Email alert
1996 Vol. 18, No. 5
Display Method:
1996, 18(5): 449-454.
Abstract:
The convergence of the inverse correlation matrix iteration algorithm is analyzed. In order to avoid computing the inverse correlation matrix in the algorithm, this paper presents an eigenvalue shift iteration algorithm, and analyzes its convergence. Finally, the simulated results are given.
The convergence of the inverse correlation matrix iteration algorithm is analyzed. In order to avoid computing the inverse correlation matrix in the algorithm, this paper presents an eigenvalue shift iteration algorithm, and analyzes its convergence. Finally, the simulated results are given.
1996, 18(5): 455-461.
Abstract:
A new quadratic time-frequency distribution (TFD) with a compound kernel is proposed and a comparative study of several popular quadratic TFD is carried out. It is shown that the new TFD with compound kernel has stronger ability than the exponential distribution(ED) and the cone-shaped kernel distribution(CKD) in reducing cross terms, meanwhile almost not decreasing the time-frequency resolution of ED or CKD.
A new quadratic time-frequency distribution (TFD) with a compound kernel is proposed and a comparative study of several popular quadratic TFD is carried out. It is shown that the new TFD with compound kernel has stronger ability than the exponential distribution(ED) and the cone-shaped kernel distribution(CKD) in reducing cross terms, meanwhile almost not decreasing the time-frequency resolution of ED or CKD.
1996, 18(5): 462-466.
Abstract:
The multisensor detection area partitioning is considered. An approach is presented to the fusion in each detection area where the sensor uses different thresholds and then at system level. The expressions of the detection probability and false alarm probability are given. An apph cation of the method is illustrated to distributed CFAR detection systems. The results show that the system detection probability may be improved by setting different thresholds for a detector.
The multisensor detection area partitioning is considered. An approach is presented to the fusion in each detection area where the sensor uses different thresholds and then at system level. The expressions of the detection probability and false alarm probability are given. An apph cation of the method is illustrated to distributed CFAR detection systems. The results show that the system detection probability may be improved by setting different thresholds for a detector.
1996, 18(5): 467-472.
Abstract:
This paper presents a new Constant-False-Alarm Ratio(CFAR) detector based on Ordered Statistics(OS) and Cell-Averaging(CA) forming local estimates, and using Greatest-Option (GO) to form interference power level estimate Z in test cell (OSCAGO). Under the Swerling II assumption, the analytic expressions of false-alarm probability (Pfa), detection probability (Pd) and Average Decision Threshold (ADT) of this detector are derived, its detection performances in homogeneous background and in strong interfering targets environment are analyzed, and compared with that of OS-, GOSGO-CFARs. Results show that the detection performances of OSCAGO in homogeneous background and in multiple-target situations are all obviously better than those of OS and GOSGO. When the number of interfering targets is equal to certain value, the CFAR loss of OSCAGO is about 3dB less than that of GOSGO.
This paper presents a new Constant-False-Alarm Ratio(CFAR) detector based on Ordered Statistics(OS) and Cell-Averaging(CA) forming local estimates, and using Greatest-Option (GO) to form interference power level estimate Z in test cell (OSCAGO). Under the Swerling II assumption, the analytic expressions of false-alarm probability (Pfa), detection probability (Pd) and Average Decision Threshold (ADT) of this detector are derived, its detection performances in homogeneous background and in strong interfering targets environment are analyzed, and compared with that of OS-, GOSGO-CFARs. Results show that the detection performances of OSCAGO in homogeneous background and in multiple-target situations are all obviously better than those of OS and GOSGO. When the number of interfering targets is equal to certain value, the CFAR loss of OSCAGO is about 3dB less than that of GOSGO.
1996, 18(5): 473-478.
Abstract:
An ATM network is expected to support a large number of bursty traffic sources. In this paper, bursty traffic is described by the Interrupted Bernoulli Process (IBP). With the use of a recursive algorithm, the cell loss probability and the average delay for ATM switching of mixed continuous-bit-rate and bursty traffic are exactly analyzed. Both numerical analytical and simulation results show that the performance of ATM switching with traffic smoothing function is improved.
An ATM network is expected to support a large number of bursty traffic sources. In this paper, bursty traffic is described by the Interrupted Bernoulli Process (IBP). With the use of a recursive algorithm, the cell loss probability and the average delay for ATM switching of mixed continuous-bit-rate and bursty traffic are exactly analyzed. Both numerical analytical and simulation results show that the performance of ATM switching with traffic smoothing function is improved.
1996, 18(5): 479-486.
Abstract:
The coding gains, that is, self-gain and mutual-gain with their variations for memories are defined and the formulas for computing the gains are derived, so that the criteria, which will enable the system designers for choosing best code, are given. Finally, the numerical results of some examples are analysed and discussed.
The coding gains, that is, self-gain and mutual-gain with their variations for memories are defined and the formulas for computing the gains are derived, so that the criteria, which will enable the system designers for choosing best code, are given. Finally, the numerical results of some examples are analysed and discussed.
1996, 18(5): 487-495.
