Email alert
2008 Vol. 30, No. 5
Display Method:
2008, 30(5): 1017-1021.
doi: 10.3724/SP.J.1146.2007.00578
Abstract:
Based on the difference of work mechanism between Frequency Modulated Continuous Wave (FMCW) Synthetic Aperture Radar (SAR) and Pulse Synthetic Aperture Radar, which leads to different form of received signal, this paper establishes the model of FMCW SAR echo, analyzes its characteristics, and reveals the Doppler frequency shift effect in range direction induced by the continuous motion while radar transmitting and receiving signal. The Residual Video Phase (RVP), inherent to Dechirp-on-receive system, is represented in the form of convolution and compensated for in the frequency domain. Also Modified Frequency Scaling Algorithm is presented. Compared to the conventional method, it almost adds no additional computation load. Finally, point target simulation validates the correctness and the effectiveness of the presented method.
Based on the difference of work mechanism between Frequency Modulated Continuous Wave (FMCW) Synthetic Aperture Radar (SAR) and Pulse Synthetic Aperture Radar, which leads to different form of received signal, this paper establishes the model of FMCW SAR echo, analyzes its characteristics, and reveals the Doppler frequency shift effect in range direction induced by the continuous motion while radar transmitting and receiving signal. The Residual Video Phase (RVP), inherent to Dechirp-on-receive system, is represented in the form of convolution and compensated for in the frequency domain. Also Modified Frequency Scaling Algorithm is presented. Compared to the conventional method, it almost adds no additional computation load. Finally, point target simulation validates the correctness and the effectiveness of the presented method.
2008, 30(5): 1022-1026.
doi: 10.3724/SP.J.1146.2006.01699
Abstract:
A wide-swath SAR scheme based on displaced phase center azimuth multibeam is presented in this paper. More than one azimuth sampling can be acquired during one pulse period in the SAR system with azimuth multibeam. Therefore, the pulse repetition frequency can be greatly reduced, which enables wide-swath SAR imaging. As the satellite velocity varies in different part of the orbit and PRF required for each view angle is different, the azimuth sampling is nonuniform in some orbit position. A signal reconstruction algorithm based on multi-channel sampling is proposed to eliminate nonuniform azimuth sampling. Compared with the wide-swath SAR based on high orbit satellite or distributed satellite, the proposed scheme is easier to implement.
A wide-swath SAR scheme based on displaced phase center azimuth multibeam is presented in this paper. More than one azimuth sampling can be acquired during one pulse period in the SAR system with azimuth multibeam. Therefore, the pulse repetition frequency can be greatly reduced, which enables wide-swath SAR imaging. As the satellite velocity varies in different part of the orbit and PRF required for each view angle is different, the azimuth sampling is nonuniform in some orbit position. A signal reconstruction algorithm based on multi-channel sampling is proposed to eliminate nonuniform azimuth sampling. Compared with the wide-swath SAR based on high orbit satellite or distributed satellite, the proposed scheme is easier to implement.
2008, 30(5): 1027-1031.
doi: 10.3724/SP.J.1146.2006.01446
Abstract:
In this paper, the statistics of the union Stokes parameters in Gaussian assumption is derived from the Whishart distribution, and through the polarization invariant the analytic expressions of the four Stokes parameters are presented. The results have a comparison with the conclusion derived before, which shows the presented expressions to be simple and exact. Then the statistical PDF(Probability Density Function) of the normalized Stokes parameters are showed ,which describes the variety of the polarization state in another way. All the above can be useful to speckle reduced, classification for terrain and target recognition in complex environments or other applications.
In this paper, the statistics of the union Stokes parameters in Gaussian assumption is derived from the Whishart distribution, and through the polarization invariant the analytic expressions of the four Stokes parameters are presented. The results have a comparison with the conclusion derived before, which shows the presented expressions to be simple and exact. Then the statistical PDF(Probability Density Function) of the normalized Stokes parameters are showed ,which describes the variety of the polarization state in another way. All the above can be useful to speckle reduced, classification for terrain and target recognition in complex environments or other applications.
2008, 30(5): 1032-1036.
doi: 10.3724/SP.J.1146.2007.00191
Abstract:
To investigate the clutter mitigation performance in non-uniform, non-linear, or sparse array radar, the number of Degrees of Freedom (DoF) estimate theory for band-limited signal is employed to estimate the clutter DoF for these array geometries. The proposed method can effectively predict the number of significant clutter eigenvalues, and reveals the relationship between the clutter DoF and the system parameters as well as the clutter scenario. It is found that in sparse array radar, the system DoF should be increased to reduce the sparse sampling section so that the clutter mitigation performance in such systems can be improved. Finally, the effectiveness of the proposed estimate method is verified with simulation.
To investigate the clutter mitigation performance in non-uniform, non-linear, or sparse array radar, the number of Degrees of Freedom (DoF) estimate theory for band-limited signal is employed to estimate the clutter DoF for these array geometries. The proposed method can effectively predict the number of significant clutter eigenvalues, and reveals the relationship between the clutter DoF and the system parameters as well as the clutter scenario. It is found that in sparse array radar, the system DoF should be increased to reduce the sparse sampling section so that the clutter mitigation performance in such systems can be improved. Finally, the effectiveness of the proposed estimate method is verified with simulation.
2008, 30(5): 1037-1040.
doi: 10.3724/SP.J.1146.2006.01626
Abstract:
For space debris radar imaging with the dimension smaller than radar range resolution, only Single Range Doppler Interferometry (SRDI) is discussed before. Considering movement features of space debris, a Single Range Matching Filtering (SRMF) approach for space debris radar imaging is proposed in this paper. The proposed approach is to obtain 2-D image of space debris by using a range unit cross-range echo data and by matching filtering the signals with different turning radius. Simulation results show that of the proposed approach, its computational load is less than that of SRDI, and its imaging effect is better.
For space debris radar imaging with the dimension smaller than radar range resolution, only Single Range Doppler Interferometry (SRDI) is discussed before. Considering movement features of space debris, a Single Range Matching Filtering (SRMF) approach for space debris radar imaging is proposed in this paper. The proposed approach is to obtain 2-D image of space debris by using a range unit cross-range echo data and by matching filtering the signals with different turning radius. Simulation results show that of the proposed approach, its computational load is less than that of SRDI, and its imaging effect is better.
2008, 30(5): 1041-1046.
doi: 10.3724/SP.J.1146.2006.01614
Abstract:
This paper starts from a signal model of a stationary receiver configuration in airborne bistatic SAR systems which are characterized by spatially separated transmitters and receivers. Through analyzing the system geometrical characteristic, the range cell migration, the influence of the difference on the range cell migration and the change of the linear FM rate both in range and azimuth direction in RD field are discussed. And then a nonlinear perturbation function is deduced to equalize the targets FM rates in azimuth direction. After that, an imaging algorithm based on the CSA for bistatic SAR with a stationary receiver is get, and the distorted image is corrected. Finally the validity and the feasibility of dealing with big mapping zone of the algorithm proposed with this paper is supported with point target simulation experiments.
This paper starts from a signal model of a stationary receiver configuration in airborne bistatic SAR systems which are characterized by spatially separated transmitters and receivers. Through analyzing the system geometrical characteristic, the range cell migration, the influence of the difference on the range cell migration and the change of the linear FM rate both in range and azimuth direction in RD field are discussed. And then a nonlinear perturbation function is deduced to equalize the targets FM rates in azimuth direction. After that, an imaging algorithm based on the CSA for bistatic SAR with a stationary receiver is get, and the distorted image is corrected. Finally the validity and the feasibility of dealing with big mapping zone of the algorithm proposed with this paper is supported with point target simulation experiments.
2008, 30(5): 1047-1051.
doi: 10.3724/SP.J.1146.2007.00754
Abstract:
A novel scheme of target enhancement in high resolution polarimetric SAR imagery is proposed in this paper, which is based on the improvement of polarization characteristics similarity between pixels of the same region in the high resolution POL-SAR imagery. Huynens decomposition, polarization orientation angle extraction and a further angle rotation transformation are employed to eliminate the effect of the speckle noise and random vector scattering on each pixels polarization characteristics. Comparing with the existing target enhancement methods, the novel scheme can improve the polarization characteristics similarity of the pixels in the same region to get a promising target enhancement effect. The validity of the novel method is indicated by experimental results with fully polarimetric SAR data sets.
A novel scheme of target enhancement in high resolution polarimetric SAR imagery is proposed in this paper, which is based on the improvement of polarization characteristics similarity between pixels of the same region in the high resolution POL-SAR imagery. Huynens decomposition, polarization orientation angle extraction and a further angle rotation transformation are employed to eliminate the effect of the speckle noise and random vector scattering on each pixels polarization characteristics. Comparing with the existing target enhancement methods, the novel scheme can improve the polarization characteristics similarity of the pixels in the same region to get a promising target enhancement effect. The validity of the novel method is indicated by experimental results with fully polarimetric SAR data sets.
