Email alert
2010 Vol. 32, No. 12
Display Method:
2010, 32(12): 2795-2801.
doi: 10.3724/SP.J.1146.2010.00158
Abstract:
In order to select a suitable time interval to reconstruct a well-focused ISAR image for maneuvering target in long recorded data, a novel method based on Phase Linear Degree (PLD) is proposed in this paper. Firstly, the scatterer phase property is deduced in detail and the imaging time condition is given. Secondly, two prominent range cells are selected based on the Average Range Profile (ARP), and the radial phase is removed by the two range cells conjugate multiplication. Then the rotation phase is curve fitting based on the unwrapping phase. Accordingly, the phase linear degree of the rotation phase is evaluated at the different time interval, and the interval with the highest phase linear degree is selected as the imaging time. Finally, the technique effectiveness is tested by simulated and real data.
In order to select a suitable time interval to reconstruct a well-focused ISAR image for maneuvering target in long recorded data, a novel method based on Phase Linear Degree (PLD) is proposed in this paper. Firstly, the scatterer phase property is deduced in detail and the imaging time condition is given. Secondly, two prominent range cells are selected based on the Average Range Profile (ARP), and the radial phase is removed by the two range cells conjugate multiplication. Then the rotation phase is curve fitting based on the unwrapping phase. Accordingly, the phase linear degree of the rotation phase is evaluated at the different time interval, and the interval with the highest phase linear degree is selected as the imaging time. Finally, the technique effectiveness is tested by simulated and real data.
2010, 32(12): 2802-2807.
doi: 10.3724/SP.J.1146.2010.00003
Abstract:
In this paper, a novel Polar Format Algorithm (PFA) is presented for Circular Synthetic Aperture Radar (CSAR) imaging. Based on the signal model analysis, it is finded that the shape of the spectrum in the wavenumber domain of a point target in CSAR is a three-dimensional curved surface. To avoid the interpolation along the third dimension and also ensure accuracy, this algorithm obtains two-dimensional images at equally spaced heights to form the final three-dimensional image, and this is done by multiplying reference function of height to project the spectrum to two-dimensional planes. By two steps of phase compensation, this algorithm also corrects the high terms of the Range Cell Migration (RCM) to enable full image quality over much larger scenes without image defocus. Finally, simulation results validate the feasibility of the proposed approach.
In this paper, a novel Polar Format Algorithm (PFA) is presented for Circular Synthetic Aperture Radar (CSAR) imaging. Based on the signal model analysis, it is finded that the shape of the spectrum in the wavenumber domain of a point target in CSAR is a three-dimensional curved surface. To avoid the interpolation along the third dimension and also ensure accuracy, this algorithm obtains two-dimensional images at equally spaced heights to form the final three-dimensional image, and this is done by multiplying reference function of height to project the spectrum to two-dimensional planes. By two steps of phase compensation, this algorithm also corrects the high terms of the Range Cell Migration (RCM) to enable full image quality over much larger scenes without image defocus. Finally, simulation results validate the feasibility of the proposed approach.
2010, 32(12): 2808-2813.
doi: 10.3724/SP.J.1146.2010.00669
Abstract:
The traditional sparse aperture methods are not suitable for signal processing of Inverse Synthetic Aperture Imaging Ladar (ISAIL) for the following reasons. One is the lack of echoes caused by the noncooperation characteristic of the target, the other is the low Signal-to-Noise Ratio (SNR) caused by the influence of atmosphere attenuation and natural background light. For this reason, a novel sparse aperture imaging method is proposed in this paper which combines the Compressed Sensing (CS) with the weighted matrix. Through the preprocessing of CS based imaging, the supporting field of the target can be obtained, with which the weighted matrix can be constructed. Then the cost function is optimized using the weighted matrix. Finally, with this new cost function, the high resolution image can be achieved using the incomplete original echoes from the sparse aperture ISAIL. This new method is robust in front of strong noise. The feasibility and effectiveness of the method are validated by the measured data of the indoor ISAIL system.
The traditional sparse aperture methods are not suitable for signal processing of Inverse Synthetic Aperture Imaging Ladar (ISAIL) for the following reasons. One is the lack of echoes caused by the noncooperation characteristic of the target, the other is the low Signal-to-Noise Ratio (SNR) caused by the influence of atmosphere attenuation and natural background light. For this reason, a novel sparse aperture imaging method is proposed in this paper which combines the Compressed Sensing (CS) with the weighted matrix. Through the preprocessing of CS based imaging, the supporting field of the target can be obtained, with which the weighted matrix can be constructed. Then the cost function is optimized using the weighted matrix. Finally, with this new cost function, the high resolution image can be achieved using the incomplete original echoes from the sparse aperture ISAIL. This new method is robust in front of strong noise. The feasibility and effectiveness of the method are validated by the measured data of the indoor ISAIL system.
2010, 32(12): 2814-2819.
doi: 10.3724/SP.J.1146.2010.00091
Abstract:
A three-stage vegetation height inversion method with compact polarimetric SAR interferometric data is proposed. The coherence optimization method applied to full PolInSAR data processing is introduced into compact PolInSAR processing. The optimum phase is used to get the initial estimation of volume decorrelation and ground phase. Volume decorrelation estimation is updated with coherence boundary extraction method. Simulated L-band full PolInSAR data provided by ESA is used to construct compact PolInSAR data, and the generated data is used for vegetation height inversion. Inversion results confirm the validity of the proposed method.
A three-stage vegetation height inversion method with compact polarimetric SAR interferometric data is proposed. The coherence optimization method applied to full PolInSAR data processing is introduced into compact PolInSAR processing. The optimum phase is used to get the initial estimation of volume decorrelation and ground phase. Volume decorrelation estimation is updated with coherence boundary extraction method. Simulated L-band full PolInSAR data provided by ESA is used to construct compact PolInSAR data, and the generated data is used for vegetation height inversion. Inversion results confirm the validity of the proposed method.
