Email alert
1996 Vol. 18, No. 2
Display Method:
1996, 18(2): 113-120.
Abstract:
A method of automatically extracting and optimizing fuzy rule base is presented. Firstly, it applies the method of CCM (Cell-to-Cell Mapping) to analyze the evolving trend and global behavior of the closed-loop fuzzy dynamical system based on a cell state space. Secondly, each rule base is evaluated to determine a performance index according to the control objective. Thirdly, GA (Genetic Algorithm) updates the old population of coded rule bases using the performance index of each rule base generation by generation. A complicated nonlinear system---an inverted pendulum is simulated to illustrate the details of the algorithms and the results demonstrate the validity of the method.
A method of automatically extracting and optimizing fuzy rule base is presented. Firstly, it applies the method of CCM (Cell-to-Cell Mapping) to analyze the evolving trend and global behavior of the closed-loop fuzzy dynamical system based on a cell state space. Secondly, each rule base is evaluated to determine a performance index according to the control objective. Thirdly, GA (Genetic Algorithm) updates the old population of coded rule bases using the performance index of each rule base generation by generation. A complicated nonlinear system---an inverted pendulum is simulated to illustrate the details of the algorithms and the results demonstrate the validity of the method.
1996, 18(2): 121-126.
Abstract:
The cost function for eigenstuctures extration is discussed in detail, one can obtain the largest eigenvector by minimizing the cost function. In order to obtain the other eigenvectors, a covariance matrix series is constructed. If one compares the cost function with the energy function of a neural network, the neural network can be introduced to extract the eigenvectors. Theoretical analysis and simulations show that the proposed method is reasonable and feasible.
The cost function for eigenstuctures extration is discussed in detail, one can obtain the largest eigenvector by minimizing the cost function. In order to obtain the other eigenvectors, a covariance matrix series is constructed. If one compares the cost function with the energy function of a neural network, the neural network can be introduced to extract the eigenvectors. Theoretical analysis and simulations show that the proposed method is reasonable and feasible.
1996, 18(2): 127-134.
Abstract:
Centralized/distributed recursive algorithms for temporal-spatial information integration are discussed by using the Dempster-shafer technique.Compared with the Bayesiar approach ,the Dempster-shafer technique has a strong capability of handling information uncertainties.All information is pooled into the central processor in the centralized integration algorithm.In contrast,the distributed algorithm shares the computation burden among the local processors,which increase the computational efficiency.The developed algorithms are applied to a target identification problem with two sensors:millimeter wave radiometer,infrared searching and tracking
Centralized/distributed recursive algorithms for temporal-spatial information integration are discussed by using the Dempster-shafer technique.Compared with the Bayesiar approach ,the Dempster-shafer technique has a strong capability of handling information uncertainties.All information is pooled into the central processor in the centralized integration algorithm.In contrast,the distributed algorithm shares the computation burden among the local processors,which increase the computational efficiency.The developed algorithms are applied to a target identification problem with two sensors:millimeter wave radiometer,infrared searching and tracking
1996, 18(2): 135-144.
Abstract:
A pair of multichannel recursive least squares (RLS) adaptive lattice algorithms based on the order recursive properties of lattice filters and the superior numerical properties of the Givens algorithms is derived. The derivation of the first algorithm is based on the QR decomposition of the input data matrix directly, and Givens rotations approach is used to compute the QR decomposition. Using first a prerotation of the input data matrix and then a repetition of the single channel Givens lattice algorithm, the second algorithm can be obtained. Both algorithms have superior numerical properties, particularly the robustness to wordlength limitations. The parameter vector to be estimated can be extracted directly from internal variables in the present algorithms without a backsolve operation with an extra triangular array. The results of computer simulation of the parameter identification of a two-channel system are presented to confirm efficientlly the derivation.