2008, 30(5): 1052-1055.
doi: 10.3724/SP.J.1146.2006.01668
Abstract:
In this paper, a novel speckle suppression method is proposed, which is based on the joint of discrete wavelet transform and independent component analysis (ICA). This method employs ICA to process the wavelet coefficients of images, extracting the independent components which represent the uncontaminated images. Also this paper compares the suppressing performance of several standard wavelet bases，investigating their influences on suppressing the speckle. Experimental results based on SAR images of MSTAR dataset show that the proposed method suppresses speckle effectively, and outperforms ICA and Lee filter.
In this paper, a novel speckle suppression method is proposed, which is based on the joint of discrete wavelet transform and independent component analysis (ICA). This method employs ICA to process the wavelet coefficients of images, extracting the independent components which represent the uncontaminated images. Also this paper compares the suppressing performance of several standard wavelet bases，investigating their influences on suppressing the speckle. Experimental results based on SAR images of MSTAR dataset show that the proposed method suppresses speckle effectively, and outperforms ICA and Lee filter.
2008, 30(5): 1056-1059.
doi: 10.3724/SP.J.1146.2006.01651
Abstract:
Distributed satellites interferometric SAR is capable of detecting ground moving targets via along-track baseline. In this paper, the signal model and phase compensation for this INSAR system is analyzed firstly. A new method based on Chirp-Z transform for GMTI in range-Doppler domain is proposed. The received data of different channels can be co-registered precisely along azimuth direction after being transformed into Doppler domain via CZT, and moving target can be detected through interferometry procedure combined with spectrum phase compensation. In the end, the constraints of multistatic SAR configuration by spectrum phase compensation and target motion by spectrum co-registration are discussed. The final simulation demonstrates that this algorithm can detect moving target efficiently.
Distributed satellites interferometric SAR is capable of detecting ground moving targets via along-track baseline. In this paper, the signal model and phase compensation for this INSAR system is analyzed firstly. A new method based on Chirp-Z transform for GMTI in range-Doppler domain is proposed. The received data of different channels can be co-registered precisely along azimuth direction after being transformed into Doppler domain via CZT, and moving target can be detected through interferometry procedure combined with spectrum phase compensation. In the end, the constraints of multistatic SAR configuration by spectrum phase compensation and target motion by spectrum co-registration are discussed. The final simulation demonstrates that this algorithm can detect moving target efficiently.
2008, 30(5): 1060-1063.
doi: 10.3724/SP.J.1146.2006.01681
Abstract:
Linear-prediction-type STAP methods greatly reduced the demand of Independent Identical Distribution(IID) samples, which made Linear-prediction-type STAP methods adapted well to non-homogeneous clutter environment especially. In this paper linear-prediction-type STAP methods is analyzed in detail, and the formula of how to determinate the order of Linear-prediction-type STAP methods is proposed. All conclusions are proved using simulation results.
Linear-prediction-type STAP methods greatly reduced the demand of Independent Identical Distribution(IID) samples, which made Linear-prediction-type STAP methods adapted well to non-homogeneous clutter environment especially. In this paper linear-prediction-type STAP methods is analyzed in detail, and the formula of how to determinate the order of Linear-prediction-type STAP methods is proposed. All conclusions are proved using simulation results.
2008, 30(5): 1064-1067.
doi: 10.3724/SP.J.1146.2006.01700
Abstract:
This paper investigats the imaging theory of the Rotate Scanning Synthetic Aperture Interferometric Radiometer (RS-SAIR) which has a non-Cartesian sampling scheme, and proposes a new imaging algorithm based on windowing back projection. The critical feature of this algorithm is that it takes back projection as the main imaging tool instead of 2-D Cartesian grid Fourier transform, and it involves only 1-D tapered FFT and back projection, which avoids the interpolations from polar sampling grids to Cartesian grids, and could perform imaging and sampling simultaneously, then it leads to a more accurate and fast imaging performance. Being a critical step the windowing effect is analyzed, and different window function is test to show the spatial resolution and noise control of the reconstructed image. Further more, the sampling is analyzed and a rational strategy is achieved. Finally, the imaging of a 2-D scene with a RS-SAIR system is simulated. The comparison of the parameters in the mean square errors and cross correlation coefficient with various methods indicates that the proposed new algorithm is superior to other Cartesian FFT based methods.
This paper investigats the imaging theory of the Rotate Scanning Synthetic Aperture Interferometric Radiometer (RS-SAIR) which has a non-Cartesian sampling scheme, and proposes a new imaging algorithm based on windowing back projection. The critical feature of this algorithm is that it takes back projection as the main imaging tool instead of 2-D Cartesian grid Fourier transform, and it involves only 1-D tapered FFT and back projection, which avoids the interpolations from polar sampling grids to Cartesian grids, and could perform imaging and sampling simultaneously, then it leads to a more accurate and fast imaging performance. Being a critical step the windowing effect is analyzed, and different window function is test to show the spatial resolution and noise control of the reconstructed image. Further more, the sampling is analyzed and a rational strategy is achieved. Finally, the imaging of a 2-D scene with a RS-SAIR system is simulated. The comparison of the parameters in the mean square errors and cross correlation coefficient with various methods indicates that the proposed new algorithm is superior to other Cartesian FFT based methods.
2008, 30(5): 1068-1071.
doi: 10.3724/SP.J.1146.2006.01683
Abstract:
The efficiency of broad band barrage Stand-Off Jamming (SOJ) is decreased quickly with such techniques as radar ultra-low sidelobe antennas and narrow-channel filter. In allusion to this situation, the weakness of SOJ and Electronic Counter-CounterMeasure (ECCM) of modern radar against blanket jamming are analyzed firstly. Then, a blanket Stand-Forward Jamming (SFJ) method with close-in mainlobe jamming tactics and coherent phase jamming technique is suggested to apply. This paper puts forward a principle block diagram of the jammer and designs jamming mode and signal in detail as well. Furthermore, it investigates the characteristics of power and time-frequency of this jamming technique. Computer simulation indicates that this jamming technique is superior to broad band barrage SOJ.
The efficiency of broad band barrage Stand-Off Jamming (SOJ) is decreased quickly with such techniques as radar ultra-low sidelobe antennas and narrow-channel filter. In allusion to this situation, the weakness of SOJ and Electronic Counter-CounterMeasure (ECCM) of modern radar against blanket jamming are analyzed firstly. Then, a blanket Stand-Forward Jamming (SFJ) method with close-in mainlobe jamming tactics and coherent phase jamming technique is suggested to apply. This paper puts forward a principle block diagram of the jammer and designs jamming mode and signal in detail as well. Furthermore, it investigates the characteristics of power and time-frequency of this jamming technique. Computer simulation indicates that this jamming technique is superior to broad band barrage SOJ.
A Detection Method for LFM Signal Embedded in NMA Interference Using Signal Phase Matching Principle
2008, 30(5): 1072-1074.
doi: 10.3724/SP.J.1146.2006.01657
Abstract:
On the basic of Signal Phase Matching Principle (SPMP), a detection method for Linear FM (LFM) signal embedded in Noise AM (NAM) interference is presented. The formula for estimation the carrier frequency of noise AM interference is derived, and the signal phase is matched using the carrier frequency of interference, then LFM signal is detected based on least squares of signal phase matching. The simulation shows that the proposed method is efficient.
On the basic of Signal Phase Matching Principle (SPMP), a detection method for Linear FM (LFM) signal embedded in Noise AM (NAM) interference is presented. The formula for estimation the carrier frequency of noise AM interference is derived, and the signal phase is matched using the carrier frequency of interference, then LFM signal is detected based on least squares of signal phase matching. The simulation shows that the proposed method is efficient.
2008, 30(5): 1075-1079.
doi: 10.3724/SP.J.1146.2006.01677
Abstract:
The low reliability of estimation to widelane ambiguity is mainly caused by pseudo-range multipath error. Therefor, a new method based on Kalman filter is proposed to separate multipath error from widelane ambiguity in real time. This method can get reliable ambiguity in short time when the multipath error is modeled as first order Gauss-Markov process. The validity and reliability of proposed method are confirmed with experiments.
The low reliability of estimation to widelane ambiguity is mainly caused by pseudo-range multipath error. Therefor, a new method based on Kalman filter is proposed to separate multipath error from widelane ambiguity in real time. This method can get reliable ambiguity in short time when the multipath error is modeled as first order Gauss-Markov process. The validity and reliability of proposed method are confirmed with experiments.
