Email alert
1989 Vol. 11, No. 5
Display Method:
1989, 11(5): 449-457.
Abstract:
The problems for calculating near-fields of EM radiation systems by using Finite Difference Time-Domain (FD-TD) method are discussed and the annular phased array of dipole antennas is simulated numerically by use of FD/TD method. For a test run the near-field and current distribution of the single dipole antenna are calculated. The near-field of the annular phased array of dipole antennas in central region filled with deionized water is computed and the interaction of near-field with a anatomically-based model of buman torso is considered as well.
The problems for calculating near-fields of EM radiation systems by using Finite Difference Time-Domain (FD-TD) method are discussed and the annular phased array of dipole antennas is simulated numerically by use of FD/TD method. For a test run the near-field and current distribution of the single dipole antenna are calculated. The near-field of the annular phased array of dipole antennas in central region filled with deionized water is computed and the interaction of near-field with a anatomically-based model of buman torso is considered as well.
1989, 11(5): 458-467.
Abstract:
A design technique for microstrip arrays is discussed. The technique is ap-plicable to comb travelling-wave microstrip arrays with any beam direction. Because of the travelling-wave make-up and specific design, the broader bandwidth and the lower side lobe level can be achieved. Two design examples of squint-beam travelling comb array are presented and the arrays are shown of good electrical characteristics by experiments.
A design technique for microstrip arrays is discussed. The technique is ap-plicable to comb travelling-wave microstrip arrays with any beam direction. Because of the travelling-wave make-up and specific design, the broader bandwidth and the lower side lobe level can be achieved. Two design examples of squint-beam travelling comb array are presented and the arrays are shown of good electrical characteristics by experiments.
1989, 11(5): 468-476.
Abstract:
The concepts of modified graphs of a composite graph with respect to two vertex-pairs and a hyperedge-decomposition are introduced. By using these concepts and the directed hypergraph theory, the topological formulae for the parameter-extraction theorem and subnetwork-extraction theorems are derived, then the topological formulae for multiterminal feedback networks are presented. In these formulae the parameters of the feedback subnetwork are separated from that of the fundamental subnetwork, so that it is convenient to find out the effect of the feedback parameters. Furthermore, since one network is decomposed into two smaller subnetworks, the computing time complexity and space complexity can be reduced.
The concepts of modified graphs of a composite graph with respect to two vertex-pairs and a hyperedge-decomposition are introduced. By using these concepts and the directed hypergraph theory, the topological formulae for the parameter-extraction theorem and subnetwork-extraction theorems are derived, then the topological formulae for multiterminal feedback networks are presented. In these formulae the parameters of the feedback subnetwork are separated from that of the fundamental subnetwork, so that it is convenient to find out the effect of the feedback parameters. Furthermore, since one network is decomposed into two smaller subnetworks, the computing time complexity and space complexity can be reduced.
1989, 11(5): 477-483.
Abstract:
The applications of the segmentation and the desegmentation methods to the planar circuits, particularly, to the ones with short boundaries, are further studied. In addition, a general node voltage equation of segmentation method for a planar circuit which can he segmented into many segments is derivd.
The applications of the segmentation and the desegmentation methods to the planar circuits, particularly, to the ones with short boundaries, are further studied. In addition, a general node voltage equation of segmentation method for a planar circuit which can he segmented into many segments is derivd.
1989, 11(5): 484-489.
Abstract:
The satisfactory approximate expression of the differential amplifier output characteristic is found by computer plotting. The synchronization band ot a differential IL circuit is derived and analysed by the frequency domain method, where the maximum synchronization band of the same frequency IL is 2/(0.372-0.019875 A2) time as large as the famous Adler s formula. It is found that the theoretical analysis agrees well with experimental curve.
The satisfactory approximate expression of the differential amplifier output characteristic is found by computer plotting. The synchronization band ot a differential IL circuit is derived and analysed by the frequency domain method, where the maximum synchronization band of the same frequency IL is 2/(0.372-0.019875 A2) time as large as the famous Adler s formula. It is found that the theoretical analysis agrees well with experimental curve.
1989, 11(5): 490-499.
