Email alert
2012 Vol. 34, No. 3
Display Method:
2012, 34(3): 509-516.
doi: 10.3724/SP.J.1146.2011.00916
Abstract:
Currently, the Coherent Point Drift (CPD) which based on Gaussian Mixture Model (GMM) is one of popular point pattern matching algorithms because of its robustness. However, The CPD is local optimization and its convergent rate is slower along with the size of point-set become larger. For resolving these problems, this paper presents a Global Optimal and Fast algorithm which based on CPD (GOF-CPD). The orthogonal normalization first reduce the general affine case to the orthogonal case, and the convex region boundary of the unoberserved datas log likelihood nearby the global optimal solutions are deduced by the properties of normalized point-sets. Then, the multi-start strategy based on the convex region boundary is introduced to achieve the global optimization. Finally, a new iterative scheme, called the Trust Region based global convergent SQUARed iterative EM (TR-gSQUAREM), is proposed to achieve the superlinear convergence. Experiments on both synthetic point-sets and real world data show that the proposed algorithm is efficient, speedy and robust.
Currently, the Coherent Point Drift (CPD) which based on Gaussian Mixture Model (GMM) is one of popular point pattern matching algorithms because of its robustness. However, The CPD is local optimization and its convergent rate is slower along with the size of point-set become larger. For resolving these problems, this paper presents a Global Optimal and Fast algorithm which based on CPD (GOF-CPD). The orthogonal normalization first reduce the general affine case to the orthogonal case, and the convex region boundary of the unoberserved datas log likelihood nearby the global optimal solutions are deduced by the properties of normalized point-sets. Then, the multi-start strategy based on the convex region boundary is introduced to achieve the global optimization. Finally, a new iterative scheme, called the Trust Region based global convergent SQUARed iterative EM (TR-gSQUAREM), is proposed to achieve the superlinear convergence. Experiments on both synthetic point-sets and real world data show that the proposed algorithm is efficient, speedy and robust.
2012, 34(3): 517-522.
doi: 10.3724/SP.J.1146.2011.00826
Abstract:
In order to improve the efficiency of stereo matching and obtain high-precision sub-pixel disparity, a fast stereo matching method based on small baseline is proposed. Firstly, integral image is used to speed up the calculation of adaptive window and normalized cross correlation. Secondly, false match is rejected according to the reliability constraints. Thirdly, sub-pixel matching method base on iterative resample-to-twice scheme is utilized to compute sub-pixel disparity for reliable pixels. Finally, the disparity plane fitting method based on image segmentation is employed to obtain dense sub-pixel disparity map. Experimental results show that the proposed algorithm not only attains high-precision sub-pixel disparity, but also boosts the efficiency of the matching algorithm, and meets the requirements of small baseline based three-dimensional reconstruction.
In order to improve the efficiency of stereo matching and obtain high-precision sub-pixel disparity, a fast stereo matching method based on small baseline is proposed. Firstly, integral image is used to speed up the calculation of adaptive window and normalized cross correlation. Secondly, false match is rejected according to the reliability constraints. Thirdly, sub-pixel matching method base on iterative resample-to-twice scheme is utilized to compute sub-pixel disparity for reliable pixels. Finally, the disparity plane fitting method based on image segmentation is employed to obtain dense sub-pixel disparity map. Experimental results show that the proposed algorithm not only attains high-precision sub-pixel disparity, but also boosts the efficiency of the matching algorithm, and meets the requirements of small baseline based three-dimensional reconstruction.
2012, 34(3): 523-528.
doi: 10.3724/SP.J.1146.2011.00633
Abstract:
Multi-classification method under t distribution is proposed in order to solve the problem of the traditional face classification methods failing to settle tailing situation with heavy-tailed noise. By adjusting the degree of freedom parameter v, t distribution will adapt to a variety of tailing conditions better after heavy-tailed noise is added in facial images, at the same time, the recognition results will be improved. The experimental results on ORL and Yale show the feasibility and the effectiveness of the proposed algorithm.
Multi-classification method under t distribution is proposed in order to solve the problem of the traditional face classification methods failing to settle tailing situation with heavy-tailed noise. By adjusting the degree of freedom parameter v, t distribution will adapt to a variety of tailing conditions better after heavy-tailed noise is added in facial images, at the same time, the recognition results will be improved. The experimental results on ORL and Yale show the feasibility and the effectiveness of the proposed algorithm.
2012, 34(3): 529-534.
doi: 10.3724/SP.J.1146.2011.00757
Abstract:
Considering the poor stability of the video object tracking methods in outdoor scenes when the background circumstance variation in object images occurs, a new method is presented for object tracking based on infrared and visible dual-channel video. It extracts the Hue, Saturation, Value (HSV) color feature in visible image and the contour feature in infrared image, and combines the Mean Shift (MS) algorithm and the level set evolution algorithm to realize object location, also, the object scalar and model update mechanism is presented. To address the multiple object occlusion problems, a method is presented to locate multiple objects by determining the object merger and separation. Experimental results on infrared and visible dual-channel video demonstrate that the proposed method can successfully cope with the cases in complex environment such as illumination changes, shadow, occlusion, etc..
Considering the poor stability of the video object tracking methods in outdoor scenes when the background circumstance variation in object images occurs, a new method is presented for object tracking based on infrared and visible dual-channel video. It extracts the Hue, Saturation, Value (HSV) color feature in visible image and the contour feature in infrared image, and combines the Mean Shift (MS) algorithm and the level set evolution algorithm to realize object location, also, the object scalar and model update mechanism is presented. To address the multiple object occlusion problems, a method is presented to locate multiple objects by determining the object merger and separation. Experimental results on infrared and visible dual-channel video demonstrate that the proposed method can successfully cope with the cases in complex environment such as illumination changes, shadow, occlusion, etc..
2012, 34(3): 535-542.
doi: 10.3724/SP.J.1146.2011.00700
Abstract:
Although the function set of Linear HyperPlane (LHP) obtained from the Separating HyperPlane (SHP) method based on direct marketing campaigns has a very low Vapnik-Chervonenkis dimension equal to 9 or lower and the corresponding optimized LHP can fast detect unseen instance and preserve users privacy, it is inefficient in training speed, sensitive to training examples and not able to apply to nonlinear datasets. For overcoming these drawbacks as above, a nonlinear classification approach is proposed in this paper, which is suitable for large datasets and called Fast Ensemble of Separating HyperPlane (FE-SHP). First, the original training data is split into several subsets and their suboptimal LHPs are respectively constructed. Then, the nonlinear ensembling effects of suboptimal LHPs are enhanced by introducing an optimized weight vector after improving their nonlinear capabilities with Radical Basis Function (RBF). Finally, the related parameters are solved by the gradient descent method to maximize a log likelihood function which is the cross-entropy error of training data following the ensembling output of suboptimal LHPs being mapped probabilities. Experimental results on UCI demonstrate that the presented FE-SHP obtains competitive effectiveness for large datasets.
Although the function set of Linear HyperPlane (LHP) obtained from the Separating HyperPlane (SHP) method based on direct marketing campaigns has a very low Vapnik-Chervonenkis dimension equal to 9 or lower and the corresponding optimized LHP can fast detect unseen instance and preserve users privacy, it is inefficient in training speed, sensitive to training examples and not able to apply to nonlinear datasets. For overcoming these drawbacks as above, a nonlinear classification approach is proposed in this paper, which is suitable for large datasets and called Fast Ensemble of Separating HyperPlane (FE-SHP). First, the original training data is split into several subsets and their suboptimal LHPs are respectively constructed. Then, the nonlinear ensembling effects of suboptimal LHPs are enhanced by introducing an optimized weight vector after improving their nonlinear capabilities with Radical Basis Function (RBF). Finally, the related parameters are solved by the gradient descent method to maximize a log likelihood function which is the cross-entropy error of training data following the ensembling output of suboptimal LHPs being mapped probabilities. Experimental results on UCI demonstrate that the presented FE-SHP obtains competitive effectiveness for large datasets.