2008, 30(5): 1080-1083.
doi: 10.3724/SP.J.1146.2006.01718
Abstract:
Robust time delay estimation is proposed using variable step-size affine projection algorithm with M-estimators. The novel algorithm, using the structure of adaptive filter and M-estimators, is free of the assumption that the statistical characteristics of singal and noise is known. Simulation experiments under three different noise environments, Gaussian, non-Gaussian and impulse noise, are conducted to compare with the proposed TDE method using the existing Higher-Order Spectra(HOS) and Least Mean Square(LMS), and the result showes that the proposed TDE method is more robust and precision.
Robust time delay estimation is proposed using variable step-size affine projection algorithm with M-estimators. The novel algorithm, using the structure of adaptive filter and M-estimators, is free of the assumption that the statistical characteristics of singal and noise is known. Simulation experiments under three different noise environments, Gaussian, non-Gaussian and impulse noise, are conducted to compare with the proposed TDE method using the existing Higher-Order Spectra(HOS) and Least Mean Square(LMS), and the result showes that the proposed TDE method is more robust and precision.
2008, 30(5): 1084-1087.
doi: 10.3724/SP.J.1146.2006.01761
Abstract:
The effect of LFM interference on pseudo-random coded fuze system is analyzed. It is found that the correlation output is deteriorated when the interference power is larger than the anti-interference limit of the fuze system. It is very necessary to excise the above interference. Because there is strong time and frequency coupling between the interference and useful signal, so the method of adaptive time-varying interference cancellation based on FRFT is proposed. The characteristic parameters of interferences are obtained by FRFT. Then the adaptive interference canceler is designed according to the estimated interference parameters. Simulation results show that the above method can effectively excise LFM interference and the correlation output is improved obviously. This method is appropriate for both single component interference and multi-component interferences and it can not be affected by across items.
The effect of LFM interference on pseudo-random coded fuze system is analyzed. It is found that the correlation output is deteriorated when the interference power is larger than the anti-interference limit of the fuze system. It is very necessary to excise the above interference. Because there is strong time and frequency coupling between the interference and useful signal, so the method of adaptive time-varying interference cancellation based on FRFT is proposed. The characteristic parameters of interferences are obtained by FRFT. Then the adaptive interference canceler is designed according to the estimated interference parameters. Simulation results show that the above method can effectively excise LFM interference and the correlation output is improved obviously. This method is appropriate for both single component interference and multi-component interferences and it can not be affected by across items.
2008, 30(5): 1088-1091.
doi: 10.3724/SP.J.1146.2006.01704
Abstract:
This paper defines the mean SNRs resolving threshold which can measure the resolving performance of MUSIC algorithm, and deduces the expression of the resolving threshold. The data simulation show that the power difference of the two signal sources can reduce the resolving ability of MUSIC algorithm, that the relative effect of the power difference on the resolving performance of MUSIC algorithm has nothing with other parameters under certain condition. This paper also discusses the effect of the power difference on the number of array element, the number of snap shot and the absolute subtract of the two signal sources orientation parameter.
This paper defines the mean SNRs resolving threshold which can measure the resolving performance of MUSIC algorithm, and deduces the expression of the resolving threshold. The data simulation show that the power difference of the two signal sources can reduce the resolving ability of MUSIC algorithm, that the relative effect of the power difference on the resolving performance of MUSIC algorithm has nothing with other parameters under certain condition. This paper also discusses the effect of the power difference on the number of array element, the number of snap shot and the absolute subtract of the two signal sources orientation parameter.
2008, 30(5): 1092-1095.
doi: 10.3724/SP.J.1146.2006.01632
Abstract:
In the process of the beamforming for the coherent signals, because of the estimation error of the coherent signal energy and the estimation error of covariance matrix R, and perturbation error on the steering vector, the output SINR of beamformer drops badly. In this paper a new beamformer is presented based on the LS shaping from quantum signal processing. As a robust beamformer it shows the good performance in the circumstance discussed above. Analysis and simulations are given to prove the efficiency and validity of the new beamformer.
In the process of the beamforming for the coherent signals, because of the estimation error of the coherent signal energy and the estimation error of covariance matrix R, and perturbation error on the steering vector, the output SINR of beamformer drops badly. In this paper a new beamformer is presented based on the LS shaping from quantum signal processing. As a robust beamformer it shows the good performance in the circumstance discussed above. Analysis and simulations are given to prove the efficiency and validity of the new beamformer.
2008, 30(5): 1096-1099.
doi: 10.3724/SP.J.1146.2006.01576
Abstract:
The MODE-TOEP beamforming algorithm is proposed to suppress the coherent interference on a Uniform Circular Array (UCA). Firstly, the mode space transformation is used to transform the UCA into a Virtual ULA (VULA). On this base, a Toeplitz matrix can be reconstructed. Then EigenSpace-Based (ESB) beamforming is applied to suppress the coherent interference successfully. MODE-TOEP beamforming algorithm has less computational burden due to non-spatial smoothing but has a better performance of the interference suppression compared with conventional MODE-SS beamforming algorithm. Simulation results show the validity and merit of the MODE-TOEP beamforming algorithm.
The MODE-TOEP beamforming algorithm is proposed to suppress the coherent interference on a Uniform Circular Array (UCA). Firstly, the mode space transformation is used to transform the UCA into a Virtual ULA (VULA). On this base, a Toeplitz matrix can be reconstructed. Then EigenSpace-Based (ESB) beamforming is applied to suppress the coherent interference successfully. MODE-TOEP beamforming algorithm has less computational burden due to non-spatial smoothing but has a better performance of the interference suppression compared with conventional MODE-SS beamforming algorithm. Simulation results show the validity and merit of the MODE-TOEP beamforming algorithm.
2008, 30(5): 1100-1103.
doi: 10.3724/SP.J.1146.2007.00481
Abstract:
A modified forward-backward averaging algorithm to estimate the covariance matrix in beamspace DBF is proposed. The computation is greatly reduced because the complex number operation is replaced with real number operation with the special structure of symmetrical array. Simulation results show that the proposed algorithm can preserve its interference suppression ability with the reduced computation.
A modified forward-backward averaging algorithm to estimate the covariance matrix in beamspace DBF is proposed. The computation is greatly reduced because the complex number operation is replaced with real number operation with the special structure of symmetrical array. Simulation results show that the proposed algorithm can preserve its interference suppression ability with the reduced computation.
2008, 30(5): 1104-1108.
doi: 10.3724/SP.J.1146.2007.01491
Abstract:
In order to avoid the default of the greedy algorithm to Approximate given function by searching a linear combination of basis functions choosing from a redundant basis function dictionary for the Kernel Matching Pursuits (KMP), we make use of the global optimal searching ability and the locally quickly searching ability of Immune Clonal Selection Algorithm (ICSA) to speed up searching basic function data in function dictionary. And a method for object recognition of Kernel matching pursuits based on immune clonal selection algorithm is presented. This method reduces greatly computer time of the KMP algorithm. The simulation result of the UCI datasets, remote images and Brodatz images show the proposed algorithm can decrease obviously training time leave the classification accuracy almost unchanged, especially for the large size datasets as compared with the standard KMP. The method has higher classification speed and more accurate recognition rate over the matching pursuits based on Genetic Algorithm (GA).
In order to avoid the default of the greedy algorithm to Approximate given function by searching a linear combination of basis functions choosing from a redundant basis function dictionary for the Kernel Matching Pursuits (KMP), we make use of the global optimal searching ability and the locally quickly searching ability of Immune Clonal Selection Algorithm (ICSA) to speed up searching basic function data in function dictionary. And a method for object recognition of Kernel matching pursuits based on immune clonal selection algorithm is presented. This method reduces greatly computer time of the KMP algorithm. The simulation result of the UCI datasets, remote images and Brodatz images show the proposed algorithm can decrease obviously training time leave the classification accuracy almost unchanged, especially for the large size datasets as compared with the standard KMP. The method has higher classification speed and more accurate recognition rate over the matching pursuits based on Genetic Algorithm (GA).
2008, 30(5): 1109-1112.
doi: 10.3724/SP.J.1146.2006.01760
Abstract:
In order to accelerate generation of confusion network with high quality, a fast algorithm with linear time complexity is proposed in this paper. The proposed algorithm is guided with maximum posteriori arc and only traverses the lattice one pass. Kullback-Leibler Divergence (KLD) is used to measure the similarity between two arcs labels, which can improve the accuracy of arc alignment in the process of generating confusion network. The experimental results show that the proposed algorithm is comparable with Xues fast algorithm at generation speed while the quality of confusion network is significantly improved. Further improvement of the quality can be obtained by using KLD as similarity measure of arcs labels.
