and experimental waveforms are presented.
Aging problem is a crucial issue nowadays. The population of old persons of age more than 65 will achieve 20% in Asia area in 2020. To care these aged persons, one of the important things is to provide early warning of the physical improper situations before it is too late to be warned by themselves. The intelligent high speed tele-care system could be a solution. It includes design, and implementation of low profile, short distance, high speed intelligent sensor network with high data rate communication, and modern signal processing techniques. The technologies behind is to promote high speed chip designs and to enhance the smart system command and control capability. A novel intelligent communication topology with smart-suit for body kinematics, blood pressure, heart beat rate and breath monitoring is presented. The prominent characteristic of the structure is that data source for every communicating unit can be determined by its current sensing state. Propagation delay is reduced and transmitting speed is improved. The topology fits to be adopted in a complex communication system, contributing to threshold control and integrated management for the early warning. Modern technologies and experimental waveforms are presented.
2011, 40(6): 815-824.
doi: 10.3969/j.issn.1001-0548.2011.06.002
Abstract:
An overview of advanced neural network methods for modeling radio-frequency (RF) and microwave electronic devices is presented. Knowledge-based engineering concept is utilized where the knowledge of RF/microwave electronics in the form of equivalent circuits and empirical formulas is combined with neural networks. Advantages of adding knowledge on the performance of the neural models in terms of generalization ability versus different sizes of training data through a knowledge based neural network (KBNN) technique are demonstrated and examples of comparisons with conventional MLP (without any knowledge-base) are given. Several methods of combining existing circuit models with neural networks, including the source difference method, the prior knowledge input method, and the space-mapped neural models, are also introduced. Application examples on modeling microwave transmission line and high electron mobility transistor (HEMT) device demonstrate that KBNN is an efficient approach for modeling various types of microwave devices.
An overview of advanced neural network methods for modeling radio-frequency (RF) and microwave electronic devices is presented. Knowledge-based engineering concept is utilized where the knowledge of RF/microwave electronics in the form of equivalent circuits and empirical formulas is combined with neural networks. Advantages of adding knowledge on the performance of the neural models in terms of generalization ability versus different sizes of training data through a knowledge based neural network (KBNN) technique are demonstrated and examples of comparisons with conventional MLP (without any knowledge-base) are given. Several methods of combining existing circuit models with neural networks, including the source difference method, the prior knowledge input method, and the space-mapped neural models, are also introduced. Application examples on modeling microwave transmission line and high electron mobility transistor (HEMT) device demonstrate that KBNN is an efficient approach for modeling various types of microwave devices.
2011, 40(6): 825-828.
doi: 10.3969/j.issn.1001-0548.2011.06.003
Abstract:
In this paper, an improved timing method is proposed for high-order QAM system. The timing processing block computes the value of weights for each data block and then combines the soft information based on the estimated results. In addition, an improved polarity decision phase detector algorithm is proposed for the carrier recovery, the soft information fed back from the decoder is modulated and then sent to the polarity module for polarity decision. The polarity decision result is fed to the phase detector (PD) module for the phase estimation. Simulation results show that the improved code-aided timing and carrier synchronization system can perform better than the unimproved algorithm at very low SNR.
In this paper, an improved timing method is proposed for high-order QAM system. The timing processing block computes the value of weights for each data block and then combines the soft information based on the estimated results. In addition, an improved polarity decision phase detector algorithm is proposed for the carrier recovery, the soft information fed back from the decoder is modulated and then sent to the polarity module for polarity decision. The polarity decision result is fed to the phase detector (PD) module for the phase estimation. Simulation results show that the improved code-aided timing and carrier synchronization system can perform better than the unimproved algorithm at very low SNR.
2011, 40(6): 829-834.
doi: 10.3969/j.issn.1001-0548.2011.06.004
Abstract:
To solve the problem that high power signal may be tracked by anti-radiation missile in coherent interference, a novel cancellation interference method based on phase modulation is proposed. The kernel thought is to produce cancellation signal according to radar signal and its phase modulation information so as to protect target through the cancellation between radar signal and interference signal. Characteristics of several typical pulse compressed signals are analyzed. Mechanism of signal cancellation's influence on filtering output was explained through the group delay function. Then connection between cancellation error and cancellation effect is explained. The implementation method of cancellation signal and the effect evaluating indicator are provided. Simulation results show that the proposed method could reduce effectively the detecting probability of radar to that below 50% through canceling interference, which can provide reference for targets' active stealth.