A pair of multichannel recursive least squares (RLS) adaptive lattice algorithms based on the order recursive properties of lattice filters and the superior numerical properties of the Givens algorithms is derived. The derivation of the first algorithm is based on the QR decomposition of the input data matrix directly, and Givens rotations approach is used to compute the QR decomposition. Using first a prerotation of the input data matrix and then a repetition of the single channel Givens lattice algorithm, the second algorithm can be obtained. Both algorithms have superior numerical properties, particularly the robustness to wordlength limitations. The parameter vector to be estimated can be extracted directly from internal variables in the present algorithms without a backsolve operation with an extra triangular array. The results of computer simulation of the parameter identification of a two-channel system are presented to confirm efficientlly the derivation.
1996, 18(2): 145-151.
Abstract:
An adaptive statistic model for mobile communication channel is studied and simulated. According to the parameters of the model set up, an error sequence describing the long burst error characteristics of the mobile channel environment is generated on computer. A test method using threshold technique is presented to verify the accuracy of the adaptive channel model. The simulation results show that the adaptive Markov model with three states is more accurate and practical than the conventional mobile channel model and is a feasible scheme for optimization of mobile communication channel model.
An adaptive statistic model for mobile communication channel is studied and simulated. According to the parameters of the model set up, an error sequence describing the long burst error characteristics of the mobile channel environment is generated on computer. A test method using threshold technique is presented to verify the accuracy of the adaptive channel model. The simulation results show that the adaptive Markov model with three states is more accurate and practical than the conventional mobile channel model and is a feasible scheme for optimization of mobile communication channel model.
1996, 18(2): 152-157.
Abstract:
The leaky bucket algorithm for priority sources in ATM network is derived and the analytical expressions of the relation among the leaky bucket performance, the leaky bucket parameters and the statistical parameters of input traffic are obtained. The adjustment factor of constant serving rate for lower priority service, , is introduced and its effect on the performance of the higher priority and the lower priority services are studied.
The leaky bucket algorithm for priority sources in ATM network is derived and the analytical expressions of the relation among the leaky bucket performance, the leaky bucket parameters and the statistical parameters of input traffic are obtained. The adjustment factor of constant serving rate for lower priority service, , is introduced and its effect on the performance of the higher priority and the lower priority services are studied.
1996, 18(2): 158-163.
Abstract:
This paper descibes all nearest-neighbors relations of feature submatrices in fault verification techniques with a Voronoi hypersphere, analyses deveation disturbance angles between the feature submatrices and an unitary basis submatrix of the -dimension maximal orientation-energy subspace of a measured voltgage-change matrix through disturbance theories. Based on the above, the paper proposes novel diagnosibility theories for the system of fault verification techniques, derives a necessary and sufficient diagnosability condition in verification techniques for analog fault diagnosis under toralence disturbance. In addition, the paper classifies diagnosis problems with the revealed diagnossability.
This paper descibes all nearest-neighbors relations of feature submatrices in fault verification techniques with a Voronoi hypersphere, analyses deveation disturbance angles between the feature submatrices and an unitary basis submatrix of the -dimension maximal orientation-energy subspace of a measured voltgage-change matrix through disturbance theories. Based on the above, the paper proposes novel diagnosibility theories for the system of fault verification techniques, derives a necessary and sufficient diagnosability condition in verification techniques for analog fault diagnosis under toralence disturbance. In addition, the paper classifies diagnosis problems with the revealed diagnossability.
1996, 18(2): 164-170.
Abstract:
The strengths of the ELF currents produced by heating the iono-spheres over Beijing, Shanghai and Haikou by powerfully modulated HF wave are calculated under a typical transmitting condition, using the practical parameters of the ionospheres. The start point of the calculation is the electron energy equation, however, a new iterative method is developed to treat the pulse modulation. This algorithm is also applied to study the heated properties of the ionosphere at typical middle latitudes and the results are compared with that by previous authors for high latitudes.