2012, 34(3): 543-548.
doi: 10.3724/SP.J.1146.2011.00782
Abstract:
Weblog co-clustering is an important research content of Weblog mining, which has ability to find out the users clusters and pages clusters simultaneously. Most of the proposed Weblog co-clustering algorithm use hard partition method to assign the users into its corresponding cluster. However, hard partition method make these clustering algorithm can not handle the clusters bond problem very well, which has significant influence for the clustering result quality. In this paper, a Fuzzy CO-clustering for Weblog (FCOW) algorithm is proposed to overcome the default of hard partition and improve the clustering results quality of Weblog co-clustering. In particularly, the underlying users model setPA={pa1,paK} is first found by using Hadamard product; and then, the rest users are assigned to its corresponding modelpak based on page subset to generate the co-clustering result {CSk, CPk}; Finally, the fuzzy membership of each user to its page clusterCPk is calculated and this information is used to do recommendation. Experimental results on five real world datasets show that FCOW has ability for improving the clustering quality of Weblog co-clustering.
Weblog co-clustering is an important research content of Weblog mining, which has ability to find out the users clusters and pages clusters simultaneously. Most of the proposed Weblog co-clustering algorithm use hard partition method to assign the users into its corresponding cluster. However, hard partition method make these clustering algorithm can not handle the clusters bond problem very well, which has significant influence for the clustering result quality. In this paper, a Fuzzy CO-clustering for Weblog (FCOW) algorithm is proposed to overcome the default of hard partition and improve the clustering results quality of Weblog co-clustering. In particularly, the underlying users model setPA={pa1,paK} is first found by using Hadamard product; and then, the rest users are assigned to its corresponding modelpak based on page subset to generate the co-clustering result {CSk, CPk}; Finally, the fuzzy membership of each user to its page clusterCPk is calculated and this information is used to do recommendation. Experimental results on five real world datasets show that FCOW has ability for improving the clustering quality of Weblog co-clustering.
2012, 34(3): 549-556.
doi: 10.3724/SP.J.1146.2011.00784
Abstract:
In the era of Web 2.0, the convenience of human-network interaction fosters new social characteristics of Web services such as rating, comment, etc., which are often ignored by service providers. So, it is difficult to achieve an efficient User eXperience (UX)-based recommendation of Web services. According to different users perspectives, this paper presents a comprehensive description model of Web services, and integrates their technical attributes (e.g., operation and message) and social characteristics by means of a multi-view approach. From the viewpoint of logical design, the conceptual model of Web Services Comprehensive Description Model (WSCDM) is given firstly. Then, the meta-model of WSCDM is constructed from the viewpoint of engineering application to guide the design of a specific model. Finally, a case study is shown to verify the feasibility of the presented method. This model and the related tools lay the foundation for active recommendation of Web services in service stores.
In the era of Web 2.0, the convenience of human-network interaction fosters new social characteristics of Web services such as rating, comment, etc., which are often ignored by service providers. So, it is difficult to achieve an efficient User eXperience (UX)-based recommendation of Web services. According to different users perspectives, this paper presents a comprehensive description model of Web services, and integrates their technical attributes (e.g., operation and message) and social characteristics by means of a multi-view approach. From the viewpoint of logical design, the conceptual model of Web Services Comprehensive Description Model (WSCDM) is given firstly. Then, the meta-model of WSCDM is constructed from the viewpoint of engineering application to guide the design of a specific model. Finally, a case study is shown to verify the feasibility of the presented method. This model and the related tools lay the foundation for active recommendation of Web services in service stores.
2012, 34(3): 557-563.
doi: 10.3724/SP.J.1146.2011.00724
Abstract:
A novel variational denoising model is proposed based on the study of two-step model. By analysing the properties of new model, an efficient and fast numerical algorithm is introduced. There are two variables in the new model, so it is firstly turned into two simple submodels by using alternative minimization method in the new algorithm, and then the two submodels are solved using split Bregman method respectively. Experimental results show new algorithm not only has faster convergence rate, but also can alleviate the staircase effect and preserve the edge information better while removing noises.
A novel variational denoising model is proposed based on the study of two-step model. By analysing the properties of new model, an efficient and fast numerical algorithm is introduced. There are two variables in the new model, so it is firstly turned into two simple submodels by using alternative minimization method in the new algorithm, and then the two submodels are solved using split Bregman method respectively. Experimental results show new algorithm not only has faster convergence rate, but also can alleviate the staircase effect and preserve the edge information better while removing noises.
2012, 34(3): 564-570.
doi: 10.3724/SP.J.1146.2011.00665
Abstract:
For improving the sinusoidal signal frequency estimation at low Signal-to-Noise Ratio (SNR), a frequency estimation algorithm based on spectrums correlation of Multi-Section Co-Frequency (MSCF) sinusoids signal is proposed. Firstly, the Optimization Weighted-Accumulation (OW-A) spectrum is gained through weighted-accumulating spectrums of the MSCF sinusoids signal by the weighted factor. Secondly, the correlation spectrum is constructed by correlating the OW-A spectrum and the accumulation spectrum of the MSCF sinusoids signal. Consequently, frequency estimation can be obtained through spectral peak searching of the correlation spectrum. Simulation results demonstrate the superior performance of the proposed algorithm.
For improving the sinusoidal signal frequency estimation at low Signal-to-Noise Ratio (SNR), a frequency estimation algorithm based on spectrums correlation of Multi-Section Co-Frequency (MSCF) sinusoids signal is proposed. Firstly, the Optimization Weighted-Accumulation (OW-A) spectrum is gained through weighted-accumulating spectrums of the MSCF sinusoids signal by the weighted factor. Secondly, the correlation spectrum is constructed by correlating the OW-A spectrum and the accumulation spectrum of the MSCF sinusoids signal. Consequently, frequency estimation can be obtained through spectral peak searching of the correlation spectrum. Simulation results demonstrate the superior performance of the proposed algorithm.
2012, 34(3): 571-575.
doi: 10.3724/SP.J.1146.2011.00839
Abstract:
A new rapid speaker adaptation method based on maximum likelihood variable subspace is proposed. A set of bases of the speaker space is obtained by performing Principal Component Analysis (PCA) on the Speaker Dependent (SD) model parameters of the training speakers. Different from conventional subspace based methods, during speaker adaptation, a subset of these bases is dynamically chosen for each speaker using maximum likelihood criteria. The new speakers model is constrained in the subspace spanned by those bases. With less free parameters required, the new method can obtain more robust SD model using very little amount of adaptation data. Speech recognition experiments show that the new method can obtain better performance than the eigenvoice method and MLLR method, both in supervised mode and in unsupervised mode.
A new rapid speaker adaptation method based on maximum likelihood variable subspace is proposed. A set of bases of the speaker space is obtained by performing Principal Component Analysis (PCA) on the Speaker Dependent (SD) model parameters of the training speakers. Different from conventional subspace based methods, during speaker adaptation, a subset of these bases is dynamically chosen for each speaker using maximum likelihood criteria. The new speakers model is constrained in the subspace spanned by those bases. With less free parameters required, the new method can obtain more robust SD model using very little amount of adaptation data. Speech recognition experiments show that the new method can obtain better performance than the eigenvoice method and MLLR method, both in supervised mode and in unsupervised mode.