To solve the problem that high power signal may be tracked by anti-radiation missile in coherent interference, a novel cancellation interference method based on phase modulation is proposed. The kernel thought is to produce cancellation signal according to radar signal and its phase modulation information so as to protect target through the cancellation between radar signal and interference signal. Characteristics of several typical pulse compressed signals are analyzed. Mechanism of signal cancellation's influence on filtering output was explained through the group delay function. Then connection between cancellation error and cancellation effect is explained. The implementation method of cancellation signal and the effect evaluating indicator are provided. Simulation results show that the proposed method could reduce effectively the detecting probability of radar to that below 50% through canceling interference, which can provide reference for targets' active stealth.
2011, 40(6): 835-838.
doi: 10.3969/j.issn.1001-0548.2011.06.005
Abstract:
A new method to design wideband transition from double-ridge waveguide to coaxial line is introduced in this investigation. The proposed design of the transition is based on the theory of mode conversion and impedance matching. The Chebyshev type of impedance matching network is employed, a new mode converter structure between the coaxial line and the ridge waveguide has been used for design. The characteristics of the transition in microwave frequencies are analyzed and simulated by full-wave analysis software. According to this ideal, a couple of transitions have been finished. The predicted superduper performance has been confirmed experimentally in the range of 6.5～18 GHz.
A new method to design wideband transition from double-ridge waveguide to coaxial line is introduced in this investigation. The proposed design of the transition is based on the theory of mode conversion and impedance matching. The Chebyshev type of impedance matching network is employed, a new mode converter structure between the coaxial line and the ridge waveguide has been used for design. The characteristics of the transition in microwave frequencies are analyzed and simulated by full-wave analysis software. According to this ideal, a couple of transitions have been finished. The predicted superduper performance has been confirmed experimentally in the range of 6.5～18 GHz.
2011, 40(6): 839-843.
doi: 10.3969/j.issn.1001-0548.2011.06.006
Abstract:
A novel track-before-detfect (TBD) algorithm based on Modified Hough Transform (MHT) and Particle Filter (PF) is proposed to solve the shortcomings that the real target trajectories cannot be obtained for many false trajectories existing in the TBD algorithm based on Hough Transform. Firstly, possible trajectories are initiated by the MHT to obtain the region of location of the target, then this region as the prior knowledge is provided to the PF and the likelihood function is established for target detection by the weights which are calculated by the PF, at last the false trajectories are eliminated and the real target trajectories are given with detection results. Simulation results show the presented algorithm can implement accurate tracking and detection weak target.
A novel track-before-detfect (TBD) algorithm based on Modified Hough Transform (MHT) and Particle Filter (PF) is proposed to solve the shortcomings that the real target trajectories cannot be obtained for many false trajectories existing in the TBD algorithm based on Hough Transform. Firstly, possible trajectories are initiated by the MHT to obtain the region of location of the target, then this region as the prior knowledge is provided to the PF and the likelihood function is established for target detection by the weights which are calculated by the PF, at last the false trajectories are eliminated and the real target trajectories are given with detection results. Simulation results show the presented algorithm can implement accurate tracking and detection weak target.
2011, 40(6): 844-848.
doi: 10.3969/j.issn.1001-0548.2011.06.007
Abstract:
In this paper, the Alpha-stable distribution and its fractional low order statistics are introduced. An array configuration is devised and a new method based on phase fractional lower-order moment (PFLOM) is proposed for 2-D direction of arrival (DOA) estimation. PFLOM-based covariance matrix of the desired signal can be derived from that of the receive signal. Making use of eigenvalue decomposition and applying least square or total least square method, we can derive DOA. Finally, the ESPRIT methods based on classic covariance matrix, SCM, FLOM, and PFLOM are compared in this paper. Simulations show that the method has robust performance and the smaller bias and mean square errors.
In this paper, the Alpha-stable distribution and its fractional low order statistics are introduced. An array configuration is devised and a new method based on phase fractional lower-order moment (PFLOM) is proposed for 2-D direction of arrival (DOA) estimation. PFLOM-based covariance matrix of the desired signal can be derived from that of the receive signal. Making use of eigenvalue decomposition and applying least square or total least square method, we can derive DOA. Finally, the ESPRIT methods based on classic covariance matrix, SCM, FLOM, and PFLOM are compared in this paper. Simulations show that the method has robust performance and the smaller bias and mean square errors.
2011, 40(6): 849-854.
doi: 10.3969/j.issn.1001-0548.2011.06.008
Abstract:
The error performance of vehicle-relayed inter-vehicular communication (IVC) systems is studied based on double-Rayleigh fading model. The statistical properties of the end-to-end signal-to-noise ratio (SNR) are analyzed and the solution to optimal power allocation (OPA) coefficient is presented for the system. A generalized expression of the lower bound on average symbol-error-rate (ASER) for various M-ary modulations is derived through moment-generating-function based approach for the cooperative IVC systems. The method to finding the solution of Chernoff union bound on error probability of the system is provided. Simulation results show that the ASER decreases by applying vehicle-relaying technique in vehicle-to-vehicle communications. The differences between the exact ASER value and its lower bound increase slowly with SNR and the range to the ASER is dete rmined by its lower bound and Chernoff upper bound. Numerical results further indicate that the transmit power can save 3 dB at most through OPA than equal power allocation method in the cooperative IVC system.