The strengths of the ELF currents produced by heating the iono-spheres over Beijing, Shanghai and Haikou by powerfully modulated HF wave are calculated under a typical transmitting condition, using the practical parameters of the ionospheres. The start point of the calculation is the electron energy equation, however, a new iterative method is developed to treat the pulse modulation. This algorithm is also applied to study the heated properties of the ionosphere at typical middle latitudes and the results are compared with that by previous authors for high latitudes.
1996, 18(2): 171-177.
Abstract:
The transmission characteristics of high-order hybrid modes in a bianisotropic loss slab bounded by perfectly conducting planes are investigated, and in such complex waveguides, the transverse field components are also expressed in terms of the longitudinal field components of hybrid modes. Numerical results are presented to demonstrate the effects of anisotropy, lossness and different constitutive parameters on the dispersion and field-distribution behavious, and it is suggested that bianisotropic materials have potential applications for the microwave and millimeter wave regimes.
The transmission characteristics of high-order hybrid modes in a bianisotropic loss slab bounded by perfectly conducting planes are investigated, and in such complex waveguides, the transverse field components are also expressed in terms of the longitudinal field components of hybrid modes. Numerical results are presented to demonstrate the effects of anisotropy, lossness and different constitutive parameters on the dispersion and field-distribution behavious, and it is suggested that bianisotropic materials have potential applications for the microwave and millimeter wave regimes.
1996, 18(2): 178-182.
Abstract:
According to the behaviours of the whispering gallery modes, a new type of integrated planar whispering gallery mode dielectric resonator is discussed. The characteristic equation is derived by applying the method proposed by X. H. Jiao(1987). The calculated values show in good agreement with that in the previous papers. The results obtained show that this type of resonator is very suitable for millimeter wave integrated circuits.
According to the behaviours of the whispering gallery modes, a new type of integrated planar whispering gallery mode dielectric resonator is discussed. The characteristic equation is derived by applying the method proposed by X. H. Jiao(1987). The calculated values show in good agreement with that in the previous papers. The results obtained show that this type of resonator is very suitable for millimeter wave integrated circuits.
1996, 18(2): 183-188.
Abstract:
This paper reports the utilization of a new type of planar whispering-gallery modes dielectric resonators and microstrip line for the design of Ka-band bandstop and directional filters. The analysis is based on the traveling wave ring resonator. Finally, experimental results of bandstop and directrional filters in Ka-band are given, which show such resonators are very suitable for use in millimeter-wave microstrip integrated circuits.
This paper reports the utilization of a new type of planar whispering-gallery modes dielectric resonators and microstrip line for the design of Ka-band bandstop and directional filters. The analysis is based on the traveling wave ring resonator. Finally, experimental results of bandstop and directrional filters in Ka-band are given, which show such resonators are very suitable for use in millimeter-wave microstrip integrated circuits.
1996, 18(2): 189-193.
Abstract:
The radar cross-section (RCS) of antennas can be divided in to two parts, one is the common structural scattering component, another is the antenna-mode scattering component produced by antenna s re-radiation. Three kinds of antenna s RCS are discussed and the measured antenna s RCS reduction by the method of minimizing the mode scattering component are given. The experimental results show that the RCS reduction is about 10-15 dB both in the operating band and outside the band. The technique expressed in th...
The radar cross-section (RCS) of antennas can be divided in to two parts, one is the common structural scattering component, another is the antenna-mode scattering component produced by antenna s re-radiation. Three kinds of antenna s RCS are discussed and the measured antenna s RCS reduction by the method of minimizing the mode scattering component are given. The experimental results show that the RCS reduction is about 10-15 dB both in the operating band and outside the band. The technique expressed in th...
1996, 18(2): 194-201.
Abstract:
Based on the analysis of DC characteristics of silicon bipolar transistors at low temperature, DC analytic models of ECL circuit at low temperature are derived, then compared with the experimental data and computer simulation results. The modification of SPICE BJT model about temperature and design of low temperature ECL circuit are discussed.
