Email alert
2003 Vol. 25, No. 7
Display Method:
2003, 25(7): 865-870.
Abstract:
The concept of Embedded Text-To-Speech (ETTS) system is first discussed. Then a real time realized Embedded Chinese Text-To-Speech (ECTTS) system based on DSP is introduced. Based on the DSP realization results, the VLSI design scheme of the ECTTS system is discussed. For the ECTTS system front-end processing, a VLSI design scheme based on the dynamic memory management is suggested. For the ECTTS system post-end processing, a VLSI design architecture based on the decoded speech frame is also proposed. The memory and the bus structure of the VLSI design of ECTTS system are discussed.
The concept of Embedded Text-To-Speech (ETTS) system is first discussed. Then a real time realized Embedded Chinese Text-To-Speech (ECTTS) system based on DSP is introduced. Based on the DSP realization results, the VLSI design scheme of the ECTTS system is discussed. For the ECTTS system front-end processing, a VLSI design scheme based on the dynamic memory management is suggested. For the ECTTS system post-end processing, a VLSI design architecture based on the decoded speech frame is also proposed. The memory and the bus structure of the VLSI design of ECTTS system are discussed.
2003, 25(7): 871-877.
Abstract:
As an important aspect in the study of attacks against information hiding and the development of anti-attack techniques, this paper describes a method to compare a pair of perceptibly identical digital images or audio signals in order to single out the one that contains embedded hidden-information. By means of statistic analyses and comparison between the two signals, and making use of short-term correlativity in the space/time domain, three criteria are proposed, which are mutually complementary, and can be used in statistical tests to identify signals with embedded data. The experimental results are satisfactory.
As an important aspect in the study of attacks against information hiding and the development of anti-attack techniques, this paper describes a method to compare a pair of perceptibly identical digital images or audio signals in order to single out the one that contains embedded hidden-information. By means of statistic analyses and comparison between the two signals, and making use of short-term correlativity in the space/time domain, three criteria are proposed, which are mutually complementary, and can be used in statistical tests to identify signals with embedded data. The experimental results are satisfactory.
2003, 25(7): 878-883.
Abstract:
The theory of detecting ridges in the modulus of the continuous wavelet transform is presented as well as reconstructing signal by using information on ridges. Application of the theory on speech signal is discussed in detail. Experimental result shows that as for signals with highest frequency at 4kHz, 3~5 ridges is enough to describe all of the information contained in the signal. The reconstructed signal reserves most of the necessary information of the original signal.
The theory of detecting ridges in the modulus of the continuous wavelet transform is presented as well as reconstructing signal by using information on ridges. Application of the theory on speech signal is discussed in detail. Experimental result shows that as for signals with highest frequency at 4kHz, 3~5 ridges is enough to describe all of the information contained in the signal. The reconstructed signal reserves most of the necessary information of the original signal.
2003, 25(7): 884-891.
Abstract:
In this paper, the cyclostationary property of output signal of the system is discussed, when a stationary(or cyclostationary) signal passes a linear period time variant(or time invariant) system. Two kinds of time variant adaptive filter are presented, which can be used in linear period time variant system. At same time, an algorithm for period of LPTV system is also presented. Simulation results indicate the efficiency of these algorithms.
In this paper, the cyclostationary property of output signal of the system is discussed, when a stationary(or cyclostationary) signal passes a linear period time variant(or time invariant) system. Two kinds of time variant adaptive filter are presented, which can be used in linear period time variant system. At same time, an algorithm for period of LPTV system is also presented. Simulation results indicate the efficiency of these algorithms.
2003, 25(7): 892-895.
Abstract:
This paper proposes a new algorithm for joint diagonalization of positive definite symmetric matrices and shows that it can be approximated by an independent component analysis problem. The good performance of the algorithm is illustrated with an example by compared with other approaches.
This paper proposes a new algorithm for joint diagonalization of positive definite symmetric matrices and shows that it can be approximated by an independent component analysis problem. The good performance of the algorithm is illustrated with an example by compared with other approaches.
2003, 25(7): 896-900.