2010, 32(12): 2820-2824.
doi: 10.3724/SP.J.1146.2010.00018
Abstract:
A novel method of multitarget localization in L shaped MIMO bistatic radar is proposed. Based on four angles information involved in L shaped receiving array from target to transmiter and receiver, the joint estimation algorithm for four angles of multitarget can be obtained by the following steps: uncorrelating the array data of received echoes first, then constructing estimation matrix based on the DOA matrix method, finally using the given relationship between the eigenvalue and the estimated parameters. As a result, the multitarget localization in three dimensions is achieved in the bistatic radar. The proposed algorithm does not refer to multi-dimensional nonlinear peak search, and need only once eigenvalue decomposition, so that the computed load of the algorithm is low, and the estimated parameters of the targets can be paired automatically. The correctness and effectiveness of the proposed method are verified with the computer simulation.
A novel method of multitarget localization in L shaped MIMO bistatic radar is proposed. Based on four angles information involved in L shaped receiving array from target to transmiter and receiver, the joint estimation algorithm for four angles of multitarget can be obtained by the following steps: uncorrelating the array data of received echoes first, then constructing estimation matrix based on the DOA matrix method, finally using the given relationship between the eigenvalue and the estimated parameters. As a result, the multitarget localization in three dimensions is achieved in the bistatic radar. The proposed algorithm does not refer to multi-dimensional nonlinear peak search, and need only once eigenvalue decomposition, so that the computed load of the algorithm is low, and the estimated parameters of the targets can be paired automatically. The correctness and effectiveness of the proposed method are verified with the computer simulation.
2010, 32(12): 2825-2829.
doi: 10.3724/SP.J.1146.2010.00006
Abstract:
In this paper, a method for dynamic measurement of non-rigid baseline of dual-antenna InSAR is presented and discussed. In this scheme, the position and orientation of the antenna is obtained by using a CCD camera and a laser range finder together. In this paper, the research of principle and experiments are carried out. In the experiments, the movement of the antenna is created by some equipments, the tracks of which can be controlled accurately. The results of experiments show that the errors of position and attitude are less than 0.2mm and 70 arcseconds. The method is applicable to determine the baseline of dual-antenna dynamically.
In this paper, a method for dynamic measurement of non-rigid baseline of dual-antenna InSAR is presented and discussed. In this scheme, the position and orientation of the antenna is obtained by using a CCD camera and a laser range finder together. In this paper, the research of principle and experiments are carried out. In the experiments, the movement of the antenna is created by some equipments, the tracks of which can be controlled accurately. The results of experiments show that the errors of position and attitude are less than 0.2mm and 70 arcseconds. The method is applicable to determine the baseline of dual-antenna dynamically.
2010, 32(12): 2830-2835.
doi: 10.3724/SP.J.1146.2010.00074
Abstract:
Based on range Digital BeamForming (DBF) on receive in DBF-SAR system, a novel method is proposed to counter deceptive jamming. The method makes full use of the difference between the direction of deceptive jamming return and the direction of ground target return. Since DBF-SAR with range multiple apertures can improve the system freedoms, the direction of deceptive jamming are estimated through spectrum estimation algorithm. Then according to the estimated direction, the weighted coefficient is calculated to form the equivalent range antenna pattern. Since range antenna pattern has the nulls in the directions of interferences, deceptive jamming is suppressed through space filter. Simulation results show that range DBF-SAR system can counter deceptive jamming effectively and enhance imaging quality.
Based on range Digital BeamForming (DBF) on receive in DBF-SAR system, a novel method is proposed to counter deceptive jamming. The method makes full use of the difference between the direction of deceptive jamming return and the direction of ground target return. Since DBF-SAR with range multiple apertures can improve the system freedoms, the direction of deceptive jamming are estimated through spectrum estimation algorithm. Then according to the estimated direction, the weighted coefficient is calculated to form the equivalent range antenna pattern. Since range antenna pattern has the nulls in the directions of interferences, deceptive jamming is suppressed through space filter. Simulation results show that range DBF-SAR system can counter deceptive jamming effectively and enhance imaging quality.
2010, 32(12): 2836-2841.
doi: 10.3724/SP.J.1146.2009.01657
Abstract:
SAR image simulation of complex target is very important to radar target recognition and SAR image interpretation. In this paper, the raw echo is simulated by assembling scattering coefficients of the target, under the different aspect and frequency. These coefficients are obtained by applying shooting and bouncing rays technology onto the targets 3D model. Via improving the traditional assumption of incident plane wave as the spherical wave, the algorithm is extended to be suitable for the near-field condition, as well as the far-field. Standard target experiment validate the algorithm can simulate full polarimetry, multiple aspect and wide bandwidth SAR echo. The simulated images of complex targets are very similar to the real data.
SAR image simulation of complex target is very important to radar target recognition and SAR image interpretation. In this paper, the raw echo is simulated by assembling scattering coefficients of the target, under the different aspect and frequency. These coefficients are obtained by applying shooting and bouncing rays technology onto the targets 3D model. Via improving the traditional assumption of incident plane wave as the spherical wave, the algorithm is extended to be suitable for the near-field condition, as well as the far-field. Standard target experiment validate the algorithm can simulate full polarimetry, multiple aspect and wide bandwidth SAR echo. The simulated images of complex targets are very similar to the real data.
2010, 32(12): 2842-2847.
doi: 10.3724/SP.J.1146.2009.01466
Abstract:
Lacking of GCPs (Ground Control Points) in hard survey area is one of the most difficult problems of geometric rectification. This paper proposes a new method of large SAR image-block adjustment based on F.leberl model. The parameters of the images and ground coordinate of pass points can be calculated with the use of alternative and iterative algorithm. The result shows that this method could get steady result with few GCPs and the error of tie points can also be decreased effectively, the accuracy reaches to the lever of single-image-calibration method.
Lacking of GCPs (Ground Control Points) in hard survey area is one of the most difficult problems of geometric rectification. This paper proposes a new method of large SAR image-block adjustment based on F.leberl model. The parameters of the images and ground coordinate of pass points can be calculated with the use of alternative and iterative algorithm. The result shows that this method could get steady result with few GCPs and the error of tie points can also be decreased effectively, the accuracy reaches to the lever of single-image-calibration method.