In order to accelerate generation of confusion network with high quality, a fast algorithm with linear time complexity is proposed in this paper. The proposed algorithm is guided with maximum posteriori arc and only traverses the lattice one pass. Kullback-Leibler Divergence (KLD) is used to measure the similarity between two arcs labels, which can improve the accuracy of arc alignment in the process of generating confusion network. The experimental results show that the proposed algorithm is comparable with Xues fast algorithm at generation speed while the quality of confusion network is significantly improved. Further improvement of the quality can be obtained by using KLD as similarity measure of arcs labels.
2008, 30(5): 1113-1117.
doi: 10.3724/SP.J.1146.2006.01588
Abstract:
A new method for color image retrieval is introduced in this paper. At first, the image is parted into some connected regions. Rational Legendre chromaticity distribution moments and texture co-occurrence matrices are computed to represent the color and texture features (Regional Color Texture: RCT). Then, Dynamic Partial distance Function (DPF) and dynamic regions matching are used to weight RCT features, for short RCT_DPF. Experimental results indicate that this method RCT-DPF has good performance in image retrieval. To compare with Texture Co-occurrence Matrices(TCM), Multi-component Co-occurrence Matrices(MCM) and RCT, RCT-DPF is better precision.
A new method for color image retrieval is introduced in this paper. At first, the image is parted into some connected regions. Rational Legendre chromaticity distribution moments and texture co-occurrence matrices are computed to represent the color and texture features (Regional Color Texture: RCT). Then, Dynamic Partial distance Function (DPF) and dynamic regions matching are used to weight RCT features, for short RCT_DPF. Experimental results indicate that this method RCT-DPF has good performance in image retrieval. To compare with Texture Co-occurrence Matrices(TCM), Multi-component Co-occurrence Matrices(MCM) and RCT, RCT-DPF is better precision.
2008, 30(5): 1118-1121.
doi: 10.3724/SP.J.1146.2006.01557
Abstract:
Compared with Self-Organizing Feature Map(SOFM), maximum-entropy clustering and K-means clustering, the Neural-Gas network algorithm has advantages of faster convergence, smaller cost distortion errors, etc. However, the fixed and regular neurons on the output space represent worse distance information when the neural gas network algorithm is used for dimension reduction and visualization of linear or nonlinear data sets with nonuniform distribution. Therefore, according to the basic idea of the probabilistic regularized SOFM, a new visualization method for hybridizing neural gas network and Sammons mapping is proposed to overcome this problem, and it reduces the computational complexity with using neural gas network algorithm for feature clustering and preserves the interneuronal distances resemblance from input space into output space by using Sammons mapping. Simulation results show that the proposed hybridizing algorithm can obtain the better visualization effect on the synthetic and real data sets, thus demonstrating the feasibility and effectiveness of the hybridizing algorithm.
Compared with Self-Organizing Feature Map(SOFM), maximum-entropy clustering and K-means clustering, the Neural-Gas network algorithm has advantages of faster convergence, smaller cost distortion errors, etc. However, the fixed and regular neurons on the output space represent worse distance information when the neural gas network algorithm is used for dimension reduction and visualization of linear or nonlinear data sets with nonuniform distribution. Therefore, according to the basic idea of the probabilistic regularized SOFM, a new visualization method for hybridizing neural gas network and Sammons mapping is proposed to overcome this problem, and it reduces the computational complexity with using neural gas network algorithm for feature clustering and preserves the interneuronal distances resemblance from input space into output space by using Sammons mapping. Simulation results show that the proposed hybridizing algorithm can obtain the better visualization effect on the synthetic and real data sets, thus demonstrating the feasibility and effectiveness of the hybridizing algorithm.
2008, 30(5): 1122-1126.
doi: 10.3724/SP.J.1146.2006.01666
Abstract:
This paper presents a method of capturing multi-curve spectrums of eye movements and expressions. Based on 3D muscle controlled eye(3D-MCE)model, eye movements and expressions are described with a group of time-varying curves of normalized muscle contractions, called multi-curve spectrum. Multi-curve spectrum is obtained from single video with tracking specific control nodes and computing time-varying curves of muscle contractions. Realistic simulation results comprising eye movements and some expressions are genarated with multi-curve spectrums captured with this method.
This paper presents a method of capturing multi-curve spectrums of eye movements and expressions. Based on 3D muscle controlled eye(3D-MCE)model, eye movements and expressions are described with a group of time-varying curves of normalized muscle contractions, called multi-curve spectrum. Multi-curve spectrum is obtained from single video with tracking specific control nodes and computing time-varying curves of muscle contractions. Realistic simulation results comprising eye movements and some expressions are genarated with multi-curve spectrums captured with this method.
2008, 30(5): 1127-1130.
doi: 10.3724/SP.J.1146.2006.01689
Abstract:
This paper presents a method on automatic question classification through incremental rule learning based on rough set theory. The core of the method is appling the machine learning approach to gain classified rules automatically through extract the features of query sentence thoroughly, and the decision table is used to construct the training collection. Comparing with the alternative means, the superiority is that it acquires the classified rule automatically and uses the rough set method to obtain the optimized smallest rule set. Especially, the incremental learning is induced to improve the precision and avoid the tedious re-training process. The performance of the approach is promising, when tested on opposite test. Meanwhile, the method obtains a very good result in the international TREC2005 Q/A track.
This paper presents a method on automatic question classification through incremental rule learning based on rough set theory. The core of the method is appling the machine learning approach to gain classified rules automatically through extract the features of query sentence thoroughly, and the decision table is used to construct the training collection. Comparing with the alternative means, the superiority is that it acquires the classified rule automatically and uses the rough set method to obtain the optimized smallest rule set. Especially, the incremental learning is induced to improve the precision and avoid the tedious re-training process. The performance of the approach is promising, when tested on opposite test. Meanwhile, the method obtains a very good result in the international TREC2005 Q/A track.
2008, 30(5): 1131-1135.
doi: 10.3724/SP.J.1146.2006.01640
Abstract:
The encoding issue is a most fundamental one in DNA based computing. In this paper, the various factors that influence the encoding problem and the general idea of the template encoding method are first introduced. Then the reason of the shift hybridisation occurred in DNA computing is presented. And a new method is proposed to search template set with high shift distance. Additionally, to increase the search space can also increase the number of template string.
The encoding issue is a most fundamental one in DNA based computing. In this paper, the various factors that influence the encoding problem and the general idea of the template encoding method are first introduced. Then the reason of the shift hybridisation occurred in DNA computing is presented. And a new method is proposed to search template set with high shift distance. Additionally, to increase the search space can also increase the number of template string.
2008, 30(5): 1136-1139.
doi: 10.3724/SP.J.1146.2006.01617
Abstract:
A new algorithm for NLOS(Non-Line-Of-Sight) passive acoustic localization of low altitude targets is presented. Based on target reflection point, the target point can be localized. This approach can reduce the NLOS error caused by acoustic propagation effectively, to increase localizing precision and provide important information for the radar to track and localize. Simulation results show that the algorithm have good effectiveness to estimate the low altitude targets position especially which hiding behind a mountain when time-delay estimation is high enough.
A new algorithm for NLOS(Non-Line-Of-Sight) passive acoustic localization of low altitude targets is presented. Based on target reflection point, the target point can be localized. This approach can reduce the NLOS error caused by acoustic propagation effectively, to increase localizing precision and provide important information for the radar to track and localize. Simulation results show that the algorithm have good effectiveness to estimate the low altitude targets position especially which hiding behind a mountain when time-delay estimation is high enough.
2008, 30(5): 1140-1143.
doi: 10.3724/SP.J.1146.2007.00739
Abstract:
An accelerated acoustic echo cancellation algorithm is proposed in this paper based on Parallel Subgradient Projection (PSP) technique. Through the angle optimization strategy, accelerated method optimizes the input signal spaces belong to different states and finds a pair of optimal projection spaces. Then adaptive algorithm can be accelerated by projection onto the pair of optimal half spaces. Experiment indicates the accelerated method possesses more rapid convergence speed than traditional algorithm. Additional, system performance gets promoted no matter evaluated by mean square error or echo path misalignment.
An accelerated acoustic echo cancellation algorithm is proposed in this paper based on Parallel Subgradient Projection (PSP) technique. Through the angle optimization strategy, accelerated method optimizes the input signal spaces belong to different states and finds a pair of optimal projection spaces. Then adaptive algorithm can be accelerated by projection onto the pair of optimal half spaces. Experiment indicates the accelerated method possesses more rapid convergence speed than traditional algorithm. Additional, system performance gets promoted no matter evaluated by mean square error or echo path misalignment.
2008, 30(5): 1144-1147.
doi: 10.3724/SP.J.1146.2007.00210
Abstract:
In this paper, the fundamental theory of voice transformation and Support Vector Regression(SVR) are introduced. The algorithm of voice transformation technology based on multi-output SVR is put forward, and the corresponding simulation experiment is carried out. The subjective test and objective test show that compared with code book mapping and GMM, this mapping rule performs well between discreteness and smoothness of parameter.