Abstract:
Blind separation of sources consists of recovering a set of signals of which only-instantaneous linear mixing is observed. This paper presents a novel blind source separation method when noises are complex isotropic SaS process. SaS processes can describe many-signals and noises with impulsive nature, but its second order and high order statistics are infinite, so, a subspace approach is used to process the observed data, then joint approximate diagonalization of eigen-matrices is used to estimate the mixing matrix and source signals. Computer simulation shows the high performance of the proposed method.
Blind separation of sources consists of recovering a set of signals of which only-instantaneous linear mixing is observed. This paper presents a novel blind source separation method when noises are complex isotropic SaS process. SaS processes can describe many-signals and noises with impulsive nature, but its second order and high order statistics are infinite, so, a subspace approach is used to process the observed data, then joint approximate diagonalization of eigen-matrices is used to estimate the mixing matrix and source signals. Computer simulation shows the high performance of the proposed method.
2003, 25(7): 901-906.
Abstract:
A symmetrized weight enumerator of Zk-linear code is defined in this paper. By using discreate Hadamard transform, a symmetrized MacWilliams identity between Zk-linear code and its dual code is given.
A symmetrized weight enumerator of Zk-linear code is defined in this paper. By using discreate Hadamard transform, a symmetrized MacWilliams identity between Zk-linear code and its dual code is given.
2003, 25(7): 907-917.
Abstract:
In this paper, the theory of Hamilton array codes for high dimension and big Hamming distance is proposed. The model of logical topology structure for d- 2k-l,d=2k(k 3) types and the existence topology condition for general high order SM array codes are analyzed. The construction principle and the synthesis algorithm of SM array codes for d=2k- 1 and d=2k(k 3) high dimension and high order are discussed. Finally, the distinction between the two models and the example design are given. The result provides a new direction for a class of new type array ciphers research.
In this paper, the theory of Hamilton array codes for high dimension and big Hamming distance is proposed. The model of logical topology structure for d- 2k-l,d=2k(k 3) types and the existence topology condition for general high order SM array codes are analyzed. The construction principle and the synthesis algorithm of SM array codes for d=2k- 1 and d=2k(k 3) high dimension and high order are discussed. Finally, the distinction between the two models and the example design are given. The result provides a new direction for a class of new type array ciphers research.
2003, 25(7): 918-924.
Abstract:
The problem of estimating the Direction-Of-Arrival(DOA) of multiple source sig-nals far from the sensors using a novel adaptive antenna suggested by Du Huiping(1999) in the presence of spatially white Gaussian noise is addressed based on the Maximum Likelihood(ML) function approach. The effects of the Signal to Noise Ratio(SNR), snapshots and space of sen-sors are investigated. The statistical performance of ML estimators as well as the Cramer-Rao Bound(CRB) are analyzed. Comparison of the simulation results with the other articles shows the validity of the theoretical analysis.
The problem of estimating the Direction-Of-Arrival(DOA) of multiple source sig-nals far from the sensors using a novel adaptive antenna suggested by Du Huiping(1999) in the presence of spatially white Gaussian noise is addressed based on the Maximum Likelihood(ML) function approach. The effects of the Signal to Noise Ratio(SNR), snapshots and space of sen-sors are investigated. The statistical performance of ML estimators as well as the Cramer-Rao Bound(CRB) are analyzed. Comparison of the simulation results with the other articles shows the validity of the theoretical analysis.
2003, 25(7): 925-930.
Abstract:
With the development of communication techniques and digital signal processing techniques, the capability of the communication receiver to identify signal parameters should be improved. In this paper, a blind receiver for MDPSK signal is proposed. Under the condition of not knowing the modulation of the received signal, the receiver could find the symbol timing and identify the modulation of the signal, realizing signal s blind receiving.
With the development of communication techniques and digital signal processing techniques, the capability of the communication receiver to identify signal parameters should be improved. In this paper, a blind receiver for MDPSK signal is proposed. Under the condition of not knowing the modulation of the received signal, the receiver could find the symbol timing and identify the modulation of the signal, realizing signal s blind receiving.