2010, 32(12): 2848-2853.
doi: 10.3724/SP.J.1146.2010.00128
Abstract:
Classification of track vehicle and wheel vehicle by using low-resolution radar echo is a challenging problem in radar ground target automatic classification community. This paper establishes an echo signal model for track vehicle and wheel vehicle based on micro-doppler effect. According to the echo signal model, a new feature extraction method is proposed by using the CLEAN algorithm. The feature reveals the energy distribution of the target doppler spectrum. The measured data results based on Relevance Vector Machine (RVM) and Support Vector Machine (SVM) show the proposed feature can not only achieve good classification performance, but also be robust to the target velocity.
Classification of track vehicle and wheel vehicle by using low-resolution radar echo is a challenging problem in radar ground target automatic classification community. This paper establishes an echo signal model for track vehicle and wheel vehicle based on micro-doppler effect. According to the echo signal model, a new feature extraction method is proposed by using the CLEAN algorithm. The feature reveals the energy distribution of the target doppler spectrum. The measured data results based on Relevance Vector Machine (RVM) and Support Vector Machine (SVM) show the proposed feature can not only achieve good classification performance, but also be robust to the target velocity.
2010, 32(12): 2854-2860.
doi: 10.3724/SP.J.1146.2009.01584
Abstract:
A novel algorithm using multilayer reconstructed-autocorrelation is proposed for range tracking with short time signal in powerful noise on a short-range millimeter wave radar. This algorithm can solve the problem of weak anti-jamming ability and low accuracy in range tracking brought by above conditions, which reconstructs a data sequence by rearrangement autocorrelation of the observed data vector and then range information can be got by doing spectrum analysis. By using of this algorithm, straining noise and analyzing the spectrum are organically combined. Not only are the estimate precision is increased by exponential but also the frequency stability is greatly improved. The simulation results test and verify the theoretical analysis result and the algorithm effectiveness.
A novel algorithm using multilayer reconstructed-autocorrelation is proposed for range tracking with short time signal in powerful noise on a short-range millimeter wave radar. This algorithm can solve the problem of weak anti-jamming ability and low accuracy in range tracking brought by above conditions, which reconstructs a data sequence by rearrangement autocorrelation of the observed data vector and then range information can be got by doing spectrum analysis. By using of this algorithm, straining noise and analyzing the spectrum are organically combined. Not only are the estimate precision is increased by exponential but also the frequency stability is greatly improved. The simulation results test and verify the theoretical analysis result and the algorithm effectiveness.
2010, 32(12): 2861-2867.
doi: 10.3724/SP.J.1146.2010.00452
Abstract:
Due to the high velocity, the complicated flight path and the need of real time processing during the diving maneuver, a new algorithm for the missile-borne SAR imaging is challenging. Thus, a new CZT (Chirp-Z Transform) imaging algorithm based on the method of series reversion is proposed in this paper. Firstly, the point target range function is established, and then it is expanded using the Taylor series. Secondly, the high order approximate expression of the tow-dimensional spectrum is obtained via the method of series reversion, based on which the mended CZT imaging algorithm is derived and the computation load is analyzed then. The algorithm is able to correct the spatially varying range-migration. Because all the operations are composed of the fast Fourier transform and the phase multiplication, the algorithm is efficient. The result of the point targets simulation shows that high resolution image is obtained by using the algorithm proposed with the diving maneuver in entire aperture.
Due to the high velocity, the complicated flight path and the need of real time processing during the diving maneuver, a new algorithm for the missile-borne SAR imaging is challenging. Thus, a new CZT (Chirp-Z Transform) imaging algorithm based on the method of series reversion is proposed in this paper. Firstly, the point target range function is established, and then it is expanded using the Taylor series. Secondly, the high order approximate expression of the tow-dimensional spectrum is obtained via the method of series reversion, based on which the mended CZT imaging algorithm is derived and the computation load is analyzed then. The algorithm is able to correct the spatially varying range-migration. Because all the operations are composed of the fast Fourier transform and the phase multiplication, the algorithm is efficient. The result of the point targets simulation shows that high resolution image is obtained by using the algorithm proposed with the diving maneuver in entire aperture.
2010, 32(12): 2868-2872.
doi: 10.3724/SP.J.1146.2010.00026
Abstract:
In this paper, multicarrier radar signal models of stationary and moving targets are analyzed. Based on the Doppler frequency difference of different subcarrier, a new method is proposed to estimate the velocity of the target with one and multiple scattering points using single echo. This method measures the velocity from the data produced by multiple auto correlation channels using nonlinear least squares criteria. The correspondence processing scheme is given and the complexity of this algorithm is analyzed. Simulation results show that this method is feasible and effective.
In this paper, multicarrier radar signal models of stationary and moving targets are analyzed. Based on the Doppler frequency difference of different subcarrier, a new method is proposed to estimate the velocity of the target with one and multiple scattering points using single echo. This method measures the velocity from the data produced by multiple auto correlation channels using nonlinear least squares criteria. The correspondence processing scheme is given and the complexity of this algorithm is analyzed. Simulation results show that this method is feasible and effective.
2010, 32(12): 2873-2878.
doi: 10.3724/SP.J.1146.2010.00079
Abstract:
Harbor detection is an important aspect for remote sensing ocean application research. Fast and accurate harbor detection can greatly improve the ability of automatic interpretation for remote sensing image. By having a detailed research on harbor object disposal, this paper establishes harbor feature model, and proposes a harbor detection method of remote sensing images based on feature. First, extraction and discrimination of harbor jetty are performed in the instruction of geometric features and topologic features belonging to harbor. Then, the jetty key points are selected and the coastline closure between key points is calculated according to the contextual and geometric features of harbor. Finally, harbor detection is finished in the guidance of closure principle. Compared with an existing harbor detection algorithm, the experiments show that novel method has a whole harbor detection, high detection ratio and fast operation.