Abstract:
The equivalent circuit model of microwave SGFET, the large signal analysis of LO injection by harmonic blance method and the small signal analysis of RF injection by multi-frequency conversion matrix method are discussed. A complete method for analysing and designing mixers in presented. The measurement results show that this method is simple and effective.
The equivalent circuit model of microwave SGFET, the large signal analysis of LO injection by harmonic blance method and the small signal analysis of RF injection by multi-frequency conversion matrix method are discussed. A complete method for analysing and designing mixers in presented. The measurement results show that this method is simple and effective.
1989, 11(5): 500-508.
Abstract:
The close relationship between the perfect binary arrays and higher dimensional hadamard matrices is discovered. Some very interesting results in Fourier transform of perfect binary arrays are shown and a few open problems are also pointed out.
The close relationship between the perfect binary arrays and higher dimensional hadamard matrices is discovered. Some very interesting results in Fourier transform of perfect binary arrays are shown and a few open problems are also pointed out.
1989, 11(5): 509-517.
Abstract:
According to the binary nature of image, a new maximum entropy restoration method with binary constraint is proposed The properties of existence and uniqueness of solution are discussed. The problem of maximum of entropy with double constraints is solved and corresponding algorithm is given. Maximum bounded entropy principle is employed concerning the prior knowledge of binary image, the maximum bounded entropy restoration method with binary constraint is put forword. The proposed methods, Wiener filter restoration method and maximum entropy restoration method are compared, it is proved by experiment that maximum entropy restoration method with binary constraint and maximum bounded entropy restoration method with binarv constraint can improve the quality of blur binary image.
According to the binary nature of image, a new maximum entropy restoration method with binary constraint is proposed The properties of existence and uniqueness of solution are discussed. The problem of maximum of entropy with double constraints is solved and corresponding algorithm is given. Maximum bounded entropy principle is employed concerning the prior knowledge of binary image, the maximum bounded entropy restoration method with binary constraint is put forword. The proposed methods, Wiener filter restoration method and maximum entropy restoration method are compared, it is proved by experiment that maximum entropy restoration method with binary constraint and maximum bounded entropy restoration method with binarv constraint can improve the quality of blur binary image.
1989, 11(5): 518-527.
Abstract:
The rational form of Helmholtz theorem in resonant cavity is discussed. On the basis of the theorem, it is proved that the electric irrotational field can be expanded into a convergent series on L vector wave function, and that the electric rotational field also can be expanded into a convergent series on M and N vector wave functions. So it is proved that L,M and N vector wave functions consist of a complete orthogonal basis for electric field in resonant cavity.
The rational form of Helmholtz theorem in resonant cavity is discussed. On the basis of the theorem, it is proved that the electric irrotational field can be expanded into a convergent series on L vector wave function, and that the electric rotational field also can be expanded into a convergent series on M and N vector wave functions. So it is proved that L,M and N vector wave functions consist of a complete orthogonal basis for electric field in resonant cavity.
1989, 11(5): 528-531.
Abstract:
A linear theory of the circular free-electron laser (FEL) with electrostatic focusing system is presented. The dispersion equation is derived. The maximum growth rate of FEL instability is found from the dispersion equation. Finally, the related problems are discussed.
A linear theory of the circular free-electron laser (FEL) with electrostatic focusing system is presented. The dispersion equation is derived. The maximum growth rate of FEL instability is found from the dispersion equation. Finally, the related problems are discussed.
1989, 11(5): 532-535.
Abstract:
A nonuniform back-projection reconstruction method for computerized tomography is presented. In the uniform back-projection reconstruction method the projection difference-between a measured ray sum and a computed ray sum is uniformly distributed to the cells in the ray, but in the nonuniform back-projection reconstruction method the projection difference is nonuniformly distributed to the cells based on the weighted grey coefficient of the cell. A qualitative improvement of reconstructed images by the new reconstruction method is evident, especially in the incomplete projection cases.
A nonuniform back-projection reconstruction method for computerized tomography is presented. In the uniform back-projection reconstruction method the projection difference-between a measured ray sum and a computed ray sum is uniformly distributed to the cells in the ray, but in the nonuniform back-projection reconstruction method the projection difference is nonuniformly distributed to the cells based on the weighted grey coefficient of the cell. A qualitative improvement of reconstructed images by the new reconstruction method is evident, especially in the incomplete projection cases.