2003, 25(7): 931-937.
Abstract:
This paper puts forward a new multilevel chaotic sequence and a multifunction circuit using three-cell CNN chaos, combining the characteristic of CNN with that of the Spread Spectrum System (SSS). It also proposes a scheme of generating the Multilevel Chaotic Sequence (MCS) and analyzes its properties. Simulation results show that this kind of sequence performs quite well, which is suitable for future large-capacity SSS.
This paper puts forward a new multilevel chaotic sequence and a multifunction circuit using three-cell CNN chaos, combining the characteristic of CNN with that of the Spread Spectrum System (SSS). It also proposes a scheme of generating the Multilevel Chaotic Sequence (MCS) and analyzes its properties. Simulation results show that this kind of sequence performs quite well, which is suitable for future large-capacity SSS.
2003, 25(7): 938-942.
Abstract:
In this paper, the optimization of channel capacity (equivalent to data throughput) is proposed for variable-rate service, which is solved with the Lagrangian multiplier method based on a simplified BER formula for m-QAM modulation. The closed-form solution for the optimization problem is given and the practical constraint of positive power and modulation mode allocation is solved by null transmission on those deep-faded subcarriers. Finally discrete modulation mode (0, QPSK, 16QAM and 64QAM) is used to approach the optimally allocated continuous modulation mode, and simulation results demonstrate that the capacity loss from this approaching is negligible and QoS is guaranteed.
In this paper, the optimization of channel capacity (equivalent to data throughput) is proposed for variable-rate service, which is solved with the Lagrangian multiplier method based on a simplified BER formula for m-QAM modulation. The closed-form solution for the optimization problem is given and the practical constraint of positive power and modulation mode allocation is solved by null transmission on those deep-faded subcarriers. Finally discrete modulation mode (0, QPSK, 16QAM and 64QAM) is used to approach the optimally allocated continuous modulation mode, and simulation results demonstrate that the capacity loss from this approaching is negligible and QoS is guaranteed.
2003, 25(7): 943-948.
Abstract:
In long-code used DS/CDMA communication systems, the convergence of perfor-mance is limited by the number of users in multi-stage Linear Soft Decision Parallel Interference Cancellation (LSD-PIC) multiuser detectors. In this paper, the precondition method in matrix computation is proposed to be applied to improve the convergence of iteration and overcome the convergence breakdown in the case of high load for the LSD-PIC multiuser detectors. Further-more, the polynomial preconditioner which could be obtained by low computational complexity is proposed, and the optimal coefficients for zero-order and first order polynomial preconditioner are deduced.
In long-code used DS/CDMA communication systems, the convergence of perfor-mance is limited by the number of users in multi-stage Linear Soft Decision Parallel Interference Cancellation (LSD-PIC) multiuser detectors. In this paper, the precondition method in matrix computation is proposed to be applied to improve the convergence of iteration and overcome the convergence breakdown in the case of high load for the LSD-PIC multiuser detectors. Further-more, the polynomial preconditioner which could be obtained by low computational complexity is proposed, and the optimal coefficients for zero-order and first order polynomial preconditioner are deduced.
2003, 25(7): 949-955.
Abstract:
This paper describes a novel shared memory switch architecture in order to adapt to local area and metroplitan area network s multiple transport speed structures. This new switch can provide variable port speed, variable packet switch, and port s self-synchronization. This paper also describes queue management for variable packet.
This paper describes a novel shared memory switch architecture in order to adapt to local area and metroplitan area network s multiple transport speed structures. This new switch can provide variable port speed, variable packet switch, and port s self-synchronization. This paper also describes queue management for variable packet.
2003, 25(7): 956-962.
Abstract:
Induction mechanism, extraction methods and classification efficiency of Jet En-gine Modulation (JEM) signatures in the aircraft returns of the conventional radar are studied in this paper. Firstly, induction mechanism of JEM signatures in the aircraft returns of the con-ventional radar is analyzed theoretically, and some differences of JEM signatures in the returns from different aircrafts including helicopter, propeller and turbojet are compared. Secondly, advantages and disadvantages of the analysis and extraction of JEM signatures are discussed in the time-domain, frequency-domain and time-frequency domain. Thirdly, a target classification and identification test is accomplished by using JEM periodic feature extracted by AR bispec-trum slice. The experimented results demonstrate that in target classification JEM periodic feature is effective but limited in the conventional radar.