Harbor detection is an important aspect for remote sensing ocean application research. Fast and accurate harbor detection can greatly improve the ability of automatic interpretation for remote sensing image. By having a detailed research on harbor object disposal, this paper establishes harbor feature model, and proposes a harbor detection method of remote sensing images based on feature. First, extraction and discrimination of harbor jetty are performed in the instruction of geometric features and topologic features belonging to harbor. Then, the jetty key points are selected and the coastline closure between key points is calculated according to the contextual and geometric features of harbor. Finally, harbor detection is finished in the guidance of closure principle. Compared with an existing harbor detection algorithm, the experiments show that novel method has a whole harbor detection, high detection ratio and fast operation.
2010, 32(12): 2879-2883.
doi: 10.3724/SP.J.1146.2010.00143
Abstract:
This paper proposes a new high accurate azimuth and elevation angle estimation algorithm using DOA matrix method to avoid the ambiguous of extended aperture. The algorithm uses 4array elements to form two-parallel-shape array geometry, high-variance unambiguous direction cosine estimates and low-variance cyclically ambiguous direction cosine estimates are obtained from the different sensors space. The key idea underlying of this algorithm is to use the high-variance unambiguous direction cosine estimates to resolve the low-variance cyclically ambiguous direction cosine estimates to extract azimuth-elevation angle estimates. The proposed algorithm requires no pair matching procedure and 2D iterative searching, and thereby, offering high estimation precision with low computational complexity. The simulation results indicate the performance of proposed algorithm.
This paper proposes a new high accurate azimuth and elevation angle estimation algorithm using DOA matrix method to avoid the ambiguous of extended aperture. The algorithm uses 4array elements to form two-parallel-shape array geometry, high-variance unambiguous direction cosine estimates and low-variance cyclically ambiguous direction cosine estimates are obtained from the different sensors space. The key idea underlying of this algorithm is to use the high-variance unambiguous direction cosine estimates to resolve the low-variance cyclically ambiguous direction cosine estimates to extract azimuth-elevation angle estimates. The proposed algorithm requires no pair matching procedure and 2D iterative searching, and thereby, offering high estimation precision with low computational complexity. The simulation results indicate the performance of proposed algorithm.
2010, 32(12): 2884-2889.
doi: 10.3724/SP.J.1146.2010.00129
Abstract:
The initial state estimation for boost phase object is an important issue to estimate the missiles tactical parameters, as well as the precondition for the surveillance system to perform the real-time tracking. In the condition of the space-based observations, the initial state estimation is the solution of the nonlinear LSE problem. Firstly, the 8-state gravity turn model for the boost phase object is established, and then the targets kinematic parameters are approximated by using the linear constant acceleration derivative model, and the nonlinear measurements are pseudo linearized. Therefore, the nonlinear LSE problem is translated into the linear LSE problem after the aforementioned steps . Specially, in this paper, the explicit derivation of the pseudo linearized space-based observation and its statistical moments are deduced, and its application conditions are illuminated. Finally, the advantage of the proposed method is illustrated in the aspects of precision and efficiency by comparing with the CRLB and classical Gauss-Newton iteration algorithm.
The initial state estimation for boost phase object is an important issue to estimate the missiles tactical parameters, as well as the precondition for the surveillance system to perform the real-time tracking. In the condition of the space-based observations, the initial state estimation is the solution of the nonlinear LSE problem. Firstly, the 8-state gravity turn model for the boost phase object is established, and then the targets kinematic parameters are approximated by using the linear constant acceleration derivative model, and the nonlinear measurements are pseudo linearized. Therefore, the nonlinear LSE problem is translated into the linear LSE problem after the aforementioned steps . Specially, in this paper, the explicit derivation of the pseudo linearized space-based observation and its statistical moments are deduced, and its application conditions are illuminated. Finally, the advantage of the proposed method is illustrated in the aspects of precision and efficiency by comparing with the CRLB and classical Gauss-Newton iteration algorithm.
2010, 32(12): 2890-2894.
doi: 10.3724/SP.J.1146.2009.01500
Abstract:
Considering the issue that two-dimensional beamforming usually takes a great deal of sampling data and has very high computation complexity by using Minimum Variance Distortionless Response (MVDR) beamformer, a Two-Dimensional Adaptive Beamforming (TDAB) algorithm based on correlation matrix is proposed. High-dimensional weight vector is written as the Kronecker product of two low-dimensional weight vectors. By utilizing a bi-iterative algorithm, two low-dimensional weight vectors can be solved on the basis of correlation matrix, which decrease the computational complexity and the number of training samples for correlation matrix estimates．Simulations results demonstrate that TDAB can converge very well and achieve better performance of interference suppression in the presence of short data records.
Considering the issue that two-dimensional beamforming usually takes a great deal of sampling data and has very high computation complexity by using Minimum Variance Distortionless Response (MVDR) beamformer, a Two-Dimensional Adaptive Beamforming (TDAB) algorithm based on correlation matrix is proposed. High-dimensional weight vector is written as the Kronecker product of two low-dimensional weight vectors. By utilizing a bi-iterative algorithm, two low-dimensional weight vectors can be solved on the basis of correlation matrix, which decrease the computational complexity and the number of training samples for correlation matrix estimates．Simulations results demonstrate that TDAB can converge very well and achieve better performance of interference suppression in the presence of short data records.
2010, 32(12): 2895-2900.
doi: 10.3724/SP.J.1146.2009.00021
Abstract:
For small sample size problem in high-dimensional space, conventional classifiers with reject option based on statistical model could not construct appropriate covering decision boundary on data distribution. In this case, a novel adaptive Minimum Spanning Tree (MST) covering model based classifier with reject option is proposed in this paper according to the data distribution in high-dimensional space. The algorithm describes the target class using MST with the assumption that the edges of the graph are also basic elements of the classifier which offers additional virtual training data for a better coverage. By this model, similar samples from the same class are divided into a connected geometric coverage area, and similar samples from different classes are divided into different geometric coverage areas. Furthermore, in order to reduce the degradation of the rejection performance due to the existence of unreasonable additional virtual training data, an adjustable coverage radius strategy is presented in coverage construction. Then the test pattern of non-training classes could be rejected by the coverage decision boundary, and if a pattern is accepted in the cross coverage area, the recognition result is decided by the data fields model. Experiments show that the method is valid and efficient.