Abstract:
It is well known that any cryptographic system can be described by a nonlinear furc-tion. This paper studies the nonlinearity of Boolean functions and the effect of certain operations on the nonlinearity of Boolean functions by using the spectral techniques. Finally, the applicability of the results to the cryptography is indicated.
It is well known that any cryptographic system can be described by a nonlinear furc-tion. This paper studies the nonlinearity of Boolean functions and the effect of certain operations on the nonlinearity of Boolean functions by using the spectral techniques. Finally, the applicability of the results to the cryptography is indicated.
1996, 18(5): 496-500.
Abstract:
By using the model of air-oil film-sea water system, the influences of the oil thickness, operating frequency of the radiometer and incidence angle of the beam on the effective emissivity are calculated and analysed. The optimal scheme for passive microwave remote sensing of sea-surface oil pollution is also proposed and used successfully for airborne microwave remote sensing experiments.
By using the model of air-oil film-sea water system, the influences of the oil thickness, operating frequency of the radiometer and incidence angle of the beam on the effective emissivity are calculated and analysed. The optimal scheme for passive microwave remote sensing of sea-surface oil pollution is also proposed and used successfully for airborne microwave remote sensing experiments.
1996, 18(5): 501-507.
Abstract:
B-spline is applied to the reconstruction of 2-D conductor cross-section via backscat-tered field in near region illuminated by plane waves with different incident directions. The whole inverse problem may be viewed as a two-stage optimal procedure, and the optimal algorithm based on gradient, such as BFGS algorithm, is adopted as the gradient of the objective functions which may be derived by using Frechet differential. In the first step, the position and the radius of optimal circular cylinder are obtained, then in the second step, with above results as its initial values, the shape of 2-D cylinder can be reconstructed. The robustness of the proposed method is tested using different numerical results, and the effects of noise are discussed.
B-spline is applied to the reconstruction of 2-D conductor cross-section via backscat-tered field in near region illuminated by plane waves with different incident directions. The whole inverse problem may be viewed as a two-stage optimal procedure, and the optimal algorithm based on gradient, such as BFGS algorithm, is adopted as the gradient of the objective functions which may be derived by using Frechet differential. In the first step, the position and the radius of optimal circular cylinder are obtained, then in the second step, with above results as its initial values, the shape of 2-D cylinder can be reconstructed. The robustness of the proposed method is tested using different numerical results, and the effects of noise are discussed.
1996, 18(5): 508-512.
Abstract:
The interaction of human body with penetrated electromagnetic fields through aperture in conducting surface is considered by using the finite-difference time-domain method and inhomogeneous man models. First, the characteristic of penetrated electromagnetic fields through aperture is studied. The comparison of obtained results with those by use of analytical method shows that the method used in this paper is available. The contributions of electromagnetic fields and absorption energy in human body are calculated when the aperture is irradiated by plane wave and the electromagnetic man model in sitting posture is placed to other side of the surface near the aperture for different polarization and frequency of incident place wave.
The interaction of human body with penetrated electromagnetic fields through aperture in conducting surface is considered by using the finite-difference time-domain method and inhomogeneous man models. First, the characteristic of penetrated electromagnetic fields through aperture is studied. The comparison of obtained results with those by use of analytical method shows that the method used in this paper is available. The contributions of electromagnetic fields and absorption energy in human body are calculated when the aperture is irradiated by plane wave and the electromagnetic man model in sitting posture is placed to other side of the surface near the aperture for different polarization and frequency of incident place wave.
1996, 18(5): 513-518.
Abstract:
The problems of the second harmonic and three coupled waves in homogeneous lossy nonlinear dielectric are analysed. Under the conditions of same damp and phase-matching, the exact solutions are derived.
The problems of the second harmonic and three coupled waves in homogeneous lossy nonlinear dielectric are analysed. Under the conditions of same damp and phase-matching, the exact solutions are derived.
1996, 18(5): 519-525.
Abstract:
An analysis of electromagnetic backscattering data using Wavelet Packet Transfrom (WPT) is presented. Due to its adaptive multiresolution property. WPT is well adapted to resolve the multiscale features of backscattering data. In particular, with respect to the consequence of analysis and computational complexity, WPT results in a better representation of backscattering data over Continuous Wavelet Tramsform (CWT) which was employed by II. Kim, H. Ling (1992, 1993). So WPT method makes a contribution to the time-frequency analysis of the electromagnetic wave. Furthermore, it is shown that WPT method offers a new approach for the further application of wavelet to instantaneous magnetic field.
An analysis of electromagnetic backscattering data using Wavelet Packet Transfrom (WPT) is presented. Due to its adaptive multiresolution property. WPT is well adapted to resolve the multiscale features of backscattering data. In particular, with respect to the consequence of analysis and computational complexity, WPT results in a better representation of backscattering data over Continuous Wavelet Tramsform (CWT) which was employed by II. Kim, H. Ling (1992, 1993). So WPT method makes a contribution to the time-frequency analysis of the electromagnetic wave. Furthermore, it is shown that WPT method offers a new approach for the further application of wavelet to instantaneous magnetic field.
1996, 18(5): 526-531.