In this paper, the fundamental theory of voice transformation and Support Vector Regression(SVR) are introduced. The algorithm of voice transformation technology based on multi-output SVR is put forward, and the corresponding simulation experiment is carried out. The subjective test and objective test show that compared with code book mapping and GMM, this mapping rule performs well between discreteness and smoothness of parameter.
2008, 30(5): 1148-1150.
doi: 10.3724/SP.J.1146.2006.01687
Abstract:
This paper considers recognition and estimation of carrier frequency from a time sampled MPSK signal in Automatic Modulation Recognition (AMR), when M and training symbol sequence are unknown. In this case, a new Non-Data-Aided (NDA) carrier frequency estimation algorithm is presented, after recognition of M. The introduction of linear phase unwrapping and high order correlation makes the new algorithm perform better than same traditional NDA carrier frequency estimation algorithms, when frequency offset is large or when the SNR is low. Simulation result shows this fact and it performs well in AMR system.
This paper considers recognition and estimation of carrier frequency from a time sampled MPSK signal in Automatic Modulation Recognition (AMR), when M and training symbol sequence are unknown. In this case, a new Non-Data-Aided (NDA) carrier frequency estimation algorithm is presented, after recognition of M. The introduction of linear phase unwrapping and high order correlation makes the new algorithm perform better than same traditional NDA carrier frequency estimation algorithms, when frequency offset is large or when the SNR is low. Simulation result shows this fact and it performs well in AMR system.
2008, 30(5): 1151-1154.
doi: 10.3724/SP.J.1146.2007.01231
Abstract:
The iterative decoding schemes such as BICM-ID and other concatenated coding systems require the knowledge of channel Signal-to-Noise Ratio (SNR). The effect of SNR estimation errors is studied for BICM-ID in the paper and the performance sensitivity of SNR mismatch for the log-APP decoding algorithm and its simplified implementations is analyzed. Analysis and simulation results show that the max-log-APP algorithm is robust against any SNR mismatch problem, while the log-APP algorithm and other simplified implementations are more tolerant of overestimation of SNR than underestimation, which is because they use different correction functions. The required SNR accuracy to prevent large performance degradation is increased with the increase of channel SNR, generally from -2dB to +6dB. Therefore, when there is no perfect channel state information, the max-log-APP algorithm is the optimum decoding algorithm for BICM-ID.
The iterative decoding schemes such as BICM-ID and other concatenated coding systems require the knowledge of channel Signal-to-Noise Ratio (SNR). The effect of SNR estimation errors is studied for BICM-ID in the paper and the performance sensitivity of SNR mismatch for the log-APP decoding algorithm and its simplified implementations is analyzed. Analysis and simulation results show that the max-log-APP algorithm is robust against any SNR mismatch problem, while the log-APP algorithm and other simplified implementations are more tolerant of overestimation of SNR than underestimation, which is because they use different correction functions. The required SNR accuracy to prevent large performance degradation is increased with the increase of channel SNR, generally from -2dB to +6dB. Therefore, when there is no perfect channel state information, the max-log-APP algorithm is the optimum decoding algorithm for BICM-ID.
2008, 30(5): 1155-1158.
doi: 10.3724/SP.J.1146.2006.01754
Abstract:
This paper realizes m sequence and FIR algorithm to generate band-limited white noise on a FPGA chip. This noise generator features excellent in-band amplitude flatness and broad bandwidth by means of a novel anti-sinc square function FIR digital filter. The detailed theoretic derivation of this new filter is given and the comparison of its theoretic results with experimental results is also made. The paper then analyzes the relations between maxim bandwidth, in-band amplutude flatness and out-band spurs position of this kind of generator, and gives the corresponding experimental results. The accordance between theoretic and experimental results indicates the validity of the presented method.
This paper realizes m sequence and FIR algorithm to generate band-limited white noise on a FPGA chip. This noise generator features excellent in-band amplitude flatness and broad bandwidth by means of a novel anti-sinc square function FIR digital filter. The detailed theoretic derivation of this new filter is given and the comparison of its theoretic results with experimental results is also made. The paper then analyzes the relations between maxim bandwidth, in-band amplutude flatness and out-band spurs position of this kind of generator, and gives the corresponding experimental results. The accordance between theoretic and experimental results indicates the validity of the presented method.
2008, 30(5): 1159-1162.
doi: 10.3724/SP.J.1146.2006.01616
Abstract:
It is a very important domain to analyze the inter-cell interference in multimedia CDMA networks. However, traditional methods have three main limitations and can hardly accurately reflect the real status of cellular system. A new method is proposed to overcome these limitations. The interference is quantitatively analyzed and the close-form expressions of interference expectation and variance are acquired. Through numeric computation, the interference is analyzed in different conditions. Results show that with the distance between two base stations increased, the decline velocity of interference is decreased.
It is a very important domain to analyze the inter-cell interference in multimedia CDMA networks. However, traditional methods have three main limitations and can hardly accurately reflect the real status of cellular system. A new method is proposed to overcome these limitations. The interference is quantitatively analyzed and the close-form expressions of interference expectation and variance are acquired. Through numeric computation, the interference is analyzed in different conditions. Results show that with the distance between two base stations increased, the decline velocity of interference is decreased.
2008, 30(5): 1163-1166.
doi: 10.3724/SP.J.1146.2006.01625
Abstract:
Due to the high decoding complexity of the conventional full rate Quasi-Orthogonal Space-Time Block Code (QOSTBC) over eight transmit antennas, two precoding-based transmit schemes are proposed. The new QOSTBC is extended from the OSTBC over four antennas. With the precoding matrix constructed with the two phase feedbacks, the orthogonalization of channel matrix cancels the Inter-Symbol Interference (ISI). The Maximum Likelihood (ML) decoding of independent symbol is realized. An extra performance improvement is achieved with the interleaving at the transmitter. Meanwhile, the ML decoding of two joint symbols is carried out. Unlike the constellation rotation scheme for the full diversity, the precoding scheme improves the performance of QOSTBC with reducing ISI to the most extent. Simulation results show that two precdoing-based QOSTBCs outperform the conventional QOSTBC with constellation rotation and achieve the lower decoding complexity.
Due to the high decoding complexity of the conventional full rate Quasi-Orthogonal Space-Time Block Code (QOSTBC) over eight transmit antennas, two precoding-based transmit schemes are proposed. The new QOSTBC is extended from the OSTBC over four antennas. With the precoding matrix constructed with the two phase feedbacks, the orthogonalization of channel matrix cancels the Inter-Symbol Interference (ISI). The Maximum Likelihood (ML) decoding of independent symbol is realized. An extra performance improvement is achieved with the interleaving at the transmitter. Meanwhile, the ML decoding of two joint symbols is carried out. Unlike the constellation rotation scheme for the full diversity, the precoding scheme improves the performance of QOSTBC with reducing ISI to the most extent. Simulation results show that two precdoing-based QOSTBCs outperform the conventional QOSTBC with constellation rotation and achieve the lower decoding complexity.
2008, 30(5): 1167-1170.
doi: 10.3724/SP.J.1146.2006.01763
Abstract:
A decoding method for LDPC codes concatenated by multiple Cyclic Redundancy Checks (CRCs) has improved the convergence of the decoding effectively. However, the error floor of the decoding appears due to the high overall Undetected Error Probability (UEP) of CRCs. Therefore, This work improves the Codeword Error Rate (CER) performance further based on the fact that the overall UEP of CRCs in decoding can be reduced when the application times of CRCs checking are decreased. The simulation results show the performance improvements in terms of CER, while decoding complexity is slightly increased.
A decoding method for LDPC codes concatenated by multiple Cyclic Redundancy Checks (CRCs) has improved the convergence of the decoding effectively. However, the error floor of the decoding appears due to the high overall Undetected Error Probability (UEP) of CRCs. Therefore, This work improves the Codeword Error Rate (CER) performance further based on the fact that the overall UEP of CRCs in decoding can be reduced when the application times of CRCs checking are decreased. The simulation results show the performance improvements in terms of CER, while decoding complexity is slightly increased.
2008, 30(5): 1171-1175.
doi: 10.3724/SP.J.1146.2006.01533
Abstract:
In this paper, a new type of Space-Time Block Code (STBC) that is similar to ABBA code is expanded and utilized in CD(Cooperative Diversity) networks. The decomposable character and destructive influence of non-orthogonality of this code is evaluated. The constraint of the number of cooperative nodes and the trade off of which and receive performance is investigated with simulation. The result shows that, for the cooperative communication system this paper proposed, it is enough for 8 relay nodes to participate in cooperation and more relay nodes have little effect on the system performance improvement.