For small sample size problem in high-dimensional space, conventional classifiers with reject option based on statistical model could not construct appropriate covering decision boundary on data distribution. In this case, a novel adaptive Minimum Spanning Tree (MST) covering model based classifier with reject option is proposed in this paper according to the data distribution in high-dimensional space. The algorithm describes the target class using MST with the assumption that the edges of the graph are also basic elements of the classifier which offers additional virtual training data for a better coverage. By this model, similar samples from the same class are divided into a connected geometric coverage area, and similar samples from different classes are divided into different geometric coverage areas. Furthermore, in order to reduce the degradation of the rejection performance due to the existence of unreasonable additional virtual training data, an adjustable coverage radius strategy is presented in coverage construction. Then the test pattern of non-training classes could be rejected by the coverage decision boundary, and if a pattern is accepted in the cross coverage area, the recognition result is decided by the data fields model. Experiments show that the method is valid and efficient.
2010, 32(12): 2901-2906.
doi: 10.3724/SP.J.1146.2010.00111
Abstract:
This paper proposes a parallel fast traffic lane detection system. The system consists of a 3232 Processing Elements (PE) array and a dual RISC core subsystem. The PE array performs pixel-parallel image preprocessing and outputs edge features, the dual RISC core subsystem performs two lanes parameters detection in parallel based on edge features. In this way, every step in the detection process is in parallel and the detection rate is rapidly increased. The system is implemented with FPGA. The experiment shows that it has good robustness and can reach up to 50 fps. This meets the demand of real-time for lane departure warning system and makes an important sense for practical application.
This paper proposes a parallel fast traffic lane detection system. The system consists of a 3232 Processing Elements (PE) array and a dual RISC core subsystem. The PE array performs pixel-parallel image preprocessing and outputs edge features, the dual RISC core subsystem performs two lanes parameters detection in parallel based on edge features. In this way, every step in the detection process is in parallel and the detection rate is rapidly increased. The system is implemented with FPGA. The experiment shows that it has good robustness and can reach up to 50 fps. This meets the demand of real-time for lane departure warning system and makes an important sense for practical application.
2010, 32(12): 2907-2912.
doi: 10.3724/SP.J.1146.2009.01630
Abstract:
Because trandition meteorological sensors range is limited and hard to be erected, video detections require good external conditon and are easliy effected by nosie. A new method of detection of visibility is proposed based on camera self-calibration. And a system of visibility detection is constructed. The camera self-calibration, wavelet transform, video contrast detecting model, curve fitting is used to overcome disadvantages of tradition detections algorithms at distance, targets, noise and stability. Now the system has been used on 1/5 of Jiangsus high-ways. The errors of result conform to national and foreign standards.
Because trandition meteorological sensors range is limited and hard to be erected, video detections require good external conditon and are easliy effected by nosie. A new method of detection of visibility is proposed based on camera self-calibration. And a system of visibility detection is constructed. The camera self-calibration, wavelet transform, video contrast detecting model, curve fitting is used to overcome disadvantages of tradition detections algorithms at distance, targets, noise and stability. Now the system has been used on 1/5 of Jiangsus high-ways. The errors of result conform to national and foreign standards.
2010, 32(12): 2913-2918.
doi: 10.3724/SP.J.1146.2010.00035
Abstract:
The context data missing is an inevitable problem of context information processing mechanism, the imputation technique of missing data also is a research hotspot in data mining. However, the existing imputation technique of missing data is not suitable for the flow data form of context information that does not make full use of data relevance between every collecting sensor. Moreover, that does not take spatial-temporal relationship into account. In order to conquer the shortcomings and deficiencies of the existing imputation technique of missing data, this paper proposes an imputation technique for context data missing based on Spatial-Temporal and Association Rule Mining (STARM) to perform spatiality and time series analysis on sensor data, and generate strong association rules to interpolate missing data. Finally, the simulation experiment verifies the rationality and efficiency of STARM through temperature sensor data acquisition. Experiments show that the algorithm is of high accuracy for the imputation of context data missing, such as Simple Linear Regression (SLR) algorithm and the EM algorithm. In addition, which is smaller time and space overhead and can guarantee Quality of Service (QoS) of real-time applications.
The context data missing is an inevitable problem of context information processing mechanism, the imputation technique of missing data also is a research hotspot in data mining. However, the existing imputation technique of missing data is not suitable for the flow data form of context information that does not make full use of data relevance between every collecting sensor. Moreover, that does not take spatial-temporal relationship into account. In order to conquer the shortcomings and deficiencies of the existing imputation technique of missing data, this paper proposes an imputation technique for context data missing based on Spatial-Temporal and Association Rule Mining (STARM) to perform spatiality and time series analysis on sensor data, and generate strong association rules to interpolate missing data. Finally, the simulation experiment verifies the rationality and efficiency of STARM through temperature sensor data acquisition. Experiments show that the algorithm is of high accuracy for the imputation of context data missing, such as Simple Linear Regression (SLR) algorithm and the EM algorithm. In addition, which is smaller time and space overhead and can guarantee Quality of Service (QoS) of real-time applications.
2010, 32(12): 2919-2924.
doi: 10.3724/SP.J.1146.2010.00056
Abstract:
In this paper, a new discrete correlation signal is defined, which can be called as First Kind of Perfect Punctured Binary Sequence Pair (FK-PPBSP). The properties of FK-PPBSP are studied. The mutual construction between FK-PPBSP and pseudorandom Hadamard matrices is proved. A FK-PPBSP based time-domain channel estimation algorithm for orthogonal frequency division multiplexing (OFDM) is proposed. Simulation results show that compared with m sequence method, a better system performance with lower overhead and lower computational complexity is obtained with the proposed method.