The error performance of vehicle-relayed inter-vehicular communication (IVC) systems is studied based on double-Rayleigh fading model. The statistical properties of the end-to-end signal-to-noise ratio (SNR) are analyzed and the solution to optimal power allocation (OPA) coefficient is presented for the system. A generalized expression of the lower bound on average symbol-error-rate (ASER) for various M-ary modulations is derived through moment-generating-function based approach for the cooperative IVC systems. The method to finding the solution of Chernoff union bound on error probability of the system is provided. Simulation results show that the ASER decreases by applying vehicle-relaying technique in vehicle-to-vehicle communications. The differences between the exact ASER value and its lower bound increase slowly with SNR and the range to the ASER is dete rmined by its lower bound and Chernoff upper bound. Numerical results further indicate that the transmit power can save 3 dB at most through OPA than equal power allocation method in the cooperative IVC system.
2011, 40(6): 855-859.
doi: 10.3969/j.issn.1001-0548.2011.06.009
Abstract:
A distributed rate-splitting (DRS) scheme is proposed to approach the rate region boundary of the two-user Gaussian interference channel. It is shown that the DRS scheme can be formulated as a non-cooperative game. Therefore, an iterative multiple waterlevels water-filling algorithm (IML-WFA) is developed to efficiently reach the Nash equilibrium (NE) of the non-cooperative game and a sufficient condition on the convergence of IML-WFA is proposed. Numerical examples show that the rate-tuples achieved by the DRS are very close to the boundary of the well-known HK regions.
A distributed rate-splitting (DRS) scheme is proposed to approach the rate region boundary of the two-user Gaussian interference channel. It is shown that the DRS scheme can be formulated as a non-cooperative game. Therefore, an iterative multiple waterlevels water-filling algorithm (IML-WFA) is developed to efficiently reach the Nash equilibrium (NE) of the non-cooperative game and a sufficient condition on the convergence of IML-WFA is proposed. Numerical examples show that the rate-tuples achieved by the DRS are very close to the boundary of the well-known HK regions.
2011, 40(6): 860-864.
doi: 10.3969/j.issn.1001-0548.2011.06.010
Abstract:
The performance and relations between base stations (BS) and mobile terminal (MT) of hybrid cooperation in cellular communication networks is investigated. Through studying the per-cell achievable uplink ergodic capacity of time-division linear cellular networks with MT-BS hybrid cooperation, a computational expression for the achievable rate of OAF combining with multi-cell processing in MIMO channel is derived and the performance of hybrid collaborations in MIMO channels is verified by simulations. Moreover, the simulation results in MIMO flat fading channels with different system parameters show that the macro-diversity gains through multi-cell decode is independent with MTs collaboration but can be linear superposition. However, high interference from adjacent cells weakens the effect of MTs cooperation in hybrid collaboration systems.
The performance and relations between base stations (BS) and mobile terminal (MT) of hybrid cooperation in cellular communication networks is investigated. Through studying the per-cell achievable uplink ergodic capacity of time-division linear cellular networks with MT-BS hybrid cooperation, a computational expression for the achievable rate of OAF combining with multi-cell processing in MIMO channel is derived and the performance of hybrid collaborations in MIMO channels is verified by simulations. Moreover, the simulation results in MIMO flat fading channels with different system parameters show that the macro-diversity gains through multi-cell decode is independent with MTs collaboration but can be linear superposition. However, high interference from adjacent cells weakens the effect of MTs cooperation in hybrid collaboration systems.
2011, 40(6): 865-868.
doi: 10.3969/j.issn.1001-0548.2011.06.011
Abstract:
As the increasing of radar resolution, the statistics of the clutter appear no longer to be Gaussian. This paper derives a covariance matrix structure of sea-clutter from maximal likelihood function of probability denisty function of no-Gaussian sea clutter. Initialization of a Toeplitz matrix of fixed point covariance matrix in no-Gaussian sea clutter is proposed by slove the covariance matrix. The simulation result shows that the proposed method can enhance the detector performance in real sea clutter.