Induction mechanism, extraction methods and classification efficiency of Jet En-gine Modulation (JEM) signatures in the aircraft returns of the conventional radar are studied in this paper. Firstly, induction mechanism of JEM signatures in the aircraft returns of the con-ventional radar is analyzed theoretically, and some differences of JEM signatures in the returns from different aircrafts including helicopter, propeller and turbojet are compared. Secondly, advantages and disadvantages of the analysis and extraction of JEM signatures are discussed in the time-domain, frequency-domain and time-frequency domain. Thirdly, a target classification and identification test is accomplished by using JEM periodic feature extracted by AR bispec-trum slice. The experimented results demonstrate that in target classification JEM periodic feature is effective but limited in the conventional radar.
2003, 25(7): 963-968.
Abstract:
This paper illustrates that Hopping Frequency (HF) communication and frequency agile radar have chaotic peculiarities, and are chaotic phenomena, with computing singular attractor and their correlative dimensions and Lyapunov exponents from simulation. It has important significance to recognizing HF communication and frequency agile radar on theory and essence.
This paper illustrates that Hopping Frequency (HF) communication and frequency agile radar have chaotic peculiarities, and are chaotic phenomena, with computing singular attractor and their correlative dimensions and Lyapunov exponents from simulation. It has important significance to recognizing HF communication and frequency agile radar on theory and essence.
2003, 25(7): 969-974.
Abstract:
The Method of Moment (MoM) in conjunction with Pade Approximation In Seg-ments (PAIS) is applied to predict the monostatic Radar Cross-Section (RCS) pattern of an arbitrarily shaped 2-D dielectric object. In the method, integro-differential equation for the total electric field is used. In MoM procedure, to reduce the need for computer memory and to speed up the iteration, a combined algorithm of conjugate gradient method and fast Fourier transform is used to solve the algebraic system. PAIS is used to accelerate the drawing of RCS pattern. PAIS here is equipped with a procedure for selecting expanding points, and so short-comings of original scheme of man-selecting location of expansion points are overcome. The new method is called CG-FFT-PAIS. Numerical results in this paper demonstrate the validity and the practicability of the method.
The Method of Moment (MoM) in conjunction with Pade Approximation In Seg-ments (PAIS) is applied to predict the monostatic Radar Cross-Section (RCS) pattern of an arbitrarily shaped 2-D dielectric object. In the method, integro-differential equation for the total electric field is used. In MoM procedure, to reduce the need for computer memory and to speed up the iteration, a combined algorithm of conjugate gradient method and fast Fourier transform is used to solve the algebraic system. PAIS is used to accelerate the drawing of RCS pattern. PAIS here is equipped with a procedure for selecting expanding points, and so short-comings of original scheme of man-selecting location of expansion points are overcome. The new method is called CG-FFT-PAIS. Numerical results in this paper demonstrate the validity and the practicability of the method.
2003, 25(7): 975-981.
Abstract:
The eigenvalues of compound waveguide system are solved by using the method of dyadic Greens function by matching the electric and magnetic field on a virtual boundary. As the chosen dyadic Greens function has no singularity and can be calculated by scalar Greens function, the method can be applied not only in the computation of dominant and higher modes for TE and TM waves, but also in the computation of compound system with twin modes, In addition, some theorems in classical field theory such as rotation theorem on surface are discussed in this paper.
The eigenvalues of compound waveguide system are solved by using the method of dyadic Greens function by matching the electric and magnetic field on a virtual boundary. As the chosen dyadic Greens function has no singularity and can be calculated by scalar Greens function, the method can be applied not only in the computation of dominant and higher modes for TE and TM waves, but also in the computation of compound system with twin modes, In addition, some theorems in classical field theory such as rotation theorem on surface are discussed in this paper.