In this paper, a new discrete correlation signal is defined, which can be called as First Kind of Perfect Punctured Binary Sequence Pair (FK-PPBSP). The properties of FK-PPBSP are studied. The mutual construction between FK-PPBSP and pseudorandom Hadamard matrices is proved. A FK-PPBSP based time-domain channel estimation algorithm for orthogonal frequency division multiplexing (OFDM) is proposed. Simulation results show that compared with m sequence method, a better system performance with lower overhead and lower computational complexity is obtained with the proposed method.
2010, 32(12): 2925-2929.
doi: 10.3724/SP.J.1146.2010.00030
Abstract:
For the Bit-Search-Generators (BSG) with a low weight feedback polynomial, a fast key recovery algorithm is presented using the ideas of the guess-and-determine attack. A candidate differential sequence is recovered firstly from the intercepted keystream sequence based on the differential construction of the BSG sequence. Then the feedback polynomial is used to check the candidate differential sequence, which will reduce the number of the linear equation systems of L dimensions thus to reduce significantly the complexity of the algorithm. Theoretical analysis and simulation experiment results show that, when the weight of the feedback polynomial is less than 10, the complexity of the attack is noticeably better than that of the existing methods. Specially, the attack complexity can be significantly reduced from the best known attack complexity O(L320.5L) to O(L20.5L) when the weight is 3.
For the Bit-Search-Generators (BSG) with a low weight feedback polynomial, a fast key recovery algorithm is presented using the ideas of the guess-and-determine attack. A candidate differential sequence is recovered firstly from the intercepted keystream sequence based on the differential construction of the BSG sequence. Then the feedback polynomial is used to check the candidate differential sequence, which will reduce the number of the linear equation systems of L dimensions thus to reduce significantly the complexity of the algorithm. Theoretical analysis and simulation experiment results show that, when the weight of the feedback polynomial is less than 10, the complexity of the attack is noticeably better than that of the existing methods. Specially, the attack complexity can be significantly reduced from the best known attack complexity O(L320.5L) to O(L20.5L) when the weight is 3.
2010, 32(12): 2930-2934.
doi: 10.3724/SP.J.1146.2010.00167
Abstract:
The braid group is a new candidate platform for constructing quantum attack-resistant cryptographic protocols. A new signature scheme is proposed based on the difficulty of the root extraction problem over braid groups, which can resist existential forgery against the adaptively chosen-message attack under the random oracle model. Compared with the Conjugacy Signature Scheme (CSS) and the Enhanced Conjugacy Signature Scheme (ECSS), the verification phase of the proposed scheme desires less computation because it does not have to determine whether two braids are conjugate. The signature is composed of an integer and a braid, which is much shorter than that of CSS and almost same as that of ECSS.
The braid group is a new candidate platform for constructing quantum attack-resistant cryptographic protocols. A new signature scheme is proposed based on the difficulty of the root extraction problem over braid groups, which can resist existential forgery against the adaptively chosen-message attack under the random oracle model. Compared with the Conjugacy Signature Scheme (CSS) and the Enhanced Conjugacy Signature Scheme (ECSS), the verification phase of the proposed scheme desires less computation because it does not have to determine whether two braids are conjugate. The signature is composed of an integer and a braid, which is much shorter than that of CSS and almost same as that of ECSS.
2010, 32(12): 2935-2941.
doi: 10.3724/SP.J.1146.2009.01152
Abstract:
In this paper, a transmit power allocation method is proposed for distributed transmit antenna Alamouti Space-Time Block Coded Orthogonal Frequency Division Multiplexing (STBC-OFDM) systems based on Cholesky-decision-feedback detector in multipath Rayleigh fading channels with multiple Residual Frequency Offsets (RFOs). The method is designed by minimizing the lower bound of the average Bit Error Ratio (BER) so as to reduce the influence of RFOs. Theoretical analysis of the lower bound on the average BER performance is presented and the closed-form expression of the optimal power allocation factor is derived. Simulation results show that in the scenario where the performance is dominated by the RFOs, the proposed method provides increasing performance gain compared with the conventional power allocation method ignoring the existence of RFOs as the ratio between the average channel gains from the two transmit antennas to receiver increasing.
In this paper, a transmit power allocation method is proposed for distributed transmit antenna Alamouti Space-Time Block Coded Orthogonal Frequency Division Multiplexing (STBC-OFDM) systems based on Cholesky-decision-feedback detector in multipath Rayleigh fading channels with multiple Residual Frequency Offsets (RFOs). The method is designed by minimizing the lower bound of the average Bit Error Ratio (BER) so as to reduce the influence of RFOs. Theoretical analysis of the lower bound on the average BER performance is presented and the closed-form expression of the optimal power allocation factor is derived. Simulation results show that in the scenario where the performance is dominated by the RFOs, the proposed method provides increasing performance gain compared with the conventional power allocation method ignoring the existence of RFOs as the ratio between the average channel gains from the two transmit antennas to receiver increasing.
2010, 32(12): 2942-2947.
doi: 10.3724/SP.J.1146.2010.00063
Abstract:
In this paper, a quick electromagnetic environment complexity evaluation method suited for important area is presented. Firstly, the index set suitable for quick evaluation is discussed and chosen; Secondly, the searching data is deeply mined to try to find abnormal signal, which consider the contribution of individual signals to the complexity; Finally, adaptive neuro-fuzzy inference system is trained and used to evaluate the complexity. The experimental results show that the evaluation model is valid.
In this paper, a quick electromagnetic environment complexity evaluation method suited for important area is presented. Firstly, the index set suitable for quick evaluation is discussed and chosen; Secondly, the searching data is deeply mined to try to find abnormal signal, which consider the contribution of individual signals to the complexity; Finally, adaptive neuro-fuzzy inference system is trained and used to evaluate the complexity. The experimental results show that the evaluation model is valid.