As the increasing of radar resolution, the statistics of the clutter appear no longer to be Gaussian. This paper derives a covariance matrix structure of sea-clutter from maximal likelihood function of probability denisty function of no-Gaussian sea clutter. Initialization of a Toeplitz matrix of fixed point covariance matrix in no-Gaussian sea clutter is proposed by slove the covariance matrix. The simulation result shows that the proposed method can enhance the detector performance in real sea clutter.
2011, 40(6): 869-872.
doi: 10.3969/j.issn.1001-0548.2011.06.012
Abstract:
A novel passive optical access network with fire alarming services is proposed in this paper. The architecture uses the fiber Bragg gratings (FBGs) with the same central wavelength and low reflectivity as sensors. Sensing signals are integrated into the Ethernet psssive optical network (EPON) system without changing its architecture. In addition, the optical time domain reflectometer (OTDR) technology us used to position. This architecture has almost no impact on the power and the distortion of the optical communication signals. By using FBGs with the same central wavelength, complex and expensive FBG's wavelength demodulation technology is not necessary.
A novel passive optical access network with fire alarming services is proposed in this paper. The architecture uses the fiber Bragg gratings (FBGs) with the same central wavelength and low reflectivity as sensors. Sensing signals are integrated into the Ethernet psssive optical network (EPON) system without changing its architecture. In addition, the optical time domain reflectometer (OTDR) technology us used to position. This architecture has almost no impact on the power and the distortion of the optical communication signals. By using FBGs with the same central wavelength, complex and expensive FBG's wavelength demodulation technology is not necessary.
2011, 40(6): 873-877.
doi: 10.3969/j.issn.1001-0548.2011.06.013
Abstract:
The performance of the Ka-band gyro-TWT has been evaluated using the self-consistent nonlinear particle-tracing. The normal mode expansion is used to solve the Maxwell equation with source. Through expand the electromagnetic field in guider-center coordinate, and combing with the electron motion equation, the self-consistent nonlinear equation is induced. The velocity spread, dielectric loading, distributed dielectric loading are taken into consideration to modify the self-consistent equation. The relation of output power with current, velocity spread, dielectric loading, and distributed dielectric loading is given through the numerical calculation. Combining the small signal analysis of self-oscillation, the optimum parameter under stable working point can be got from the nonlinear analysis.
The performance of the Ka-band gyro-TWT has been evaluated using the self-consistent nonlinear particle-tracing. The normal mode expansion is used to solve the Maxwell equation with source. Through expand the electromagnetic field in guider-center coordinate, and combing with the electron motion equation, the self-consistent nonlinear equation is induced. The velocity spread, dielectric loading, distributed dielectric loading are taken into consideration to modify the self-consistent equation. The relation of output power with current, velocity spread, dielectric loading, and distributed dielectric loading is given through the numerical calculation. Combining the small signal analysis of self-oscillation, the optimum parameter under stable working point can be got from the nonlinear analysis.
2011, 40(6): 878-881.
doi: 10.3969/j.issn.1001-0548.2011.06.014
Abstract:
Complex linear systems will be generated when scattering problems are simulated by hybrid FEM-MoM methods, such complex linear systems are difficult to solve with iterative methods alone, so preconditioning technique is usually applied to accelerate the convergence rate of iterative methods. For such complex linear systems, several preconditioners are proposed in this paper. Numerical results show the effectiveness of the proposed preconditioning technique.
Complex linear systems will be generated when scattering problems are simulated by hybrid FEM-MoM methods, such complex linear systems are difficult to solve with iterative methods alone, so preconditioning technique is usually applied to accelerate the convergence rate of iterative methods. For such complex linear systems, several preconditioners are proposed in this paper. Numerical results show the effectiveness of the proposed preconditioning technique.
2011, 40(6): 882-886.
doi: 10.3969/j.issn.1001-0548.2011.06.015
Abstract:
Aiming to the undesirable switching between running and stop state during the process of active power cutback according to the demands of grid operators or preset value, an active power dispatching algorithm is presented in this paper, which can smooth the wind farm output power and impulse on grid. The wind turbines are divided into two categories of controllable and uncontrollable based on their adjustable margins to the whole active power decreasing adjustment. The models for stopping sets and power dispatching are proposed. The application results indicate that the power dispatching algorithm can make the active power decreasing process accurate and smooth with less wind turbines switching between running and stopping, and reduce the real active power overshoot of wind turbines.
Aiming to the undesirable switching between running and stop state during the process of active power cutback according to the demands of grid operators or preset value, an active power dispatching algorithm is presented in this paper, which can smooth the wind farm output power and impulse on grid. The wind turbines are divided into two categories of controllable and uncontrollable based on their adjustable margins to the whole active power decreasing adjustment. The models for stopping sets and power dispatching are proposed. The application results indicate that the power dispatching algorithm can make the active power decreasing process accurate and smooth with less wind turbines switching between running and stopping, and reduce the real active power overshoot of wind turbines.