2012, 34(3): 576-581.
doi: 10.3724/SP.J.1146.2011.00611
Abstract:
Two novel DOA (Direction Of Arrival) estimation methods are proposed using sparse representation when the signal number is unknown. One is the method using sparse representation based on the eigenvector of covariance matrix. The biggest eigenvector of covariance matrix is proved to be the linear combination of all steer vectors and is extracted to build sparse representation model for DOA estimation. The other is the method using sparse representation of high-order power of covariance Matrix. This method approximates the signal sub-space through the high order power of the spatial covariance matrix on the basic of signal eigenvalue being larger than noise eigenvalue. Then the column vector of high order power of the spatial covariance matrix is extracted to construct the sparse representation model for DOA estimation. The theoretical analysis and experimental results show the two methods have a better performance than the MUSIC algorithm in the aspects of accuracy, resolution and adaptability to coherent signals without estimating the number of signals.
Two novel DOA (Direction Of Arrival) estimation methods are proposed using sparse representation when the signal number is unknown. One is the method using sparse representation based on the eigenvector of covariance matrix. The biggest eigenvector of covariance matrix is proved to be the linear combination of all steer vectors and is extracted to build sparse representation model for DOA estimation. The other is the method using sparse representation of high-order power of covariance Matrix. This method approximates the signal sub-space through the high order power of the spatial covariance matrix on the basic of signal eigenvalue being larger than noise eigenvalue. Then the column vector of high order power of the spatial covariance matrix is extracted to construct the sparse representation model for DOA estimation. The theoretical analysis and experimental results show the two methods have a better performance than the MUSIC algorithm in the aspects of accuracy, resolution and adaptability to coherent signals without estimating the number of signals.
2012, 34(3): 582-586.
doi: 10.3724/SP.J.1146.2011.00801
Abstract:
A multi-target location algorithm is proposed for MIMO radar with extended-aperture array of electromagnetic vector sensors. A novel bistatic MIMO radar system with multiple transmit sensors and multiple receive electromagnetic vectors is introduced, and receive element spacing is greater than half-wavelength. The ESPRIT method is employed to extract a set of high accurate but ambiguous Direction Of Arrive (DOA), then the low accurate but unambiguous estimates which are obtained utilizing the internal structure feature of the vector sensors and subspace rotation invariance are used to disambiguate the ambiguous DOA to yield high accurate and unambiguous DOA. One-dimensional MUSIC method is employed to get the Departure Of Direction (DOD) estimates with the known DOA. The algorithm improves greatly the accuracy of MIMO radar location, requires no two-dimensional searches and parameter pairing, thus showing low computational complexity. Simulation results verify the effectiveness of the proposed algorithm, and the estimation accuracy is close to the Cramer-Rao Bound (CRB).
A multi-target location algorithm is proposed for MIMO radar with extended-aperture array of electromagnetic vector sensors. A novel bistatic MIMO radar system with multiple transmit sensors and multiple receive electromagnetic vectors is introduced, and receive element spacing is greater than half-wavelength. The ESPRIT method is employed to extract a set of high accurate but ambiguous Direction Of Arrive (DOA), then the low accurate but unambiguous estimates which are obtained utilizing the internal structure feature of the vector sensors and subspace rotation invariance are used to disambiguate the ambiguous DOA to yield high accurate and unambiguous DOA. One-dimensional MUSIC method is employed to get the Departure Of Direction (DOD) estimates with the known DOA. The algorithm improves greatly the accuracy of MIMO radar location, requires no two-dimensional searches and parameter pairing, thus showing low computational complexity. Simulation results verify the effectiveness of the proposed algorithm, and the estimation accuracy is close to the Cramer-Rao Bound (CRB).
2012, 34(3): 587-593.
doi: 10.3724/SP.J.1146.2011.00713
Abstract:
In the conventional SAR Ground Moving Targets Indication (GMTI) method, the sample number is heavily large, increasing severely data transmission and storage load. To mitigate this problem, a SAR GMTI method is proposed based on Compressive Sensing (CS) in this paper. In the presented method, raw radar data of dual channels are firstly acquired by sampling randomly and sparsely in the azimuth direction. Secondly, the matched filter is used to perform the range direction focus and compressive sensing method is used to perform the azimuth direction focus. Finally, the conventional Displaced Phase Center Antenna (DPCA) technique is adopted to suppress clutter. Theoretical analysis shows that the proposed method can be applied to clutter suppression of the sparse sampling data of dual channels and the effects of motion parameters on target imaging are analysed in detail. The results of simulated and real data processing verify that the proposed method has excellent clutter suppression performance even if few samples can be obtained in the azimuth direction.
In the conventional SAR Ground Moving Targets Indication (GMTI) method, the sample number is heavily large, increasing severely data transmission and storage load. To mitigate this problem, a SAR GMTI method is proposed based on Compressive Sensing (CS) in this paper. In the presented method, raw radar data of dual channels are firstly acquired by sampling randomly and sparsely in the azimuth direction. Secondly, the matched filter is used to perform the range direction focus and compressive sensing method is used to perform the azimuth direction focus. Finally, the conventional Displaced Phase Center Antenna (DPCA) technique is adopted to suppress clutter. Theoretical analysis shows that the proposed method can be applied to clutter suppression of the sparse sampling data of dual channels and the effects of motion parameters on target imaging are analysed in detail. The results of simulated and real data processing verify that the proposed method has excellent clutter suppression performance even if few samples can be obtained in the azimuth direction.
2012, 34(3): 594-601.
doi: 10.3724/SP.J.1146.2011.00751
Abstract:
Detection of multiple unresolved targets is a frontier topic and pressing task in radar technology field. A signal fusion method of a T/R-R2 hybrid multistatic radar system is presented, target echo signal models of a monostatic radar receiver and two multistatic radar receivers are established, considering three-dimensional space situation, the signal fusion model is presented based on range information and the multiple unresolved targets are detected with maximum likelihood method, and the effects of SNR, number of targets, target fluctuation, detection threshold and hybrid multistatic radar geometrical configuration on the detection performance are analyzed. The Monte Carlo simulation results identify the effectiveness of the method. The detection method is presented under the assumed condition of Swerling II targets, but it is still suitable for Swerling 0 and IV targets.
Detection of multiple unresolved targets is a frontier topic and pressing task in radar technology field. A signal fusion method of a T/R-R2 hybrid multistatic radar system is presented, target echo signal models of a monostatic radar receiver and two multistatic radar receivers are established, considering three-dimensional space situation, the signal fusion model is presented based on range information and the multiple unresolved targets are detected with maximum likelihood method, and the effects of SNR, number of targets, target fluctuation, detection threshold and hybrid multistatic radar geometrical configuration on the detection performance are analyzed. The Monte Carlo simulation results identify the effectiveness of the method. The detection method is presented under the assumed condition of Swerling II targets, but it is still suitable for Swerling 0 and IV targets.
2012, 34(3): 602-608.
doi: 10.3724/SP.J.1146.2011.00607
Abstract:
The translational motion compensation methods for conventional Inverse Synthetic Aperture Radar (ISAR) are not suitable for ballistic targets. In the paper, an algorithm for ballistic targets translational motion compensation and micro-Doppler (m-D) feature extraction is proposed. Based on the analysis of m-D effect of ballistic targets with tails without translational motion compensation, the skeleton extraction algorithm in morphology image processing is firstly utilized to suppress the sidelobes of range profile, and then the m-D curves on the range-slowtime plane are separated. The Empirical-Mode Decomposition (EMD) algorithm is then utilized to obtain the translational motion component from the m-D curves for translational motion compensation. The micro-motion features of target such as spinning frequency and coning frequency are also obtained from the EMD results. Simulations are given to validate the effectiveness and robustness of the proposed algorithm.