1989, 11(5): 536-541.
Abstract:
Strict relations between the multiplication in transform domain and the convolution in time domain are derived for odd and even discrete sine and cosine transforms of version I and II, thus it is possible to use these orthogonal transforms of the sinusoidal family in digital filtering.
Strict relations between the multiplication in transform domain and the convolution in time domain are derived for odd and even discrete sine and cosine transforms of version I and II, thus it is possible to use these orthogonal transforms of the sinusoidal family in digital filtering.
1989, 11(5): 542-546.
Abstract:
A concept of general transconductance of V/Q (voltage/charge) transformation i.. proposed, and applied to the simulation of switching capacitance impedance. By Using V/Q transverter, the expected impedance function can be implemented by means of volrage transfer function. Starting from this concept, the switched capacitor-FDNR (frequency dependent negative resistor) circuits of three kinds of S/Z transformation (i.e. forward differential SfZ transformation, backward differential S/Z transformation, bilinear S/Z transformation) are respectively deduced. If the selected voltage transfer function circuit is not sensitive to stray capacitance, then the implemented simulation impedance circuit is also not sensitive to stray capacitance. As an example, a resonant loop is composed of forward differential FDNR circuit. The experiments show that the characteristics of its frequency response coincide with the theoretical analysis.
A concept of general transconductance of V/Q (voltage/charge) transformation i.. proposed, and applied to the simulation of switching capacitance impedance. By Using V/Q transverter, the expected impedance function can be implemented by means of volrage transfer function. Starting from this concept, the switched capacitor-FDNR (frequency dependent negative resistor) circuits of three kinds of S/Z transformation (i.e. forward differential SfZ transformation, backward differential S/Z transformation, bilinear S/Z transformation) are respectively deduced. If the selected voltage transfer function circuit is not sensitive to stray capacitance, then the implemented simulation impedance circuit is also not sensitive to stray capacitance. As an example, a resonant loop is composed of forward differential FDNR circuit. The experiments show that the characteristics of its frequency response coincide with the theoretical analysis.
1989, 11(5): 547-550.
Abstract:
A high-precision regulated power supply for a new type electron beam lithography system with 30 kV high voltage is recommended. The design scheme of double closed loop regulation and the design scheme of centralized compensation combined with dispersed compensation are adopted. In accordance with the technical requirements of regulated power supply, some key technigues have been properly handled. All it s qualifications satisfy or exceed the original design requirements.
A high-precision regulated power supply for a new type electron beam lithography system with 30 kV high voltage is recommended. The design scheme of double closed loop regulation and the design scheme of centralized compensation combined with dispersed compensation are adopted. In accordance with the technical requirements of regulated power supply, some key technigues have been properly handled. All it s qualifications satisfy or exceed the original design requirements.
1989, 11(5): 551-556.
Abstract:
A simple model is used to describle the fractional phase-locking of a relaxation oscillator by periodically applied impulses. For any definite fraction p/q, analytic expressions are derived for the phase-locking region in the parameter space. Experimental results for a bidirectional negative resistance transistor (BNRT) relaxation oscillator agree with the computed ones.
A simple model is used to describle the fractional phase-locking of a relaxation oscillator by periodically applied impulses. For any definite fraction p/q, analytic expressions are derived for the phase-locking region in the parameter space. Experimental results for a bidirectional negative resistance transistor (BNRT) relaxation oscillator agree with the computed ones.
1989, 11(5): 557-560.
Abstract:
The deep level in InGaAsP/InP DH LED have been studied by DLTS method. The results show that when the p-n junction located in the interface of the p-InP/n-In GaAsP, there is deep level in some InGaAsP/InP DH LED. The activation energy △E is 0.24eV.
The deep level in InGaAsP/InP DH LED have been studied by DLTS method. The results show that when the p-n junction located in the interface of the p-InP/n-In GaAsP, there is deep level in some InGaAsP/InP DH LED. The activation energy △E is 0.24eV.