2011, 40(6): 887-891.
doi: 10.3969/j.issn.1001-0548.2011.06.016
Abstract:
A flight control system conceptual prototype design method based on integration of SysML and Simulink is researched. Firstly, the improvements between SysML and UML /STATEMATE that meet the flight control system design requirements are introduced. Secondly, a SysML extension mechanism that integrates SysML and Simulink model through code embedding is researched, the key problems it faces are analyzed and its profile definition is given. In the last, taking an unmanned aircraft system for example, the conceptual virtual prototype of its flight control system predator is designed, and functions, behaviors of predator are verified and validated. The design practice shows this method could meet the design requirements of flight control system conceptual prototype.
A flight control system conceptual prototype design method based on integration of SysML and Simulink is researched. Firstly, the improvements between SysML and UML /STATEMATE that meet the flight control system design requirements are introduced. Secondly, a SysML extension mechanism that integrates SysML and Simulink model through code embedding is researched, the key problems it faces are analyzed and its profile definition is given. In the last, taking an unmanned aircraft system for example, the conceptual virtual prototype of its flight control system predator is designed, and functions, behaviors of predator are verified and validated. The design practice shows this method could meet the design requirements of flight control system conceptual prototype.
2011, 40(6): 892-897.
doi: 10.3969/j.issn.1001-0548.2011.06.017
Abstract:
Some traditional anomaly detection mechanisms (such as data mining and Bayes methods) have much poorer performance in terms of detection rate and false alarm rate because they consider only the normal behavior feature of IP networks, and neglect that of the abnormal behaviors. Motivate by the situations, this paper proposed a new characterization model of abnormal behaviors, and also developed an anomaly detection mechanism based on cusp-catastrophe for IP networks. They not only make the best of the prominent features of cusp-catastrophe in terms of multiple steady states and discontinuous catastrophe, and also can describe the normal behavior features and abnormal ones. Finally under Kdd-Cup 99 datasets, the proposed mechanism is evaluated, and the evaluation result shows that its detection rate and the false detection have greatly been improved compared with BN and C4.5.
Some traditional anomaly detection mechanisms (such as data mining and Bayes methods) have much poorer performance in terms of detection rate and false alarm rate because they consider only the normal behavior feature of IP networks, and neglect that of the abnormal behaviors. Motivate by the situations, this paper proposed a new characterization model of abnormal behaviors, and also developed an anomaly detection mechanism based on cusp-catastrophe for IP networks. They not only make the best of the prominent features of cusp-catastrophe in terms of multiple steady states and discontinuous catastrophe, and also can describe the normal behavior features and abnormal ones. Finally under Kdd-Cup 99 datasets, the proposed mechanism is evaluated, and the evaluation result shows that its detection rate and the false detection have greatly been improved compared with BN and C4.5.
2011, 40(6): 898-904.
doi: 10.3969/j.issn.1001-0548.2011.06.018
Abstract:
In order to meet the high demand of computation intensive and band-width exhausting media applications for embedded system, a heterogeneous multi-core embedded visual media processor, named EVMPSoC, is proposed. The chip consists of a main processor called EPStar3, which is a 32 bit RISC embedded processor, and two SIMD coprocessor, which are designed by application-specific instruction set. According to the communication characteristics of media applications, the hierachy high/low speed bus and dual band-width parallel memory access with multi-channel are used as the on-chip bus Architecture of EVMPSoC. The chip was taped out sucessful using SMIC 0.13 μm LVT CMOS technology and packaged by Amkor with PBGA 400. It runs well at peak frequency 416 MHz, and shows its high efficiency and avaliability.
In order to meet the high demand of computation intensive and band-width exhausting media applications for embedded system, a heterogeneous multi-core embedded visual media processor, named EVMPSoC, is proposed. The chip consists of a main processor called EPStar3, which is a 32 bit RISC embedded processor, and two SIMD coprocessor, which are designed by application-specific instruction set. According to the communication characteristics of media applications, the hierachy high/low speed bus and dual band-width parallel memory access with multi-channel are used as the on-chip bus Architecture of EVMPSoC. The chip was taped out sucessful using SMIC 0.13 μm LVT CMOS technology and packaged by Amkor with PBGA 400. It runs well at peak frequency 416 MHz, and shows its high efficiency and avaliability.