The translational motion compensation methods for conventional Inverse Synthetic Aperture Radar (ISAR) are not suitable for ballistic targets. In the paper, an algorithm for ballistic targets translational motion compensation and micro-Doppler (m-D) feature extraction is proposed. Based on the analysis of m-D effect of ballistic targets with tails without translational motion compensation, the skeleton extraction algorithm in morphology image processing is firstly utilized to suppress the sidelobes of range profile, and then the m-D curves on the range-slowtime plane are separated. The Empirical-Mode Decomposition (EMD) algorithm is then utilized to obtain the translational motion component from the m-D curves for translational motion compensation. The micro-motion features of target such as spinning frequency and coning frequency are also obtained from the EMD results. Simulations are given to validate the effectiveness and robustness of the proposed algorithm.
2012, 34(3): 609-615.
doi: 10.3724/SP.J.1146.2011.00546
Abstract:
Micro-motion feature is one of the effective features used for radar target recognition in the middle section of the ballistic curve. Using the scattering centers of the cone-shaped warhead analyzed by the Geometrical Theory of Diffraction (GTD), it is concluded that the cone-shaped warhead has three scattering centers, one is at the conical point, and the other two are the points where the projection of the bistatic bisector on the conical bottom intersects the edge of the conical bottom, firstly. Then the micro-Doppler (m-D) models of the scattering center at the conical point and two slip-type scattering centers on the edge are derived. They are similar with the monostatic m-D models of the scattering centers when the radar locates on the bistatic bisector, and the amplitude ratio of the m-D in the bistatic radar and monostatic radar is the cosine of the half of the bistatic angle. It coincides with the monostatic-bistatic equivalence theorem. Three correlations of the m-D of the three scattering centers are founded by analyzing their m-D expressions. Finally, the bistatic Radar Cross Section(RCS) of the cone-shaped warhead model are calculated by using FEKO. The theoretic analysis of the m-D expressions is verified from the time-frequency analysis on the bistatic scattering coefficients.
Micro-motion feature is one of the effective features used for radar target recognition in the middle section of the ballistic curve. Using the scattering centers of the cone-shaped warhead analyzed by the Geometrical Theory of Diffraction (GTD), it is concluded that the cone-shaped warhead has three scattering centers, one is at the conical point, and the other two are the points where the projection of the bistatic bisector on the conical bottom intersects the edge of the conical bottom, firstly. Then the micro-Doppler (m-D) models of the scattering center at the conical point and two slip-type scattering centers on the edge are derived. They are similar with the monostatic m-D models of the scattering centers when the radar locates on the bistatic bisector, and the amplitude ratio of the m-D in the bistatic radar and monostatic radar is the cosine of the half of the bistatic angle. It coincides with the monostatic-bistatic equivalence theorem. Three correlations of the m-D of the three scattering centers are founded by analyzing their m-D expressions. Finally, the bistatic Radar Cross Section(RCS) of the cone-shaped warhead model are calculated by using FEKO. The theoretic analysis of the m-D expressions is verified from the time-frequency analysis on the bistatic scattering coefficients.
2012, 34(3): 616-621.
doi: 10.3724/SP.J.1146.2011.00650
Abstract:
Wide Angle Synthetic Aperture Radar (WASAR) undergoes a long data collection time and strong electromagnetic, radio frequency interference and azimuth disturbance could lead to missing-data problems. It suffers from aliasing errors, ambiguity and high sidelobes if using FFT method for imaging. This paper first expands the Adaptive Tuning Amplitude and Phase EStimation (ATAPES) method into two-dimension, then, the tuning factor is introduced to data extrapolating process and an Adaptive Tuning Missing-data Amplitude and Phase EStimation (ATMAPES) is proposed. The optimized extrapolated result is filled back into the missing data and the accurate estimating results of target location and amplitude is obtained through circular iteration. The feasibility and effectiveness of the proposed method are validated by simulation results.
Wide Angle Synthetic Aperture Radar (WASAR) undergoes a long data collection time and strong electromagnetic, radio frequency interference and azimuth disturbance could lead to missing-data problems. It suffers from aliasing errors, ambiguity and high sidelobes if using FFT method for imaging. This paper first expands the Adaptive Tuning Amplitude and Phase EStimation (ATAPES) method into two-dimension, then, the tuning factor is introduced to data extrapolating process and an Adaptive Tuning Missing-data Amplitude and Phase EStimation (ATMAPES) is proposed. The optimized extrapolated result is filled back into the missing data and the accurate estimating results of target location and amplitude is obtained through circular iteration. The feasibility and effectiveness of the proposed method are validated by simulation results.
2012, 34(3): 622-627.
doi: 10.3724/SP.J.1146.2011.00472
Abstract:
Bistatic ISAR systems can receive echoes from different observation angles simultaneously, thus a finer image with more abundant information of a certain target can be obtained by fusing the two images attained with bistatic radars. However, the dis-match of dimensions and shapes of the two target images caused by bistatic angle highly complicates the image fusion process. This paper, based on the bistatic ISAR turn table model, proposes an ISAR image fusion method based on sub-aperture parameter estimation. The algorithm first extracts the Doppler difference between two sub-apertures of an isolated prominent scattering point to estimate the Rotation Velocity (RV) and half bistatic angle, then maps the two data sets from different observation angles onto a global coordinate system to perform the image fusion process, resulting in an ISAR image of higher quality. Finally, simulation experiments are provided to demonstrate the effectiveness of the method proposed.
Bistatic ISAR systems can receive echoes from different observation angles simultaneously, thus a finer image with more abundant information of a certain target can be obtained by fusing the two images attained with bistatic radars. However, the dis-match of dimensions and shapes of the two target images caused by bistatic angle highly complicates the image fusion process. This paper, based on the bistatic ISAR turn table model, proposes an ISAR image fusion method based on sub-aperture parameter estimation. The algorithm first extracts the Doppler difference between two sub-apertures of an isolated prominent scattering point to estimate the Rotation Velocity (RV) and half bistatic angle, then maps the two data sets from different observation angles onto a global coordinate system to perform the image fusion process, resulting in an ISAR image of higher quality. Finally, simulation experiments are provided to demonstrate the effectiveness of the method proposed.
2012, 34(3): 628-632.
doi: 10.3724/SP.J.1146.2011.00702
Abstract:
Multi-channel SAR can achieve high resolution and wide swath at the same time, which overcoming the inherent limitations of conventional SAR system. Azimuth uniformly sampled data must be reconstructed in imaging procession. A novel method based on fall-rise chirp rate technology is proposed. The original signal is dechirped firstly; then, it is up sampled, and reconstructed by inverse processing of dechirping. Data processing flow is given, and the reconstruction effect of this new method and classical ways using reconstruction filter banks are compared. All simulation results prove the correctness of theoretic analysis.
Multi-channel SAR can achieve high resolution and wide swath at the same time, which overcoming the inherent limitations of conventional SAR system. Azimuth uniformly sampled data must be reconstructed in imaging procession. A novel method based on fall-rise chirp rate technology is proposed. The original signal is dechirped firstly; then, it is up sampled, and reconstructed by inverse processing of dechirping. Data processing flow is given, and the reconstruction effect of this new method and classical ways using reconstruction filter banks are compared. All simulation results prove the correctness of theoretic analysis.
2012, 34(3): 633-638.
doi: 10.3724/SP.J.1146.2011.00750
Abstract:
It is a common issue for Ballistic Missile (BM) to separate warheads in order to improve the penetration probability during reentry phase. For the reasons of unknown warhead number, closeness between target and warheads and similarity of reentry velocities of the warheads which make them moving as a group, how to rapidly associate the separating warheads without any prior information has become an urgent problem. This paper proposes an improved real-time sliding window Markov Chain-Monte Carlo (MCMC) suboptimal association algorithm. By calculating the maximum posterior probability of combination from the surveillance area observations using Monte Carlo method, the algorithm approximates the Markov Chains stable distribution. Furthermore, considering the warhead separation reality, the sliding window MCMC reassigns the weights of the probability association hypothesis and optimizes the inheritance yielding greatly reduction in computation. Simulation results show that the proposed algorithm yields significant improvements both in association and computation performance under heavy dense targets compared with classical Multiple Hypothesis Tracking (MHT).