2010, 32(12): 2948-2953.
doi: 10.3724/SP.J.1146.2010.00055
Abstract:
This paper propose a QoS-aware resource allocation algorithm for cooperative relay-based OFDM cellular network with mixed services of different QoS requirements. The proposed algorithm solves relay selection, subcarrier allocation and power control issues. The design of resource allocation is formulated as a non-linear combinatorial optimization problem, which aims at maximizing the system sum utility while considering the data rate requirements of QoS users and the transmit power constrains for the base station as well as each relay station. To reduce the computational complexity, the global optimization issue is decomposed into subcarrier allocation and power control sub-issues. Numerical results demonstrate that the proposed algorithm is able to improve significantly the network performance in terms of power saving, system utility and system throughput.
This paper propose a QoS-aware resource allocation algorithm for cooperative relay-based OFDM cellular network with mixed services of different QoS requirements. The proposed algorithm solves relay selection, subcarrier allocation and power control issues. The design of resource allocation is formulated as a non-linear combinatorial optimization problem, which aims at maximizing the system sum utility while considering the data rate requirements of QoS users and the transmit power constrains for the base station as well as each relay station. To reduce the computational complexity, the global optimization issue is decomposed into subcarrier allocation and power control sub-issues. Numerical results demonstrate that the proposed algorithm is able to improve significantly the network performance in terms of power saving, system utility and system throughput.
2010, 32(12): 2954-2958.
doi: 10.3724/SP.J.1146.2009.01609
Abstract:
To overcoming some drawbacks for the existing multiple base-stations cooperative systems in which a central control unit and low-latency backhaul links are necessary in order to gather channel state information, a new framework and its signaling properties are discussed that allow cooperative communication on downlink links without the need of substantial modification on the conventional cellular systems. Furthermore, limited rate feedback strategies based on the index of the chosen code are explicitly analyzed, and then codebook design criteria and construction algorithm by adopting the chordal distance are presented. Some results show that the proposed scheme with relatively small feedback overhead improves the system performance significantly, compared to the conventional analog method and the Euclidean distance based quantization method.
To overcoming some drawbacks for the existing multiple base-stations cooperative systems in which a central control unit and low-latency backhaul links are necessary in order to gather channel state information, a new framework and its signaling properties are discussed that allow cooperative communication on downlink links without the need of substantial modification on the conventional cellular systems. Furthermore, limited rate feedback strategies based on the index of the chosen code are explicitly analyzed, and then codebook design criteria and construction algorithm by adopting the chordal distance are presented. Some results show that the proposed scheme with relatively small feedback overhead improves the system performance significantly, compared to the conventional analog method and the Euclidean distance based quantization method.
2010, 32(12): 2959-2963.
doi: 10.3724/SP.J.1146.2010.00017
Abstract:
Considering the condition of low Signal to Noise Ratio (SNR), the Code-aided carrier recovery algorithm performs high computational complexity. Based on the analysis of the probability density function of posterior information, modified posterior information mean with three states and new Log-Likelihood Ratio (LLR) are proposed according a probability threshold that represents the information reliability. Then a new carrier synchronization algorithm is obtained when the new posterior information mean and log-likelihood ratio are separately applied into the expectation maximization and soft frequency sweep algorithm. There is a simplified expectation maximization algorithm and evaluation function. In last, the simulations by the Turbo code with 1/6 code rate verify that the new algorithm improves the performance of computational complexity, increases the synchronization range and achieves perfect synchronization with only 0.05dB loss when SNR exceeds -7.9 dB.
Considering the condition of low Signal to Noise Ratio (SNR), the Code-aided carrier recovery algorithm performs high computational complexity. Based on the analysis of the probability density function of posterior information, modified posterior information mean with three states and new Log-Likelihood Ratio (LLR) are proposed according a probability threshold that represents the information reliability. Then a new carrier synchronization algorithm is obtained when the new posterior information mean and log-likelihood ratio are separately applied into the expectation maximization and soft frequency sweep algorithm. There is a simplified expectation maximization algorithm and evaluation function. In last, the simulations by the Turbo code with 1/6 code rate verify that the new algorithm improves the performance of computational complexity, increases the synchronization range and achieves perfect synchronization with only 0.05dB loss when SNR exceeds -7.9 dB.
2010, 32(12): 2964-2969.
doi: 10.3724/SP.J.1146.2010.00374
Abstract:
In underlay cognitive radio scene, secondary users can opportunistically access the licensed spectrum under the constraint of interference temperature. In this paper, Stackelberg game is adopted for allocating power resource. In this model, the primary user acts as leader, secondary users act as followers. When using the licensed spectrum, secondary users must make payments to the primary user based on the induced interference power to the primary user. Meanwhile, the primary user aims at maximizing its own revenue through adjusting its price, under the constraint of interference temperature. Given specified price, the secondary users compete with each other using the non-cooperative game theory. Simulation shows that, compared with the centralized optimal power allocation algorithm, the proposed distributed power allocation algorithm not only can achieve close-to-optimal performance, but also needs much less communicating overhead than the optimal algorithm which is very beneficial.
In underlay cognitive radio scene, secondary users can opportunistically access the licensed spectrum under the constraint of interference temperature. In this paper, Stackelberg game is adopted for allocating power resource. In this model, the primary user acts as leader, secondary users act as followers. When using the licensed spectrum, secondary users must make payments to the primary user based on the induced interference power to the primary user. Meanwhile, the primary user aims at maximizing its own revenue through adjusting its price, under the constraint of interference temperature. Given specified price, the secondary users compete with each other using the non-cooperative game theory. Simulation shows that, compared with the centralized optimal power allocation algorithm, the proposed distributed power allocation algorithm not only can achieve close-to-optimal performance, but also needs much less communicating overhead than the optimal algorithm which is very beneficial.
2010, 32(12): 2970-2975.
doi: 10.3724/SP.J.1146.2009.01524
Abstract:
The industrial wireless sensor networks may suffer serious random fading of wireless channel in industrial environment which is induced by the metallic equipments and apparatus. By analyzing the radio signals in industrial environment, an antenna attitude optimized system is presented based on the polarization diversity technology considering the limitation of size and cost for wireless sensor networks. The performance comparison between selection-diversity system and antenna attitude optimized system is carried out, and an antenna optimization scheme is put forward suitable for wireless sensor networks, with high effectiveness verified by measurements in actual industrial environment.