2011, 40(6): 905-910.
doi: 10.3969/j.issn.1001-0548.2011.06.019
Abstract:
An orthogonal non-negative CANDECOMP/PARAFAC factorization algorithm (ONNCP) is proposed. With the orthogonal constrain, the low-dimensional presentations of samples are kept non-negative in ONNCP. The relationship between NNCP and NMF is analyzed theoretically. The solution process and the convergence of the algorithm are discussed. The experiments indicate that, compared with other non-negative factorization algorithms, ONNCP can reduce the redundancy of the base images and enhance the sparseness of the base images due to its orthogonality. It also ensures the low-dimensional feature is non-negative. The algorithm can achieve better recognition rate in facial expression recognition and will convergence a fixed point. Furthermore, the algorithm can be generalized to any order tensor.
An orthogonal non-negative CANDECOMP/PARAFAC factorization algorithm (ONNCP) is proposed. With the orthogonal constrain, the low-dimensional presentations of samples are kept non-negative in ONNCP. The relationship between NNCP and NMF is analyzed theoretically. The solution process and the convergence of the algorithm are discussed. The experiments indicate that, compared with other non-negative factorization algorithms, ONNCP can reduce the redundancy of the base images and enhance the sparseness of the base images due to its orthogonality. It also ensures the low-dimensional feature is non-negative. The algorithm can achieve better recognition rate in facial expression recognition and will convergence a fixed point. Furthermore, the algorithm can be generalized to any order tensor.
2011, 40(6): 911-914.
doi: 10.3969/j.issn.1001-0548.2011.06.020
Abstract:
According to the drawback of the background value in GM(1,1) model, a new method for computing background value is proposed which uses the non-homogeneous exponential law to fit the accumulated sequence in order to get the area where the practical curve is on as background value. Experiment by using traffic data of Shenzhen between 8:00 to 8:50 on October 9 and 10 in 2007 shows that fitting precision and prediction precision of the new reconstructing background value of GM(1,1) model are improved.
According to the drawback of the background value in GM(1,1) model, a new method for computing background value is proposed which uses the non-homogeneous exponential law to fit the accumulated sequence in order to get the area where the practical curve is on as background value. Experiment by using traffic data of Shenzhen between 8:00 to 8:50 on October 9 and 10 in 2007 shows that fitting precision and prediction precision of the new reconstructing background value of GM(1,1) model are improved.
2011, 40(6): 915-920.
doi: 10.3969/j.issn.1001-0548.2011.06.021
Abstract:
If obvious objects in an image are deleted, the image would be meaningless. The protection of the image contents is imminence. A novel image watermarking scheme combined with the behavior of primary visual cortex is proposed. The Berkeley wavelet transform shares many characteristics with the receptive fields of neurons in mammalian primary visual cortex (V1). The watermark is embedded into the region-of-interest (ROI) position of an image. Independent component analysis (ICA) is adapted to the watermark extraction procedure and the original image is not necessary. Experimental results demonstrate that the proposed watermarking technique successfully locates ROI position automatically, has a good vision performance and robustness against some attacks as JPEG compress, noises, and little rotation and scale, luminance and contrast enhancement and low pass filtering. More importantly, the proposed scheme can extract the watermarks with high quality even if majority background of an original image is removed.
If obvious objects in an image are deleted, the image would be meaningless. The protection of the image contents is imminence. A novel image watermarking scheme combined with the behavior of primary visual cortex is proposed. The Berkeley wavelet transform shares many characteristics with the receptive fields of neurons in mammalian primary visual cortex (V1). The watermark is embedded into the region-of-interest (ROI) position of an image. Independent component analysis (ICA) is adapted to the watermark extraction procedure and the original image is not necessary. Experimental results demonstrate that the proposed watermarking technique successfully locates ROI position automatically, has a good vision performance and robustness against some attacks as JPEG compress, noises, and little rotation and scale, luminance and contrast enhancement and low pass filtering. More importantly, the proposed scheme can extract the watermarks with high quality even if majority background of an original image is removed.
2011, 40(6): 921-926.
doi: 10.3969/j.issn.1001-0548.2011.06.022
Abstract:
Power consumption in embedded systems is becoming a hot issue that attracts more attention. Different assemble instruction set, software algorithm, and high-level software architecture can significantly affect the system energy consumption. In this paper, we firstly analyze the relations between software energy consumption and some software characteristics on algorithm level. Through measuring three software characteristics, i.e., average time complexity, space complexity, and input scale, we propose a BP neural network software power model based on algorithm complexity. Then, we design and train a kind of BP neural network to accomplish energy consumption function approximation. Simulation experiment results show that the error between the estimation value of this energy consumption function and the real energy consumption value is below 10%. Therefore, it could quickly estimate the energy consumption of software in some input scale, which is an important fundament to explore the energy consumption optimization in the future.