It is a common issue for Ballistic Missile (BM) to separate warheads in order to improve the penetration probability during reentry phase. For the reasons of unknown warhead number, closeness between target and warheads and similarity of reentry velocities of the warheads which make them moving as a group, how to rapidly associate the separating warheads without any prior information has become an urgent problem. This paper proposes an improved real-time sliding window Markov Chain-Monte Carlo (MCMC) suboptimal association algorithm. By calculating the maximum posterior probability of combination from the surveillance area observations using Monte Carlo method, the algorithm approximates the Markov Chains stable distribution. Furthermore, considering the warhead separation reality, the sliding window MCMC reassigns the weights of the probability association hypothesis and optimizes the inheritance yielding greatly reduction in computation. Simulation results show that the proposed algorithm yields significant improvements both in association and computation performance under heavy dense targets compared with classical Multiple Hypothesis Tracking (MHT).
2012, 34(3): 639-643.
doi: 10.3724/SP.J.1146.2011.00673
Abstract:
Class B noise model is statistical-physical model for atmospheric noise. This paper proposes a nonlinear regression-type algorithm to estimate the parameters of Class B model. The algorithm based on characteristic function is derived from nonlinear regression model, which has low iterations. An initial estimators are also designed to accelerate the convergence of algorithm, and a special series are utilized to calculate the log characteristic function to solve the issue of many zero points of characteristic function. The result shows that new method has high precisions and low iterations, which can be applied excellently to practice.
Class B noise model is statistical-physical model for atmospheric noise. This paper proposes a nonlinear regression-type algorithm to estimate the parameters of Class B model. The algorithm based on characteristic function is derived from nonlinear regression model, which has low iterations. An initial estimators are also designed to accelerate the convergence of algorithm, and a special series are utilized to calculate the log characteristic function to solve the issue of many zero points of characteristic function. The result shows that new method has high precisions and low iterations, which can be applied excellently to practice.
2012, 34(3): 644-649.
doi: 10.3724/SP.J.1146.2011.00567
Abstract:
A deconvolution algorithm based on Compressive Sensing (CS) is proposed for the post-processing of Ultra WideBand (UWB) channel modeling using frequency-domain measurements. A window with Gaussian transition band is used to extract the measurements according to the UWB frequency regulation policy of China. The time-domain waveform of the quasi-Gaussian window is used as the apriori information of the CS based deconvolution algorithm. The deconvolution results are with high-resolution characteristic. Furthermore, flexible zero-padding of windowing and the design of parameterized waveform dictionary lead to different resolutions of the deconvolution results. Matching Pursuit (MP) algorithm is used as the reconstruction algorithm of CS. Both LOS and NLOS measurements of offices are exploited to demonstrate that the proposed CS based deconvolution algorithm can achieve comparable performance with CLEAN algorithm using fewer samples.
A deconvolution algorithm based on Compressive Sensing (CS) is proposed for the post-processing of Ultra WideBand (UWB) channel modeling using frequency-domain measurements. A window with Gaussian transition band is used to extract the measurements according to the UWB frequency regulation policy of China. The time-domain waveform of the quasi-Gaussian window is used as the apriori information of the CS based deconvolution algorithm. The deconvolution results are with high-resolution characteristic. Furthermore, flexible zero-padding of windowing and the design of parameterized waveform dictionary lead to different resolutions of the deconvolution results. Matching Pursuit (MP) algorithm is used as the reconstruction algorithm of CS. Both LOS and NLOS measurements of offices are exploited to demonstrate that the proposed CS based deconvolution algorithm can achieve comparable performance with CLEAN algorithm using fewer samples.
2012, 34(3): 650-656.
doi: 10.3724/SP.J.1146.2011.00394
Abstract:
To utilize the polarization state of primary signal which is the crucial vector characteristic for spectrum sensing, a rapid polarization adaptation algorithm is proposed based on the K-armed bandit. The proposed algorithm can reduce the complexity and converge rapidly by transferring the two-degree-of-freedom search issue to K single-dimensional issues, and identify the polarization state of primary signal to realize real-time spectrum sensing for Cognitive Radios (CR). Further, the theoretical boundary of the proposed algorithm is derived. The spectrum sensing performance on the basis of the polarization state identification is discussed by receiver operation characteristic curve finally. The simulation results show that the proposed algorithm converges fast, and obtains a high accuracy on polarization state identification for primary signal. The detection performance is more effective with the rapid polarization adaptation.
To utilize the polarization state of primary signal which is the crucial vector characteristic for spectrum sensing, a rapid polarization adaptation algorithm is proposed based on the K-armed bandit. The proposed algorithm can reduce the complexity and converge rapidly by transferring the two-degree-of-freedom search issue to K single-dimensional issues, and identify the polarization state of primary signal to realize real-time spectrum sensing for Cognitive Radios (CR). Further, the theoretical boundary of the proposed algorithm is derived. The spectrum sensing performance on the basis of the polarization state identification is discussed by receiver operation characteristic curve finally. The simulation results show that the proposed algorithm converges fast, and obtains a high accuracy on polarization state identification for primary signal. The detection performance is more effective with the rapid polarization adaptation.
2012, 34(3): 657-663.
doi: 10.3724/SP.J.1146.2011.00330
Abstract:
This paper investigates transceiver design, that minimizes the transmit power and balances all SINRs, for multiuser Multi-Input Multi-Output (MIMO) systems in the presence of imperfect Channel State Information (CSI). A worst-case SINR uplink-downlink duality is established under bounded channel uncertainties, and with the aids of this duality, a new scheme is proposed for the robust transceiver design. This scheme obviates convex optimization which uses interior-point iterative algorithm, and thus the whole system complexity can be greatly reduced. Numerical results show that the proposed algorithms exhibit the same performance when compared with the exiting robust algorithm.
This paper investigates transceiver design, that minimizes the transmit power and balances all SINRs, for multiuser Multi-Input Multi-Output (MIMO) systems in the presence of imperfect Channel State Information (CSI). A worst-case SINR uplink-downlink duality is established under bounded channel uncertainties, and with the aids of this duality, a new scheme is proposed for the robust transceiver design. This scheme obviates convex optimization which uses interior-point iterative algorithm, and thus the whole system complexity can be greatly reduced. Numerical results show that the proposed algorithms exhibit the same performance when compared with the exiting robust algorithm.
2012, 34(3): 664-669.
doi: 10.3724/SP.J.1146.2011.00655
Abstract:
The presence of the Line-Of-Sight (LOS) component will increase the condition number of the multi antenna channel matrix, and severely reduce detection performance of the synchronous multi-carrier layered space-time code. As a remedy to this problem, a multi-carrier layered space-time architecture is proposed based on the asynchronous transmission of each antennas signal in frequency domain. Closed-form Bit Error Rate (BER) expressions of the Zero Forcing (ZF) joint detection for a Ricean fading channel is derived for both coded and uncoded modulation. Comparisons are also made with those obtained in a traditional synchronous system. The theoretical analysis and simulation results show that the asynchronous multi-carrier transmission scheme can break through the antenna number limitation of synchronous multi-carrier layered space-time code system, while achieving full receive diversity with low complexity linear detector.
The presence of the Line-Of-Sight (LOS) component will increase the condition number of the multi antenna channel matrix, and severely reduce detection performance of the synchronous multi-carrier layered space-time code. As a remedy to this problem, a multi-carrier layered space-time architecture is proposed based on the asynchronous transmission of each antennas signal in frequency domain. Closed-form Bit Error Rate (BER) expressions of the Zero Forcing (ZF) joint detection for a Ricean fading channel is derived for both coded and uncoded modulation. Comparisons are also made with those obtained in a traditional synchronous system. The theoretical analysis and simulation results show that the asynchronous multi-carrier transmission scheme can break through the antenna number limitation of synchronous multi-carrier layered space-time code system, while achieving full receive diversity with low complexity linear detector.