2003, 25(7): 982-989.
Abstract:
A new kind of microwave cavity-strongly coupling two-gap cavities, which can be used in broadband multi-beam klystrons that work in lower frequency band, is designed in this paper. The important parameters of the cavity are calculated with ISFEL3D, a reliable program for solving three dimension electric-magnetic field problems, and the characteristics of this kind of cavity are analyzed carefully. It is showed that strongly coupling two-gap cavities have some prominent virtues, such as higher characteristic resistance and smaller volume when they work in tt mode. Besides, the frequency interval of n mode and 2tt mode is much wider. Furthermore, two kinds of concentrated element equivalent circuits are used in this paper to prove the calculated data and conclusions, meanwhile the precision of two equivalent circuits is discussed.
A new kind of microwave cavity-strongly coupling two-gap cavities, which can be used in broadband multi-beam klystrons that work in lower frequency band, is designed in this paper. The important parameters of the cavity are calculated with ISFEL3D, a reliable program for solving three dimension electric-magnetic field problems, and the characteristics of this kind of cavity are analyzed carefully. It is showed that strongly coupling two-gap cavities have some prominent virtues, such as higher characteristic resistance and smaller volume when they work in tt mode. Besides, the frequency interval of n mode and 2tt mode is much wider. Furthermore, two kinds of concentrated element equivalent circuits are used in this paper to prove the calculated data and conclusions, meanwhile the precision of two equivalent circuits is discussed.
2003, 25(7): 990-994.
Abstract:
This paper proposes a method for narrowband interference suppression in Direct Sequence Spread System(DS-SS). The distribution of the transform domain data is analysed, and it is approximated by Gaussian distribution. The variance of Gaussian distribution is the sum of the variance of the spread spectrum signal and the variance of the noise, so if it; is estimated, it can be used as the threshold in designing the interference suppression filter. Maximum Likelihood (ML) technique is used for estimating the variance to achieve the auto-matic threshold setup. Theoretical analysis and simulation show that this method has good performance in suppressing the narrowband interference in DS-SS.
This paper proposes a method for narrowband interference suppression in Direct Sequence Spread System(DS-SS). The distribution of the transform domain data is analysed, and it is approximated by Gaussian distribution. The variance of Gaussian distribution is the sum of the variance of the spread spectrum signal and the variance of the noise, so if it; is estimated, it can be used as the threshold in designing the interference suppression filter. Maximum Likelihood (ML) technique is used for estimating the variance to achieve the auto-matic threshold setup. Theoretical analysis and simulation show that this method has good performance in suppressing the narrowband interference in DS-SS.
2003, 25(7): 995-999.
Abstract:
The effect of mutual coupling among array elements on the performance of MUSIC algorithm is investigated. The presence of mutual coupling distorts array response vector and the eigenstructure of the covariance matrix. The Method of Moment (MoM) is used to evaluate the mutual coupling among the dipoles. The admittance matrix is then used to eliminate the effects of mutual coupling. Finally, the results of computer simulation are given.
The effect of mutual coupling among array elements on the performance of MUSIC algorithm is investigated. The presence of mutual coupling distorts array response vector and the eigenstructure of the covariance matrix. The Method of Moment (MoM) is used to evaluate the mutual coupling among the dipoles. The admittance matrix is then used to eliminate the effects of mutual coupling. Finally, the results of computer simulation are given.
2003, 25(7): 1000-1005.
Abstract:
In this paper the channel capacity in MIMO system is introduced, then the effects of antenna selection in the transmitter and receiver on channel capacity are analyzed. Simulation results indicate that the transmit antenna selection can increase channel capacity, the selection of the receiver and transmitter antennas simultaneously can reduce complexity, but cannot improve the channel capacity.
In this paper the channel capacity in MIMO system is introduced, then the effects of antenna selection in the transmitter and receiver on channel capacity are analyzed. Simulation results indicate that the transmit antenna selection can increase channel capacity, the selection of the receiver and transmitter antennas simultaneously can reduce complexity, but cannot improve the channel capacity.