Power consumption in embedded systems is becoming a hot issue that attracts more attention. Different assemble instruction set, software algorithm, and high-level software architecture can significantly affect the system energy consumption. In this paper, we firstly analyze the relations between software energy consumption and some software characteristics on algorithm level. Through measuring three software characteristics, i.e., average time complexity, space complexity, and input scale, we propose a BP neural network software power model based on algorithm complexity. Then, we design and train a kind of BP neural network to accomplish energy consumption function approximation. Simulation experiment results show that the error between the estimation value of this energy consumption function and the real energy consumption value is below 10%. Therefore, it could quickly estimate the energy consumption of software in some input scale, which is an important fundament to explore the energy consumption optimization in the future.
2011, 40(6): 927-932.
doi: 10.3969/j.issn.1001-0548.2011.06.023
Abstract:
Simulation is a very important way to computer architecture research. To reduce simulation time, the dynamic binary translation (DBT) technique has bee proposed to extract representative simulation points of programs. The performance characteristics of programs can then be obtained by simulating only the extracted representative simulation points. However, the accuracy implications of the DBT method have not been studied. For some programs, the DBT method can incur a nearly 20% error in the simulation results. In this paper, the cause of the execution trace deviation is studied carefully and several methods are proposed to overcome the problem. Experimental results show that with the proposed methods, the accuracy of the DBT method is the same as the simulation method.
Simulation is a very important way to computer architecture research. To reduce simulation time, the dynamic binary translation (DBT) technique has bee proposed to extract representative simulation points of programs. The performance characteristics of programs can then be obtained by simulating only the extracted representative simulation points. However, the accuracy implications of the DBT method have not been studied. For some programs, the DBT method can incur a nearly 20% error in the simulation results. In this paper, the cause of the execution trace deviation is studied carefully and several methods are proposed to overcome the problem. Experimental results show that with the proposed methods, the accuracy of the DBT method is the same as the simulation method.
2011, 40(6): 933-936.
doi: 10.3969/j.issn.1001-0548.2011.06.024
Abstract:
MoO3 films were deposited on glass substrates by electron beam evaporation method. X-ray diffraction patterns revealed that the prepared samples presented orthorhombic structure with preferred orientation of . As substrate temperature increased, the grain size increases firstly then decreases, but increases once more after annealing. Moreover, the diffraction peak of (040) plane shifts to lower angle monotonically with increased substrate temperature. Scanning electron microscope images show needle-like grains and their size becomes more uniform after annealing. With increased subsutrate temperature, the sample optical absorption edge in ultraviolet-visible spectra shift to longer wavelength and the transmittance also decreases. On the other hand, the optical gap reduces from 2.93 eV to 2.59 eV when substrate temperature increases from 100 ℃ to 300 ℃, and even reduces to 2.46 eV after air annealing under 500 ℃.
MoO3 films were deposited on glass substrates by electron beam evaporation method. X-ray diffraction patterns revealed that the prepared samples presented orthorhombic structure with preferred orientation of . As substrate temperature increased, the grain size increases firstly then decreases, but increases once more after annealing. Moreover, the diffraction peak of (040) plane shifts to lower angle monotonically with increased substrate temperature. Scanning electron microscope images show needle-like grains and their size becomes more uniform after annealing. With increased subsutrate temperature, the sample optical absorption edge in ultraviolet-visible spectra shift to longer wavelength and the transmittance also decreases. On the other hand, the optical gap reduces from 2.93 eV to 2.59 eV when substrate temperature increases from 100 ℃ to 300 ℃, and even reduces to 2.46 eV after air annealing under 500 ℃.
2011, 40(6): 937-940.
doi: 10.3969/j.issn.1001-0548.2011.06.025
Abstract:
A novel design methodology for programmable logic cores (PLCs) is proposed. A structured hardware description strategy is applied where some repeatable tiles are created and then replicated to form the fabric of the PLC, afterwards, the semi-custom design flow is used for layout generation. Compared with the existing ‘soft' PLCs design methodlogy, the problem of ‘combinational cycle’ is solved through bypassing the process of synthesis. Experimental results show that the area of PLCs designed following this methodology is reduced by about 50%. The feasibility of this methodology is verified through the implementation of desired logic functionality on the proof-of-concept prototype chip.
A novel design methodology for programmable logic cores (PLCs) is proposed. A structured hardware description strategy is applied where some repeatable tiles are created and then replicated to form the fabric of the PLC, afterwards, the semi-custom design flow is used for layout generation. Compared with the existing ‘soft' PLCs design methodlogy, the problem of ‘combinational cycle’ is solved through bypassing the process of synthesis. Experimental results show that the area of PLCs designed following this methodology is reduced by about 50%. The feasibility of this methodology is verified through the implementation of desired logic functionality on the proof-of-concept prototype chip.