Based on the analysis of DC characteristics of silicon bipolar transistors at low temperature, DC analytic models of ECL circuit at low temperature are derived, then compared with the experimental data and computer simulation results. The modification of SPICE BJT model about temperature and design of low temperature ECL circuit are discussed.
1996, 18(2): 209-212.
Abstract:
By use of Hopfield model and basis solution of homogeneous linear equations which are established in accordance with consistent state, a practical decision method for the existence of optimal Hopfield model of combinational circuits is provided. Finally, several examples are given.
By use of Hopfield model and basis solution of homogeneous linear equations which are established in accordance with consistent state, a practical decision method for the existence of optimal Hopfield model of combinational circuits is provided. Finally, several examples are given.
1996, 18(2): 213-216.
Abstract:
Leapfrog technique is discussed and used to realize current-mode lowpass active RC ladder structure. The current-mode lowpass filter realized by this method only takes CCII, CCI as active devices, and all RC elements axe grounded, therefore, this structure is simpler than the relating voltage-mode one. It also has low sensitivity as the other ladder structures do.
Leapfrog technique is discussed and used to realize current-mode lowpass active RC ladder structure. The current-mode lowpass filter realized by this method only takes CCII, CCI as active devices, and all RC elements axe grounded, therefore, this structure is simpler than the relating voltage-mode one. It also has low sensitivity as the other ladder structures do.
1996, 18(2): 217-220.
Abstract:
The effect of Ge doped in CZSi on the precipitation and the defect-free zone (DFZ) formation in Ge-doped CZSi wafers after a three step gettering annealing was studied. It is found that Ge not only can promote the out-diffusion of oxygen and form wider DFZ in Ge-Doped CZSi wafer than that in the control sample, but also can suppress the formation of precipitation.
The effect of Ge doped in CZSi on the precipitation and the defect-free zone (DFZ) formation in Ge-doped CZSi wafers after a three step gettering annealing was studied. It is found that Ge not only can promote the out-diffusion of oxygen and form wider DFZ in Ge-Doped CZSi wafer than that in the control sample, but also can suppress the formation of precipitation.
1996, 18(2): 221-224.
Abstract:
Multi-beam klystron (MBK) is a new type of high power microwave amplifier. The S-band multi-beam broadband klystron doveloped in Institute of Electronics of Academia Sinica (IEAS) is described in this paper. The research progresses on the design and calculation of the tube, multi-beam electron gun and focusing system, broadband output circuit are given. For beam voltage of 19kV and bias voltage of control electrode of --7kV, the measured performance of the tube are as follows: the beam current of 32.25A, beam perveance of 12.3P, DC beam transmission of 91.8, peak power of 252kW, average power of 3.5kW, the efficiency of 41, the gain of 49dB, the band width of 9.1 with 1.7dB power variation.
Multi-beam klystron (MBK) is a new type of high power microwave amplifier. The S-band multi-beam broadband klystron doveloped in Institute of Electronics of Academia Sinica (IEAS) is described in this paper. The research progresses on the design and calculation of the tube, multi-beam electron gun and focusing system, broadband output circuit are given. For beam voltage of 19kV and bias voltage of control electrode of --7kV, the measured performance of the tube are as follows: the beam current of 32.25A, beam perveance of 12.3P, DC beam transmission of 91.8, peak power of 252kW, average power of 3.5kW, the efficiency of 41, the gain of 49dB, the band width of 9.1 with 1.7dB power variation.
1996, 18(2): 202-208.
Abstract:
Driven by the increasing need to integrated analog and digital circuits on the same silicon chip, design automation for analog integrated ckcuits has gained huge momentum and close attention in recent years. Kernel subjects and achievements in topology selection, device sizing, and layout synthesis are reviewed in this paper.
Driven by the increasing need to integrated analog and digital circuits on the same silicon chip, design automation for analog integrated ckcuits has gained huge momentum and close attention in recent years. Kernel subjects and achievements in topology selection, device sizing, and layout synthesis are reviewed in this paper.