Abstract:
A novel three layers microstrip antenna elect that has the advantages of wide band and high gain is proposed. The characteristic equation and the frequency characteristic formulas of the input stand wave ratio are obtained by using the spectral domain method and equivalent circuit method, respectively. With the aid of the numerical results, a C-band microstrip antenna element with band width of 16 (SWR1.5) or 25 (VSWR2) and gain of 10.2-11.3 dB is developed, which are much larger than the bandwidth of 5-6 and the gin of 6-7 dB of the common microstrip antenna element.
A novel three layers microstrip antenna elect that has the advantages of wide band and high gain is proposed. The characteristic equation and the frequency characteristic formulas of the input stand wave ratio are obtained by using the spectral domain method and equivalent circuit method, respectively. With the aid of the numerical results, a C-band microstrip antenna element with band width of 16 (SWR1.5) or 25 (VSWR2) and gain of 10.2-11.3 dB is developed, which are much larger than the bandwidth of 5-6 and the gin of 6-7 dB of the common microstrip antenna element.
1996, 18(5): 532-536.
Abstract:
This paper discusses the definition and properties of multivalued symmetric functions, and points out that a multivalued symmetric function can be decomposed according to the value of the function j. The subfunction L j corresponding to j will certainly be a symmetric function, and it may be expressed as the sum-of-products form of degenerated multivalued fundamental symmetric functions. Based on this consideration, the logic synthesis circuit realization for the multivalued symmetric functions based upon full-adders is proposed.
This paper discusses the definition and properties of multivalued symmetric functions, and points out that a multivalued symmetric function can be decomposed according to the value of the function j. The subfunction L j corresponding to j will certainly be a symmetric function, and it may be expressed as the sum-of-products form of degenerated multivalued fundamental symmetric functions. Based on this consideration, the logic synthesis circuit realization for the multivalued symmetric functions based upon full-adders is proposed.
1996, 18(5): 537-543.
Abstract:
Detailed analysis of transient characteristics of ECL circuits is performed in this paper, then a relatively exact propagation delay expression applied to all temperatures is presented. The cryogenic characteristics of some dominant parameters contributed to propagation delay are also discussed. The model achieved is suitable for optimum designs of high speed devices and circuits at all temperatures.
Detailed analysis of transient characteristics of ECL circuits is performed in this paper, then a relatively exact propagation delay expression applied to all temperatures is presented. The cryogenic characteristics of some dominant parameters contributed to propagation delay are also discussed. The model achieved is suitable for optimum designs of high speed devices and circuits at all temperatures.
1996, 18(5): 544-548.
Abstract:
Novel distributed parameter neural networks are proposed for solving partial differential equations, and their dynamic performances are studied in Hilbert space. The locally connected neural networks are obtained by separating distributed parameter neural networks. Two simulations are also given. Both theoretical and practical results illustrate that the distributed parameter neural networks are effective and efficient for solving partial differential equation problems.
Novel distributed parameter neural networks are proposed for solving partial differential equations, and their dynamic performances are studied in Hilbert space. The locally connected neural networks are obtained by separating distributed parameter neural networks. Two simulations are also given. Both theoretical and practical results illustrate that the distributed parameter neural networks are effective and efficient for solving partial differential equation problems.
1996, 18(5): 549-552.
Abstract:
This paper proposes an HB weighted adaptive time delay estimator. The theoretical analysis and the computer simulation show that this new method does not depend on the knowledge about input signals and it has a higher precision and faster convergence speed. Under low SNR conditions, the performances of this method are better than those of LMS TDE obviously.
This paper proposes an HB weighted adaptive time delay estimator. The theoretical analysis and the computer simulation show that this new method does not depend on the knowledge about input signals and it has a higher precision and faster convergence speed. Under low SNR conditions, the performances of this method are better than those of LMS TDE obviously.
1996, 18(5): 553-557.
Abstract:
The error performance of continuous phase modulation (CPM) with optimal coherent detection depends on the minimum squared Euclidean distance (MSED). In this paper, a reduced-state trellis for full response CPM is presented under the condition that the MSED is not reduced, and therefore the complexity of the Viterbi decoding is reduced.
The error performance of continuous phase modulation (CPM) with optimal coherent detection depends on the minimum squared Euclidean distance (MSED). In this paper, a reduced-state trellis for full response CPM is presented under the condition that the MSED is not reduced, and therefore the complexity of the Viterbi decoding is reduced.
1996, 18(5): 558-560.
Abstract:
A variational expression of the propagation coefficient of the dominant mode in twisted rectangular waveguide is derived on the basis of the theory of the nonstandard eigenvalue problems, and the numerical results are compared with that of perturbation approach. It illustrates that there is no infinite series in the variational expression, and the variational results are more accurate than the perturbational ones, especially when the twisted angular period becomes bigger.
A variational expression of the propagation coefficient of the dominant mode in twisted rectangular waveguide is derived on the basis of the theory of the nonstandard eigenvalue problems, and the numerical results are compared with that of perturbation approach. It illustrates that there is no infinite series in the variational expression, and the variational results are more accurate than the perturbational ones, especially when the twisted angular period becomes bigger.