In this paper, a new type of Space-Time Block Code (STBC) that is similar to ABBA code is expanded and utilized in CD(Cooperative Diversity) networks. The decomposable character and destructive influence of non-orthogonality of this code is evaluated. The constraint of the number of cooperative nodes and the trade off of which and receive performance is investigated with simulation. The result shows that, for the cooperative communication system this paper proposed, it is enough for 8 relay nodes to participate in cooperation and more relay nodes have little effect on the system performance improvement.
2008, 30(5): 1176-1179.
doi: 10.3724/SP.J.1146.2007.00541
Abstract:
In the paper, a novel overlapping multiplexing technology named OvHDM (Overlapped Hybrid Division Multiplexing) is proposed. In OvHDM, the overlapping multiplexing is developed in both time and frequency domain. The optimum detection, spectrum efficiency, detection complexity, performance and Peak-to-Average Power Ratio (PAPR) of OvHDM are studied in detail. The signal model of OvHDM is the convolutional constraint model between vectors. In order to implement optimum detection, a novel algorithm named MU-MLSD (Multi-User Maximum-Likelihood Sequence Detection) is proposed. The simulation results prove that the performance of OvHDM is better 4.5dB than that of 64QAM and better 1.5dB than that of OvTDM (Overlapped Time Division Multiplexing) when the spectrum efficiency is 6bit/s/Hz. Meanwhile, the PAPR of OvHDM is better 2dB than that of 128-subcarrier OFDM and better 2.6dB than that of 1024-subcarrier OFDM.
In the paper, a novel overlapping multiplexing technology named OvHDM (Overlapped Hybrid Division Multiplexing) is proposed. In OvHDM, the overlapping multiplexing is developed in both time and frequency domain. The optimum detection, spectrum efficiency, detection complexity, performance and Peak-to-Average Power Ratio (PAPR) of OvHDM are studied in detail. The signal model of OvHDM is the convolutional constraint model between vectors. In order to implement optimum detection, a novel algorithm named MU-MLSD (Multi-User Maximum-Likelihood Sequence Detection) is proposed. The simulation results prove that the performance of OvHDM is better 4.5dB than that of 64QAM and better 1.5dB than that of OvTDM (Overlapped Time Division Multiplexing) when the spectrum efficiency is 6bit/s/Hz. Meanwhile, the PAPR of OvHDM is better 2dB than that of 128-subcarrier OFDM and better 2.6dB than that of 1024-subcarrier OFDM.
2008, 30(5): 1180-1184.
doi: 10.3724/SP.J.1146.2006.01377
Abstract:
A new Joint Detection (JD) algorithm, named as Block-Decision-Feedback based Joint Detection (BDF-JD), is proposed for TD-SCDMA. Comparing to the existing algorithms, BDF-JD has much less computational complexity and comparable performance, and is very suitable for terminals. Based on BDF-JD, the architecture of the JD coprocessor is proposed. Implemented with the 0.18 m process of SMIC, the coprocessor only consumes 160 k gates.
A new Joint Detection (JD) algorithm, named as Block-Decision-Feedback based Joint Detection (BDF-JD), is proposed for TD-SCDMA. Comparing to the existing algorithms, BDF-JD has much less computational complexity and comparable performance, and is very suitable for terminals. Based on BDF-JD, the architecture of the JD coprocessor is proposed. Implemented with the 0.18 m process of SMIC, the coprocessor only consumes 160 k gates.
2008, 30(5): 1185-1188.
doi: 10.3724/SP.J.1146.2006.01621
Abstract:
Transmission optimization for Multiple-Input Multiple-Output Orthogonal Frequency Division Multiplexing (MIMO-OFDM) broadcast channel with power-limited constraints is analyzed in this paper. Applying different optimization rules, two different transmission strategies are proposed. One is Optimal Space Distance Dirty Paper Coding(OSD-DPC), the other is Max Signal to Noise Ratio Dirty Paper Coding (MSNR-DPC). Numerical results show that proposed schemes have the similar performance to the optimal schemes and are capable of improving the system capacity and performance with power-limited constraints.
Transmission optimization for Multiple-Input Multiple-Output Orthogonal Frequency Division Multiplexing (MIMO-OFDM) broadcast channel with power-limited constraints is analyzed in this paper. Applying different optimization rules, two different transmission strategies are proposed. One is Optimal Space Distance Dirty Paper Coding(OSD-DPC), the other is Max Signal to Noise Ratio Dirty Paper Coding (MSNR-DPC). Numerical results show that proposed schemes have the similar performance to the optimal schemes and are capable of improving the system capacity and performance with power-limited constraints.
2008, 30(5): 1189-1192.
doi: 10.3724/SP.J.1146.2006.01407
Abstract:
A fast generalized complex sphere decoder is proposed for under-determined MIMO systems with fewer receive antennas than transmit antennas in this paper. It can handle both the lattice-structured and non-lattice-structured constellations in two-dimensional space. This algorithm adopts optimal search scheme on the under-determined part of the transmitted signal, thus avoids the high complexity of the exhaustive search scheme. The simulation results show that the complexity of the generalized complex sphere algorithm is significantly lower than that of the exhaustive search scheme.
A fast generalized complex sphere decoder is proposed for under-determined MIMO systems with fewer receive antennas than transmit antennas in this paper. It can handle both the lattice-structured and non-lattice-structured constellations in two-dimensional space. This algorithm adopts optimal search scheme on the under-determined part of the transmitted signal, thus avoids the high complexity of the exhaustive search scheme. The simulation results show that the complexity of the generalized complex sphere algorithm is significantly lower than that of the exhaustive search scheme.
2008, 30(5): 1193-1197.
doi: 10.3724/SP.J.1146.2006.01366
Abstract:
This paper begins with the investigation on exhaustive search algorithm, the Gerschgorin circle algorithm, on the basis of which, a Gerschgorin circle with low complexity is presented. In the proposed algorithm, the column corresponding to the minimum eigen-value is eliminated to increase the lower bound of the eigen-values and to reduce the correlation between the antennas during each iteration, correspondingly, capacity is maximized or BER is minimized. The results of the research reveal that both the capacity and the BER performance of the proposed algorithm are better than that of the norm-based algorithm, and are close to the performance of the optimal exhaustive search algorithm with low complexity in the cases that more antennas are selected.
This paper begins with the investigation on exhaustive search algorithm, the Gerschgorin circle algorithm, on the basis of which, a Gerschgorin circle with low complexity is presented. In the proposed algorithm, the column corresponding to the minimum eigen-value is eliminated to increase the lower bound of the eigen-values and to reduce the correlation between the antennas during each iteration, correspondingly, capacity is maximized or BER is minimized. The results of the research reveal that both the capacity and the BER performance of the proposed algorithm are better than that of the norm-based algorithm, and are close to the performance of the optimal exhaustive search algorithm with low complexity in the cases that more antennas are selected.
2008, 30(5): 1198-1202.
doi: 10.3724/SP.J.1146.2006.01593
Abstract:
The image distortion from JPEG2000 is the loss of structure information, which makes perceived quality dropping in particular at the lower bit rate. So, a rate control algorithm for JPEG2000 is proposed based on the prediction of perceptual structural information distortion without the reference image. The model of measuring the perceived image quality dropping is proposed at first. Then a prediction model of perceived quality dropping for JPEG2000 is constructed. Based on the model efficient rate control is realized. Numerical experiments verify the validity of the proposed method.
The image distortion from JPEG2000 is the loss of structure information, which makes perceived quality dropping in particular at the lower bit rate. So, a rate control algorithm for JPEG2000 is proposed based on the prediction of perceptual structural information distortion without the reference image. The model of measuring the perceived image quality dropping is proposed at first. Then a prediction model of perceived quality dropping for JPEG2000 is constructed. Based on the model efficient rate control is realized. Numerical experiments verify the validity of the proposed method.
2008, 30(5): 1203-1205.
doi: 10.3724/SP.J.1146.2007.00515
Abstract:
On automatic modulation there are two approaches, decision-theoretic and statistical pattem. An automatic modulation recognition system to recognize four digital signal classes as MASK, MFSK, MPSK, MQAM is proposed in this paper, which using decision-theoretic based feature set addition to statistical pattem based feature set with momentum auto-adapted weight BP neural network. Performance is generally good when Signal to Noise Ratios (SNR) in 0-10dB, and the estimated carrier frequency differs from the actual carrier frequency of 0-100Hz, simulations show the results even larger than 97%, that confirm the robustness and practicability of this recognition method.
On automatic modulation there are two approaches, decision-theoretic and statistical pattem. An automatic modulation recognition system to recognize four digital signal classes as MASK, MFSK, MPSK, MQAM is proposed in this paper, which using decision-theoretic based feature set addition to statistical pattem based feature set with momentum auto-adapted weight BP neural network. Performance is generally good when Signal to Noise Ratios (SNR) in 0-10dB, and the estimated carrier frequency differs from the actual carrier frequency of 0-100Hz, simulations show the results even larger than 97%, that confirm the robustness and practicability of this recognition method.