2011, 40(6): 941-945.
doi: 10.3969/j.issn.1001-0548.2011.06.026
Abstract:
The influence of refractive index distribution on the frequency property of optical modes in random media is analyzed by using the finite difference time domain method to numerically solve Maxwell equations. The results indicate that the number and the spectral intensity of the modes existed in random media is growing up with the increase of the randomness intensity of the refractive index distribution. The minor variety of the refractive index of scattering granules also can conduce to the increase of the number of the modes in random media. The spectral intensity of these modes is also determined by the excursion, frequency, and velocity of the variety of refractive index of scattering granules. Anisotropic random media can achieve the modulation of the refractive index in terms of time and space. The agreement between experimental observations and simulations is helpful in understanding the physical mechanism in random systems.
The influence of refractive index distribution on the frequency property of optical modes in random media is analyzed by using the finite difference time domain method to numerically solve Maxwell equations. The results indicate that the number and the spectral intensity of the modes existed in random media is growing up with the increase of the randomness intensity of the refractive index distribution. The minor variety of the refractive index of scattering granules also can conduce to the increase of the number of the modes in random media. The spectral intensity of these modes is also determined by the excursion, frequency, and velocity of the variety of refractive index of scattering granules. Anisotropic random media can achieve the modulation of the refractive index in terms of time and space. The agreement between experimental observations and simulations is helpful in understanding the physical mechanism in random systems.
2011, 40(6): 946-950.
doi: 10.3969/j.issn.1001-0548.2011.06.027
Abstract:
The thermal diffusivities and thermal effusivities of metal materials have been measured by the Photoacoustic Technique. Firstly, based on the simplified R-G model, the one-dimensional theory of photoacoustic was analyzed. The corresponding experimental system has been set up, by which the photoaoustic signal of the solid photoacoustic cell can be obtained. Secondly, the thermal diffusivity and thermal effusivity of reference samples were measured, by which the experimental system can be validated. Finally, the diffusivities and effusivities of several kinds of metal material were studied by the photoacoustic cell system. Comparing with the corresponding reference data, it is shown that the photoacoustic techniques can be used to measure both the thermal diffusivity and effusivity of metal materials effectively. The relative error between the experiment data and the reference data are less than 3%.
The thermal diffusivities and thermal effusivities of metal materials have been measured by the Photoacoustic Technique. Firstly, based on the simplified R-G model, the one-dimensional theory of photoacoustic was analyzed. The corresponding experimental system has been set up, by which the photoaoustic signal of the solid photoacoustic cell can be obtained. Secondly, the thermal diffusivity and thermal effusivity of reference samples were measured, by which the experimental system can be validated. Finally, the diffusivities and effusivities of several kinds of metal material were studied by the photoacoustic cell system. Comparing with the corresponding reference data, it is shown that the photoacoustic techniques can be used to measure both the thermal diffusivity and effusivity of metal materials effectively. The relative error between the experiment data and the reference data are less than 3%.
2011, 40(6): 951-955.
doi: 10.3969/j.issn.1001-0548.2011.06.028
Abstract:
The radio frequency interference (RFI) is a dominant factor to degrade the image quality in microwave medical imaging systems. This paper analyzes the effects of RFI incurred in microwave medical imaging systems, and proposes a phase random coding method to transform RFI into zero mean random signal. The RFI can be suppressed easily using traditional mean value filter algorithm. A set of 2D breast cancer imaging experiments with simulated data demonstrate the proposed method is feasible and effective.
The radio frequency interference (RFI) is a dominant factor to degrade the image quality in microwave medical imaging systems. This paper analyzes the effects of RFI incurred in microwave medical imaging systems, and proposes a phase random coding method to transform RFI into zero mean random signal. The RFI can be suppressed easily using traditional mean value filter algorithm. A set of 2D breast cancer imaging experiments with simulated data demonstrate the proposed method is feasible and effective.
2011, 40(6): 956-960.
Abstract:
For the data mining based on time series estimation, the existed studies reveal that the Intra-Cranial Pressure (ICP) time series cannot be well estimated when the linear mapping function is used to delineate the relationship between error and feature. To improve the accuracy for ICP estimation, the non-linear support vector regression (SVR) is used to construct the nonlinear function between feature and error. The experiment results showed that the SVR based mapping function is superior to the linear least square based one.
For the data mining based on time series estimation, the existed studies reveal that the Intra-Cranial Pressure (ICP) time series cannot be well estimated when the linear mapping function is used to delineate the relationship between error and feature. To improve the accuracy for ICP estimation, the non-linear support vector regression (SVR) is used to construct the nonlinear function between feature and error. The experiment results showed that the SVR based mapping function is superior to the linear least square based one.