2012, 34(3): 670-676.
doi: 10.3724/SP.J.1146.2011.00515
Abstract:
This paper considers multi-carrier heterogeneous cellular networks where only limited information can be shared among Base Stations (BSs) via backhaul links. Considering that there is frequency offset between macro and micro BSs, an optimal precoder and a suboptimal precoder are proposed, aimed at maximizing the date rate of micro systems while ensuring the resulting interference to macro users to be less than an interference threshold, where either instantaneous or statistical interference channel information is known at micro BS. Analysis and simulation results show that the performance of the suboptimal algorithm with only one iteration is close to the performance of the optimal one, and hence it has a lower computational complexity. The proposed method requires only little amount of information to be exchanged between BSs via backhaul links, and thus has a low overhead for sharing information. The proposed method is applicable when there are multiple marco users around the micro BS.
This paper considers multi-carrier heterogeneous cellular networks where only limited information can be shared among Base Stations (BSs) via backhaul links. Considering that there is frequency offset between macro and micro BSs, an optimal precoder and a suboptimal precoder are proposed, aimed at maximizing the date rate of micro systems while ensuring the resulting interference to macro users to be less than an interference threshold, where either instantaneous or statistical interference channel information is known at micro BS. Analysis and simulation results show that the performance of the suboptimal algorithm with only one iteration is close to the performance of the optimal one, and hence it has a lower computational complexity. The proposed method requires only little amount of information to be exchanged between BSs via backhaul links, and thus has a low overhead for sharing information. The proposed method is applicable when there are multiple marco users around the micro BS.
2012, 34(3): 677-682.
doi: 10.3724/SP.J.1146.2011.00707
Abstract:
For improving broadcasting efficiency of wireless network, a novel Network-Coding (NC)-based type-III Hybrid Automatic Repeat reQuest (HARQ) mechanism with Joint Network and Channel Decoding (JNCD) strategy is presented. A combination strategy is developed to perform XORing Network Coding Combined (XNCC) retransmission. The main idea is to recover packets for most users simultaneously using the network coding technology and obtain the soft information of missing packets to implement JNCD in each retransmission phase. Moreover, The proposed NC-HARQ III mechanism is extended to relay system by opportunistic relaying selected strategy. Analytical results and simulation illustrate that the average number of transmissions can be sharply decreased in the whole network especially in severe fading circumstance. Compared with type-I HARQ scheme and NC-HARQ I scheme, the delay gains can be attributed to the potential advantage of network coding and the better PER performance of JNCD strategy.
For improving broadcasting efficiency of wireless network, a novel Network-Coding (NC)-based type-III Hybrid Automatic Repeat reQuest (HARQ) mechanism with Joint Network and Channel Decoding (JNCD) strategy is presented. A combination strategy is developed to perform XORing Network Coding Combined (XNCC) retransmission. The main idea is to recover packets for most users simultaneously using the network coding technology and obtain the soft information of missing packets to implement JNCD in each retransmission phase. Moreover, The proposed NC-HARQ III mechanism is extended to relay system by opportunistic relaying selected strategy. Analytical results and simulation illustrate that the average number of transmissions can be sharply decreased in the whole network especially in severe fading circumstance. Compared with type-I HARQ scheme and NC-HARQ I scheme, the delay gains can be attributed to the potential advantage of network coding and the better PER performance of JNCD strategy.
2012, 34(3): 683-688.
doi: 10.3724/SP.J.1146.2011.00436
Abstract:
Millimeter-wave signals make the integration of antenna arrays technology possible, which can be employed to implement beamforming. However, with the increasing of antenna elements, the complexity of beam searching will also significantly increase. By fully exploring beam characteristics generated with different numbers of array elements, an efficient beam searching algorithm is proposed relying on a novel conception of dividing and conquering. By increasing progressively the number of antenna elements, the search space is noticeably reduced, and the optimal beam pair can be iteratively discovered. Compared with the existing popular schemes, the proposed algorithm can significantly reduce preamble transmissions accompanying energy consumption. Numerical simulation results validate the superiority of the proposed algorithm.
Millimeter-wave signals make the integration of antenna arrays technology possible, which can be employed to implement beamforming. However, with the increasing of antenna elements, the complexity of beam searching will also significantly increase. By fully exploring beam characteristics generated with different numbers of array elements, an efficient beam searching algorithm is proposed relying on a novel conception of dividing and conquering. By increasing progressively the number of antenna elements, the search space is noticeably reduced, and the optimal beam pair can be iteratively discovered. Compared with the existing popular schemes, the proposed algorithm can significantly reduce preamble transmissions accompanying energy consumption. Numerical simulation results validate the superiority of the proposed algorithm.
2012, 34(3): 689-693.
doi: 10.3724/SP.J.1146.2011.00732
Abstract:
Distributed algorithms which can find the Maximum Weight Independent Set (MWIS) of graphs are very important for the design of wireless networks. Based on the max-product belief propagation framework, a new distributed MWIS algorithm is proposed. In this algorithm, each node is assumed knowing the local connection information among its neighbors. Helped by this local information, the proposed algorithm outperforms existing algorithms which requirs that the graph must be tree or bipartite, which is verified by simulation results.
Distributed algorithms which can find the Maximum Weight Independent Set (MWIS) of graphs are very important for the design of wireless networks. Based on the max-product belief propagation framework, a new distributed MWIS algorithm is proposed. In this algorithm, each node is assumed knowing the local connection information among its neighbors. Helped by this local information, the proposed algorithm outperforms existing algorithms which requirs that the graph must be tree or bipartite, which is verified by simulation results.
2012, 34(3): 694-697.
doi: 10.3724/SP.J.1146.2011.00888
Abstract:
3D cipher is a new block cipher with Substitution-Permutation Network (SPN) structure. Unlike the Advanced Encryption Standard (AES), 3D cipher uses a 3-dimensional state. Based on the structure of 3D cipher, a 5-round meet-in-the-middle distinguisher and a new attack on 10-round 3D are found. It is shown that the attack requires a data complexity of about 2128 chosen plaintexts and a time complexity of about 2331.1 10-round 3D encryptions. Compared with the previous published cryptanalytic results, the attack reduces the data complexity and time complexity efficiently.
3D cipher is a new block cipher with Substitution-Permutation Network (SPN) structure. Unlike the Advanced Encryption Standard (AES), 3D cipher uses a 3-dimensional state. Based on the structure of 3D cipher, a 5-round meet-in-the-middle distinguisher and a new attack on 10-round 3D are found. It is shown that the attack requires a data complexity of about 2128 chosen plaintexts and a time complexity of about 2331.1 10-round 3D encryptions. Compared with the previous published cryptanalytic results, the attack reduces the data complexity and time complexity efficiently.
2012, 34(3): 698-703.
doi: 10.3724/SP.J.1146.2011.00799
Abstract:
Since current video quality assessment method based on Just Noticeable Distortion (JND) model can only assess monscopic video, a binocular JND model considering binocular luminance relationship is explored. First, binocular luminance relationship model is used according to stereo video characteristic. Then, many perception factors such as the background luminance masking effect, texture masking effect, intra-frame masking effect, spatiotemporal contrast sensitivity function, eye movement are considered. Finally, binocular JND model is applied to stereo video quality assessment. Experimental results prove that the proposed method can reflect the human visual perception characteristics and has a more consistency with the subjective judgments.
Since current video quality assessment method based on Just Noticeable Distortion (JND) model can only assess monscopic video, a binocular JND model considering binocular luminance relationship is explored. First, binocular luminance relationship model is used according to stereo video characteristic. Then, many perception factors such as the background luminance masking effect, texture masking effect, intra-frame masking effect, spatiotemporal contrast sensitivity function, eye movement are considered. Finally, binocular JND model is applied to stereo video quality assessment. Experimental results prove that the proposed method can reflect the human visual perception characteristics and has a more consistency with the subjective judgments.