2008, 30(5): 1206-1209.
doi: 10.3724/SP.J.1146.2006.01563
Abstract:
Based on some basic handover rules, a new routing handover strategy is proposed and the performance is analyzed. The handover strategy is under considerations of handover, delay and load. Its performance is compared with other handover strategies The simulation results show that the new strategy performs better with less delay, lower handover frequency and can equipoise the network flux.
Based on some basic handover rules, a new routing handover strategy is proposed and the performance is analyzed. The handover strategy is under considerations of handover, delay and load. Its performance is compared with other handover strategies The simulation results show that the new strategy performs better with less delay, lower handover frequency and can equipoise the network flux.
2008, 30(5): 1210-1213.
doi: 10.3724/SP.J.1146.2006.01759
Abstract:
An iterative method based on orthogonal arrays is presented in this paper. And this method is used as the primary means to construct Cartesian authentication codes with k sources and n authentication symbols, where k and n are two arbitrary positive integers. When each source occurs equally often and encoding rule distribution is uniform, the probabilities of successful impersonation and substitution attack are 1/n. Compared with the Cartesian authentication codes constructed by Descartes product, the authentication codes produced with iterative means have much less encoding rules, when they contain the same parameters k and n.
An iterative method based on orthogonal arrays is presented in this paper. And this method is used as the primary means to construct Cartesian authentication codes with k sources and n authentication symbols, where k and n are two arbitrary positive integers. When each source occurs equally often and encoding rule distribution is uniform, the probabilities of successful impersonation and substitution attack are 1/n. Compared with the Cartesian authentication codes constructed by Descartes product, the authentication codes produced with iterative means have much less encoding rules, when they contain the same parameters k and n.
2008, 30(5): 1214-1217.
doi: 10.3724/SP.J.1146.2006.01627
Abstract:
On the basis of BB short signature scheme, this paper derives a new signature scheme, from which a new kind of group signature scheme is constructed. The security of the new group signature scheme is based on the q- SDH assumption and the decisional Diffie-Hellman assumption in the random oracle model. The length of the new group signature is a little longer than that of BBS short group signature. However, in the group signature scheme, giving certificates and private keys to group members do not need any third trusted party, while in BBS short group signature scheme it does need.
On the basis of BB short signature scheme, this paper derives a new signature scheme, from which a new kind of group signature scheme is constructed. The security of the new group signature scheme is based on the q- SDH assumption and the decisional Diffie-Hellman assumption in the random oracle model. The length of the new group signature is a little longer than that of BBS short group signature. However, in the group signature scheme, giving certificates and private keys to group members do not need any third trusted party, while in BBS short group signature scheme it does need.
2008, 30(5): 1218-1221.
doi: 10.3724/SP.J.1146.2007.00689
Abstract:
A fuzzy particle swarm optimization is proposed for solving complex constrained optimization problems. Firstly, a new perturbation operator is designed, and the concepts of fuzzy personal best value and fuzzy global best value are given based on the new operator. Particle updating equations are revised based upon the two new concepts to discourage the premature convergence. Secondly, a new comparison strategy is proposed based on the new concept of infeasible threshold value. It can preserve some infeasible solutions with high quality. Finally, the convergence of this algorithm is proved. The simulation results show that the proposed algorithm is effective, especially for the problems with high dimensions.
A fuzzy particle swarm optimization is proposed for solving complex constrained optimization problems. Firstly, a new perturbation operator is designed, and the concepts of fuzzy personal best value and fuzzy global best value are given based on the new operator. Particle updating equations are revised based upon the two new concepts to discourage the premature convergence. Secondly, a new comparison strategy is proposed based on the new concept of infeasible threshold value. It can preserve some infeasible solutions with high quality. Finally, the convergence of this algorithm is proved. The simulation results show that the proposed algorithm is effective, especially for the problems with high dimensions.
2008, 30(5): 1222-1225.
doi: 10.3724/SP.J.1146.2007.00735
Abstract:
Localization is one of the key technologies in wireless sensor network. Only the average one-hop distance estimated by the nearest anchor was used in DV-Hop localization algorithm, but the value estimated by a single anchor can not reflect the whole networks actual average one-hop distance. A novel algorithm is proposed to estimate the average one-hop distance based on weighted disposal. The average one-hop distance estimated by several anchors is used in this algorithm and assigned different weights according to the hops to the unknown node. The unknown nodes average one-hop distance is estimated more accurate and the localization precision is improved. Simulation results demonstrate that the average one-hop distance is estimated more exactly in the proposed algorithm than that of DV-Hop, the root mean square error is decreased and the location precision is improved.
Localization is one of the key technologies in wireless sensor network. Only the average one-hop distance estimated by the nearest anchor was used in DV-Hop localization algorithm, but the value estimated by a single anchor can not reflect the whole networks actual average one-hop distance. A novel algorithm is proposed to estimate the average one-hop distance based on weighted disposal. The average one-hop distance estimated by several anchors is used in this algorithm and assigned different weights according to the hops to the unknown node. The unknown nodes average one-hop distance is estimated more accurate and the localization precision is improved. Simulation results demonstrate that the average one-hop distance is estimated more exactly in the proposed algorithm than that of DV-Hop, the root mean square error is decreased and the location precision is improved.
2008, 30(5): 1226-1229.
doi: 10.3724/SP.J.1146.2007.01125
Abstract:
In the wireless sensor networks, unbalanced energy consumption has a strong impact on the network lifetime. The novel energy balance algorithms based on hybrid transmission probabilities are developed to prolong the network lifetime. The circle network model is divided into some ring sections (or slices), and the sensor nodes in each slice transmit data by hybrid transmission. The hybrid transmission probabilities finding algorithm is then presented to balance energy consumption of each sensor node. A set of probabilities for each slice can be obtained to decide the transmission mode of each sensor node and to better even out the energy consumption. This research further analyzes and optimizes the slice width. Simulations show that the algorithms can efficiently decrease energy consumption and prolong the network lifetime.
In the wireless sensor networks, unbalanced energy consumption has a strong impact on the network lifetime. The novel energy balance algorithms based on hybrid transmission probabilities are developed to prolong the network lifetime. The circle network model is divided into some ring sections (or slices), and the sensor nodes in each slice transmit data by hybrid transmission. The hybrid transmission probabilities finding algorithm is then presented to balance energy consumption of each sensor node. A set of probabilities for each slice can be obtained to decide the transmission mode of each sensor node and to better even out the energy consumption. This research further analyzes and optimizes the slice width. Simulations show that the algorithms can efficiently decrease energy consumption and prolong the network lifetime.
2008, 30(5): 1230-1233.
doi: 10.3724/SP.J.1146.2006.01755
Abstract:
In order to efficiently solve the energy problem of routing, the game-theory is borrowed and an energy efficient routing algorithm is proposed. Based on the research of game model and other routing schemes for wireless sensor networks, the Distributed Energy-Economical Routing (DEER) is designed to save the energy of the whole network through paying attention to both remained energy and the distribution of head nodes. The simulation results prove that this scheme can effectively balance the load and prolong the life of the wireless sensor networks.
In order to efficiently solve the energy problem of routing, the game-theory is borrowed and an energy efficient routing algorithm is proposed. Based on the research of game model and other routing schemes for wireless sensor networks, the Distributed Energy-Economical Routing (DEER) is designed to save the energy of the whole network through paying attention to both remained energy and the distribution of head nodes. The simulation results prove that this scheme can effectively balance the load and prolong the life of the wireless sensor networks.
2008, 30(5): 1234-1238.
doi: 10.3724/SP.J.1146.2006.01559
Abstract:
In this paper, the concept of IEEE 802.11 DCF error frame model is proposed. This paper points out the error of the simulation algorithm of the DCF error frame model in GloMoSim in detail, based on the analysis of the model. To solve this problem, the algorithm is modified strictly according to the specifications of the DCF error frame model. The simulation results show that compared with the original simulation algorithm, the modified algorithm can simulate the related specifications of the DCF error frame model more correctly, thus can provide more reliable simulation results.
In this paper, the concept of IEEE 802.11 DCF error frame model is proposed. This paper points out the error of the simulation algorithm of the DCF error frame model in GloMoSim in detail, based on the analysis of the model. To solve this problem, the algorithm is modified strictly according to the specifications of the DCF error frame model. The simulation results show that compared with the original simulation algorithm, the modified algorithm can simulate the related specifications of the DCF error frame model more correctly, thus can provide more reliable simulation results.
2008, 30(5): 1239-1242.
doi: 10.3724/SP.J.1146.2007.00787
Abstract:
There are two restrictions in the Versatile Place and Route tool, VPR. It can only support three kinds of switch box architecture, which are Disjoint, Wilton and Universal, and the same type of wires in a channel must be distributed next to each other. To break through these two restrictions, this paper proposes a hierarchical versatile switch box model, covering arbitrary switch box architecture in FPGA. Based on this model, this paper designs new switch box architecture, JSB. Comparing with Disjoint, Wilton and Universal architecture, JSB improves greatly routability by 10.1%, 3.3% and 4.6% respectively. Furthermore, in this paper, optimizing the distribution of wires reduces the timing of critical path by 10.4% on average, compared with VPR.
There are two restrictions in the Versatile Place and Route tool, VPR. It can only support three kinds of switch box architecture, which are Disjoint, Wilton and Universal, and the same type of wires in a channel must be distributed next to each other. To break through these two restrictions, this paper proposes a hierarchical versatile switch box model, covering arbitrary switch box architecture in FPGA. Based on this model, this paper designs new switch box architecture, JSB. Comparing with Disjoint, Wilton and Universal architecture, JSB improves greatly routability by 10.1%, 3.3% and 4.6% respectively. Furthermore, in this paper, optimizing the distribution of wires reduces the timing of critical path by 10.4% on average, compared with VPR.
2008, 30(5): 1243-1246.
doi: 10.3724/SP.J.1146.2007.01497
Abstract:
The electric field component of TM310 mode in cylindrical coaxial resonator presents periodic distribution along with circumference, the cylindrical resonator circumference borders with tunable inductors is irregularity and uniformity of the distribution is destruction. In order to accurately describe the distribution situation, in the paper, a irregular circumference border coaxial resonator physical model is established. The formula of the electric field distribution through CST and the numerical simulation, simulation results with theoretical analysis concluded unanimously.
The electric field component of TM310 mode in cylindrical coaxial resonator presents periodic distribution along with circumference, the cylindrical resonator circumference borders with tunable inductors is irregularity and uniformity of the distribution is destruction. In order to accurately describe the distribution situation, in the paper, a irregular circumference border coaxial resonator physical model is established. The formula of the electric field distribution through CST and the numerical simulation, simulation results with theoretical analysis concluded unanimously.
2008, 30(5): 1247-1250.
doi: 10.3724/SP.J.1146.2006.01733
Abstract:
The Multistage Depressed Collector (MDC) is widely used to improve the efficiency of microwave amplifiers. The emission of secondary electron in MDC has important influence on the efficiency. In this paper, the emission models of true secondary electron, elastic and inelastic reflected electron are discussed, and the corresponding emission ratio, angle and energy distribution are deeply analyzed. Moreover, the convergence criteria in MDC simulation is discussed, when the secondary electron are considered. The results of the analysis and discussion are used in the CAD software about MDC design, and the effects of secondary electron on the efficiency of MDC are quantitatively analyzed.
The Multistage Depressed Collector (MDC) is widely used to improve the efficiency of microwave amplifiers. The emission of secondary electron in MDC has important influence on the efficiency. In this paper, the emission models of true secondary electron, elastic and inelastic reflected electron are discussed, and the corresponding emission ratio, angle and energy distribution are deeply analyzed. Moreover, the convergence criteria in MDC simulation is discussed, when the secondary electron are considered. The results of the analysis and discussion are used in the CAD software about MDC design, and the effects of secondary electron on the efficiency of MDC are quantitatively analyzed.
2008, 30(5): 1257-1260.
doi: 10.3724/SP.J.1146.2006.01697
Abstract:
The pseudo-median filter is a computationally efficient alternative to the median filter. This paper extends the definition of pseudo-median filter when the size of filter windows is even, and proposes the Negative Weighted Pseudo-Median Filter (NWPMF). The NWPM uses the Mallows theory of sample selection probability to design the weight coefficients according to the respective impulse response of FIR filter with the desired spectral response. Because the weight coefficients can be negative or positive, the NWPM is characteristic of the frequency selection. The simulation results show that the NWPM can reduce the computations and is superior on reserving fine details, removing impulsive noise and high frequency periodic noise.
The pseudo-median filter is a computationally efficient alternative to the median filter. This paper extends the definition of pseudo-median filter when the size of filter windows is even, and proposes the Negative Weighted Pseudo-Median Filter (NWPMF). The NWPM uses the Mallows theory of sample selection probability to design the weight coefficients according to the respective impulse response of FIR filter with the desired spectral response. Because the weight coefficients can be negative or positive, the NWPM is characteristic of the frequency selection. The simulation results show that the NWPM can reduce the computations and is superior on reserving fine details, removing impulsive noise and high frequency periodic noise.
2008, 30(5): 1261-1263.
doi: 10.3724/SP.J.1146.2006.01656
Abstract:
In this paper, a cross-layer design in OFDM (Orthogonal Frequency Division Multiplexing) systems with a space-time coded transmit diversity is provided, which combines the use of AM (Adaptive Modulation) at the physical layer and ARQ (Automatic Repeat reQuest protocol) at the data link layer. The packet error rate and channel estimation are used to compute the adaptive threshold in order to maximize spectral efficiency under given delay and error performance constraints. Simulation results show that this cross-layer design has 1.5dB gain in spectral efficiency comparing the traditional adaptive transaction at physical layer. Nevertheless, with the increasing of maximum number of retransmissions, the gain in spectral efficiency decreases, which suggest that enough spectral efficiency gain can be gotten by a desirable delay in practical systems.
In this paper, a cross-layer design in OFDM (Orthogonal Frequency Division Multiplexing) systems with a space-time coded transmit diversity is provided, which combines the use of AM (Adaptive Modulation) at the physical layer and ARQ (Automatic Repeat reQuest protocol) at the data link layer. The packet error rate and channel estimation are used to compute the adaptive threshold in order to maximize spectral efficiency under given delay and error performance constraints. Simulation results show that this cross-layer design has 1.5dB gain in spectral efficiency comparing the traditional adaptive transaction at physical layer. Nevertheless, with the increasing of maximum number of retransmissions, the gain in spectral efficiency decreases, which suggest that enough spectral efficiency gain can be gotten by a desirable delay in practical systems.
2008, 30(5): 1264-1266.
doi: 10.3724/SP.J.1146.2006.01726
Abstract:
The advantage of majority-logic decoding algorithm is its very simple implementation; however, it provides modest error-correcting capability. In this paper, an improved scheme is proposed, which employs information provided by error symbols as well as correct ones in a received vector. The improved algorithm carries out minimum distance decoding, and complies with two optimal decoding rules, which make the error possibility of code word and the error possibility of code symbol minimum, respectively. Theoretical analysis and computer simulations show that the improvement can enhance efficiently error-correcting capability.
The advantage of majority-logic decoding algorithm is its very simple implementation; however, it provides modest error-correcting capability. In this paper, an improved scheme is proposed, which employs information provided by error symbols as well as correct ones in a received vector. The improved algorithm carries out minimum distance decoding, and complies with two optimal decoding rules, which make the error possibility of code word and the error possibility of code symbol minimum, respectively. Theoretical analysis and computer simulations show that the improvement can enhance efficiently error-correcting capability.
2008, 30(5): 1267-1270.
doi: 10.3724/SP.J.1146.2006.01560
Abstract:
ANSYS is used to analysis the heat dissipation capability of the slow-wave structures with different rods. The influence of the rods on heat dissipation capability is studied and compared separately under the steady and transient state. Several influencing factors are obtained, which could make some reference to the selection of suitable structures and dimensions of rods of the slow-wave structures.
ANSYS is used to analysis the heat dissipation capability of the slow-wave structures with different rods. The influence of the rods on heat dissipation capability is studied and compared separately under the steady and transient state. Several influencing factors are obtained, which could make some reference to the selection of suitable structures and dimensions of rods of the slow-wave structures.
2008, 30(5): 1251-1256.
doi: 10.3724/SP.J.1146.2006.01701
Abstract:
The robustness of digital watermarking to various attacks is a focus of researchers, while geometric attacks are very common attacks in practical applications. But conventional watermarking algorithms are helpless in face of geometric attacks, so the robustness to geometric attacks is called a bottleneck of commercial digital watermarking. The existing typical watermarking algorithms robust to geometric attacks are summarized, then they are compared and analyzed from the points of theoretical analyses and experimental validations. The possible development trends are pointed out, which can instruct the development of watermarking algorithms resistant to geometric attacks.
The robustness of digital watermarking to various attacks is a focus of researchers, while geometric attacks are very common attacks in practical applications. But conventional watermarking algorithms are helpless in face of geometric attacks, so the robustness to geometric attacks is called a bottleneck of commercial digital watermarking. The existing typical watermarking algorithms robust to geometric attacks are summarized, then they are compared and analyzed from the points of theoretical analyses and experimental validations. The possible development trends are pointed out, which can instruct the development of watermarking algorithms resistant to geometric attacks.