2012, 34(3): 704-709.
doi: 10.3724/SP.J.1146.2011.00828
Abstract:
Device-to-Device (D2D) communication is a new technology that allows mobile terminals communicate with each other directly by using the licensed frequency resources under the control of cellular networks. D2D communications help to improve the efficiency of data dissemination services in cellular networks, via enabling cooperative inter-recipient data forwarding within a D2D cluster. However, the existing D2D forwarding algorithms do not take into account the heterogeneity of D2D channel capacity, and thus the resource utilization efficiency can not be maximized. In this paper, a cooperative multi-hop D2D forwarding algorithm which includes multicast and unicast modes is proposed. The proposed algorithm can fully exploit the multi-channel diversity gain by adaptively selecting the optimal relay, routing and hops of D2D transmission according to the D2D channel conditions. Numerical results show that the proposed algorithm achieves a significant gain in terms of resource utilization efficiency and therefore improves greatly the throughput of cellular data dissemination.
Device-to-Device (D2D) communication is a new technology that allows mobile terminals communicate with each other directly by using the licensed frequency resources under the control of cellular networks. D2D communications help to improve the efficiency of data dissemination services in cellular networks, via enabling cooperative inter-recipient data forwarding within a D2D cluster. However, the existing D2D forwarding algorithms do not take into account the heterogeneity of D2D channel capacity, and thus the resource utilization efficiency can not be maximized. In this paper, a cooperative multi-hop D2D forwarding algorithm which includes multicast and unicast modes is proposed. The proposed algorithm can fully exploit the multi-channel diversity gain by adaptively selecting the optimal relay, routing and hops of D2D transmission according to the D2D channel conditions. Numerical results show that the proposed algorithm achieves a significant gain in terms of resource utilization efficiency and therefore improves greatly the throughput of cellular data dissemination.
2012, 34(3): 710-715.
doi: 10.3724/SP.J.1146.2011.01032
Abstract:
Routing and Wavelength Assignment (RWA) is a key issue in next-generation optical networks. This paper presents an improved ant colony algorithm (Smart Ant-Dynamic RWA, SA-DRWA) to solve the RWA problem in optical networks. The link idle ratio is introduced into the ant transition probability as new constraint. At the same time, the random disturbance is added to prevent premature convergence into local optimal path within the less iteration. Theoretical analysis shows that the SA-DRWA algorithm can effectively achieve load balancing in optical network. Simulation for different topologies shows that the SA-DRWA algorithm has both improvements of blocking probability and resources utilization comparing with the traditional Dijkstra+FF algorithm. An optimal improvement scenario is obtained in regular mesh network with up to 0.23 improved for blocking probability and resources utilization.
Routing and Wavelength Assignment (RWA) is a key issue in next-generation optical networks. This paper presents an improved ant colony algorithm (Smart Ant-Dynamic RWA, SA-DRWA) to solve the RWA problem in optical networks. The link idle ratio is introduced into the ant transition probability as new constraint. At the same time, the random disturbance is added to prevent premature convergence into local optimal path within the less iteration. Theoretical analysis shows that the SA-DRWA algorithm can effectively achieve load balancing in optical network. Simulation for different topologies shows that the SA-DRWA algorithm has both improvements of blocking probability and resources utilization comparing with the traditional Dijkstra+FF algorithm. An optimal improvement scenario is obtained in regular mesh network with up to 0.23 improved for blocking probability and resources utilization.
2012, 34(3): 716-721.
doi: 10.3724/SP.J.1146.2011.00405
Abstract:
Target localization is one of the most challenging and important issues in Wireless Sensor Networks (WSNs). A multiple target localization method is proposed in WSNs. The multiple target localization issue is transformed into compressed sensing issue by designing iteration backtracking algorithm using multi-resolution analysis idea. The achievement of this algorithm is to save the energy of WSN nodes, by minimizing inter-node communication, in the result of which the lifetime of the WSN is prolonged, at the cost of increasing the computation complexity in the fusion center instead. Emulation results demonstrated that the localization performance is improved by more than 50% by the proposed algorithm.
Target localization is one of the most challenging and important issues in Wireless Sensor Networks (WSNs). A multiple target localization method is proposed in WSNs. The multiple target localization issue is transformed into compressed sensing issue by designing iteration backtracking algorithm using multi-resolution analysis idea. The achievement of this algorithm is to save the energy of WSN nodes, by minimizing inter-node communication, in the result of which the lifetime of the WSN is prolonged, at the cost of increasing the computation complexity in the fusion center instead. Emulation results demonstrated that the localization performance is improved by more than 50% by the proposed algorithm.
2012, 34(3): 722-727.
doi: 10.3724/SP.J.1146.2011.00753
Abstract:
Controlled synchronizibility of non-diffusively coupled complex networks is studied. After the analysis of eigenvalues distribution of outer coupling matrix of non-diffusively coupled complex networks, control law for synchronize non-diffusively coupled complex networks is given. It is found that the smaller the distance between largest and smallest eigenvalues of outer coupling matrix, the stronger the controlled synchronizibility of non-diffusively coupled complex networks. It is also found that the smaller the coupling strength, the greater the possibility that the network can achieve controlled synchronization. However under certain condition a smaller coupling strength requires a larger control gain value. Finally a small-world network formed by coupled Lorenz oscillators which needs to be controlled to synchronize onto chaotic orbit is given as an example to illustrate the effectiveness of the results.
Controlled synchronizibility of non-diffusively coupled complex networks is studied. After the analysis of eigenvalues distribution of outer coupling matrix of non-diffusively coupled complex networks, control law for synchronize non-diffusively coupled complex networks is given. It is found that the smaller the distance between largest and smallest eigenvalues of outer coupling matrix, the stronger the controlled synchronizibility of non-diffusively coupled complex networks. It is also found that the smaller the coupling strength, the greater the possibility that the network can achieve controlled synchronization. However under certain condition a smaller coupling strength requires a larger control gain value. Finally a small-world network formed by coupled Lorenz oscillators which needs to be controlled to synchronize onto chaotic orbit is given as an example to illustrate the effectiveness of the results.
2012, 34(3): 728-732.
doi: 10.3724/SP.J.1146.2011.00832
Abstract:
This paper presents the method of designing and fabricating 6-pole TeraHertz (THz) iris band-pass filter by adopting micromachining technology. Based on the existing process conditions, through analyzing the influence of process factors on electromagnetic property of filters, the process and designing parameters are mutually compromised to avoid sharp deterioration of device performance due to process factors so that the design can be optimized, and ultimately the integratable THz filter of small insertion loss and high reliability is obtained. By applying the processing methods including Inductively Coupled Plasma (ICP) etching, evaporating, electroplating gold and bonding, the single after-dicing micromachined filter with the volume of 24.0 mm 5.0 mm 1.66 mm is fabricated. The filter is fixed by applying tunable measurement clamp, and the power meter is adopted to test its power attenuation. Measurements results show that the center frequency of a 10.6% 3 dB fractional bandwidth is 141.5 GHz, and the power attenuation is less than 1 dB at 141.5 GHz, thus proving the effectiveness of the proposed processing method.
This paper presents the method of designing and fabricating 6-pole TeraHertz (THz) iris band-pass filter by adopting micromachining technology. Based on the existing process conditions, through analyzing the influence of process factors on electromagnetic property of filters, the process and designing parameters are mutually compromised to avoid sharp deterioration of device performance due to process factors so that the design can be optimized, and ultimately the integratable THz filter of small insertion loss and high reliability is obtained. By applying the processing methods including Inductively Coupled Plasma (ICP) etching, evaporating, electroplating gold and bonding, the single after-dicing micromachined filter with the volume of 24.0 mm 5.0 mm 1.66 mm is fabricated. The filter is fixed by applying tunable measurement clamp, and the power meter is adopted to test its power attenuation. Measurements results show that the center frequency of a 10.6% 3 dB fractional bandwidth is 141.5 GHz, and the power attenuation is less than 1 dB at 141.5 GHz, thus proving the effectiveness of the proposed processing method.
2012, 34(3): 733-738.
doi: 10.3724/SP.J.1146.2011.00654
Abstract:
To obtain the good performance for linearization, a digital predistortion algorithm is presented. Look-Up-Table (LUT) and memory-effect compensation techniques is combed, and the linear interpolation method is applied to the LUT. Compared to memory polynomial methods, this predistortion technique can reduce the complexity of calculation and obtain the good performance for linearization. Under power amplifiers memory polynomial model, the simulation results for Orthogonal Frequency Division Multiplexing (OFDM) wideband signal show the excellent linearization purpose for memory nonlinear power amplifier.
To obtain the good performance for linearization, a digital predistortion algorithm is presented. Look-Up-Table (LUT) and memory-effect compensation techniques is combed, and the linear interpolation method is applied to the LUT. Compared to memory polynomial methods, this predistortion technique can reduce the complexity of calculation and obtain the good performance for linearization. Under power amplifiers memory polynomial model, the simulation results for Orthogonal Frequency Division Multiplexing (OFDM) wideband signal show the excellent linearization purpose for memory nonlinear power amplifier.
2012, 34(3): 739-743.
doi: 10.3724/SP.J.1146.2011.00580
Abstract:
A large signal computation model of the beam-wave interaction in the External Interaction Klystron (EIK) is presented. This mode bases on the one dimensional electron disk model with considering the space charge force between the electrons. The equivalent circuit of the double-gap coupled cavity is used in the calculation of the interaction between electrons and gap field, and the electron conductivity in each gap of the coupled cavity is taken into account. The simulation code has been compiled based on this model and it shows good agreement with the simulation results by the Particle-In-Cell (PIC) code, MAGIC3D. This model can be used in the large signal computer simulation of the beam-wave interaction in the three or more gaps coupled cavity in the EIK.
A large signal computation model of the beam-wave interaction in the External Interaction Klystron (EIK) is presented. This mode bases on the one dimensional electron disk model with considering the space charge force between the electrons. The equivalent circuit of the double-gap coupled cavity is used in the calculation of the interaction between electrons and gap field, and the electron conductivity in each gap of the coupled cavity is taken into account. The simulation code has been compiled based on this model and it shows good agreement with the simulation results by the Particle-In-Cell (PIC) code, MAGIC3D. This model can be used in the large signal computer simulation of the beam-wave interaction in the three or more gaps coupled cavity in the EIK.
An Optimization Algorithm Based on Computation Structure of Correlation in Digital Lock-in Detection
2012, 34(3): 744-748.
doi: 10.3724/SP.J.1146.2011.00333
Abstract:
In order to give consideration to both of the precision and speed to improve the comprehensive properties of the detecting system, a new high-speed algorithm is proposed based on computation structure of correlation in digital lock-in detection and it is optimized by oversampling. The algorithm based on oversampling reduces the sampling frequency to four times of the frequency of an input signal. Thus, the digital lock-in algorithm can be implemented by the addition and subtraction operations of 4q points in q periods. This process eliminates almost all the multiplication operations compared with the traditional method. And the speed of the digital lock-in algorithm is greatly enhanced. Moreover, a correction factor is introduced to improve the error caused by down-sampling. The results of the experiments show that the new algorithm not only has high precision of oversampling and digital lock-in detection for signal with low signal-to-noise ratio, but also has high speed, which makes it possible for a general microprocessor to implement weak signal detection by the new algorithm in real time.
In order to give consideration to both of the precision and speed to improve the comprehensive properties of the detecting system, a new high-speed algorithm is proposed based on computation structure of correlation in digital lock-in detection and it is optimized by oversampling. The algorithm based on oversampling reduces the sampling frequency to four times of the frequency of an input signal. Thus, the digital lock-in algorithm can be implemented by the addition and subtraction operations of 4q points in q periods. This process eliminates almost all the multiplication operations compared with the traditional method. And the speed of the digital lock-in algorithm is greatly enhanced. Moreover, a correction factor is introduced to improve the error caused by down-sampling. The results of the experiments show that the new algorithm not only has high precision of oversampling and digital lock-in detection for signal with low signal-to-noise ratio, but also has high speed, which makes it possible for a general microprocessor to implement weak signal detection by the new algorithm in real time.
2012, 34(3): 749-753.
doi: 10.3724/SP.J.1146.2011.00417
Abstract:
The Direction-Of-Arrival (DOA) tracking of coherent signals is researched based on polarization vector- sensor array. The decorrelative performance of a Single Snapshot Polarization Smoothing Algorithm (SSPSA), which is available for the fast estimation of DOA in multipath environment, is analyzed. The algorithm can improve the effective array aperture. Then a subspace tracking algorithm of coherent sources is presented based on iteration. The iterated algorithm has higher tracking speed and requires neither the eigen-decomposition into signal/noise subspace nor the inverse operation of matrix. Simulation results verify the effectiveness of the proposed algorithm.
The Direction-Of-Arrival (DOA) tracking of coherent signals is researched based on polarization vector- sensor array. The decorrelative performance of a Single Snapshot Polarization Smoothing Algorithm (SSPSA), which is available for the fast estimation of DOA in multipath environment, is analyzed. The algorithm can improve the effective array aperture. Then a subspace tracking algorithm of coherent sources is presented based on iteration. The iterated algorithm has higher tracking speed and requires neither the eigen-decomposition into signal/noise subspace nor the inverse operation of matrix. Simulation results verify the effectiveness of the proposed algorithm.
2012, 34(3): 754-757.
doi: 10.3724/SP.J.1146.2011.00621
Abstract:
For high-speed weak targets, a fast method of target parameter estimation is presented in this paper. After analyzing the factors which influence high-speed target,s range migration, it pointed out that the range migration exhibits a straight line with accumulation time. Then, using the feature of Hough Transform (HT) to detect straight line, the relationship between the parameters of HT and initial distance and initial radial velocity are analyzed and an improved method is proposed based on Parallel HT (PHT) which can improve operation speed and meet the needs of real-time processing by taking advantage of its parallel principle. Finally, the algorithm flowchart is given and the validity of this algorithm is verified with the simulation results.
For high-speed weak targets, a fast method of target parameter estimation is presented in this paper. After analyzing the factors which influence high-speed target,s range migration, it pointed out that the range migration exhibits a straight line with accumulation time. Then, using the feature of Hough Transform (HT) to detect straight line, the relationship between the parameters of HT and initial distance and initial radial velocity are analyzed and an improved method is proposed based on Parallel HT (PHT) which can improve operation speed and meet the needs of real-time processing by taking advantage of its parallel principle. Finally, the algorithm flowchart is given and the validity of this algorithm is verified with the simulation results.
2012, 34(3): 758-762.
doi: 10.3724/SP.J.1146.2011.00725
Abstract:
Interval vector is a crucial parameter of primitive-LFSR, but the reserch about the interval vector is little. Based on interval vector, the correctness of decimation conjecture is first proved when m=2, and then some properties about the decimation sequence are given. Finally some new methods are given to construct Z- primitive-LFSR and primitive-LFSR sequences.
Interval vector is a crucial parameter of primitive-LFSR, but the reserch about the interval vector is little. Based on interval vector, the correctness of decimation conjecture is first proved when m=2, and then some properties about the decimation sequence are given. Finally some new methods are given to construct Z- primitive-LFSR and primitive-LFSR sequences.