The industrial wireless sensor networks may suffer serious random fading of wireless channel in industrial environment which is induced by the metallic equipments and apparatus. By analyzing the radio signals in industrial environment, an antenna attitude optimized system is presented based on the polarization diversity technology considering the limitation of size and cost for wireless sensor networks. The performance comparison between selection-diversity system and antenna attitude optimized system is carried out, and an antenna optimization scheme is put forward suitable for wireless sensor networks, with high effectiveness verified by measurements in actual industrial environment.
2010, 32(12): 2976-2980.
doi: 10.3724/SP.J.1146.2009.01656
Abstract:
In order to solve the decision fusion issue under a dynamic large-scale Wireless Sensor Network (WSN), a voting decision fusion rule is proposed. It is assumed that the fading of channels submits a Rayleigh distribution and the number of sensors follows a Poisson distribution in the WSN. The analysis and simulation on detection performance of the fusion rule are made for different system parameters. It is shown that this fusion rule can achieve a good system-level detection performance even at low Signal-to-Noise Ratio (SNR), as long as the average number of sensors is sufficiently large, and this fusion rule can also provide better system-level detection performance than the existing nonparametric fusion rules. Moreover, a method to obtain the optimal receiving threshold of sensor output is given, by using the deflection coefficient of decision fusion system.
In order to solve the decision fusion issue under a dynamic large-scale Wireless Sensor Network (WSN), a voting decision fusion rule is proposed. It is assumed that the fading of channels submits a Rayleigh distribution and the number of sensors follows a Poisson distribution in the WSN. The analysis and simulation on detection performance of the fusion rule are made for different system parameters. It is shown that this fusion rule can achieve a good system-level detection performance even at low Signal-to-Noise Ratio (SNR), as long as the average number of sensors is sufficiently large, and this fusion rule can also provide better system-level detection performance than the existing nonparametric fusion rules. Moreover, a method to obtain the optimal receiving threshold of sensor output is given, by using the deflection coefficient of decision fusion system.
2010, 32(12): 2981-2986.
doi: 10.3724/SP.J.1146.2010.00130
Abstract:
Currently, traffic matrices have been applied to anomaly detection, traffic forecasting and traffic engineering widely, but existing researches only find the linear structure of traffic matrix. In order to search the nonlinear structure of traffic matrix, a traffic matrix model is constructed and traffic matrix datasets are collected from real Internet backbone Abilene. Using classical manifold learning algorithms, based on measurement data from Abilene find that these traffic matrix datasets with high dimensionality (81 or 121 dimensions) have a intrinsic dimensionality of 5 and have all kinds of manifold structures in low-dimension embedding space, influenced by sampling density and noise data.
Currently, traffic matrices have been applied to anomaly detection, traffic forecasting and traffic engineering widely, but existing researches only find the linear structure of traffic matrix. In order to search the nonlinear structure of traffic matrix, a traffic matrix model is constructed and traffic matrix datasets are collected from real Internet backbone Abilene. Using classical manifold learning algorithms, based on measurement data from Abilene find that these traffic matrix datasets with high dimensionality (81 or 121 dimensions) have a intrinsic dimensionality of 5 and have all kinds of manifold structures in low-dimension embedding space, influenced by sampling density and noise data.
2010, 32(12): 2987-2992.
doi: 10.3724/SP.J.1146.2010.00025
Abstract:
Adaptive optics has wide applications in astronomical observation realm because of its high capacity in improving the spatial resolution. Compared with night astronomical adaptive technology for star observation, the observation and correction for low contrast extended target at the solar surface is always a difficulty in adaptive optics. In reference to low contrast extended character of adoptive optics in solar observation, FFT correlation coefficients algorithm is utilized and a large array adaptive optoelectronic system is designed. The system, with DSP+FPGA architecture, consists of low-order correlation tracking part and high-order wave-front processing part. The experimental result shows that the system well satisfies pre-design requirements and thus demonstrates good flexibility, versatility and scalability of the scheme.
Adaptive optics has wide applications in astronomical observation realm because of its high capacity in improving the spatial resolution. Compared with night astronomical adaptive technology for star observation, the observation and correction for low contrast extended target at the solar surface is always a difficulty in adaptive optics. In reference to low contrast extended character of adoptive optics in solar observation, FFT correlation coefficients algorithm is utilized and a large array adaptive optoelectronic system is designed. The system, with DSP+FPGA architecture, consists of low-order correlation tracking part and high-order wave-front processing part. The experimental result shows that the system well satisfies pre-design requirements and thus demonstrates good flexibility, versatility and scalability of the scheme.
2010, 32(12): 2993-2998.
doi: 10.3724/SP.J.1146.2010.00688
Abstract:
Based on 65 nm CMOS low leakage process, an 8-channel 10-bit 200 kS/s SAR (Successive Approximation Register) ADC (Analog-to-Digital Converter) IP core for touch screen SoC is realized. In the D/A converter design, a 7MSB (Most-Significant-Bit)-plus-3LSB (Least-Significant-Bit) R-C hybrid conversion approach is utilized to reduce the area of the converter, and by reusing the MSB resistor string, the matching requirement of the capacitors is alleviated. With a low-offset pseuso-differential comparison approach, the input offset of the comparator is reduced. In the layout design, capacitor array symmetrical layout routing approach and resistor string dummy surrounding method are utilized to improve the matching performance. The area of the IP core is 322m267m. This converter operates with a 2.5 V analog supply and a 1.2 V digital supply. With the input frequency of 1.03 kHz at 200 kS/s sampling rate, the SFDR (Spurious-Free Dynamic Range) and ENOB (Effective Number Of Bits) are measured to be 68.2 dB and 9.27 respectively, and the power dissipation is just measured to be 440W. The design results prove the applicability of this converter to embedded SoC.
Based on 65 nm CMOS low leakage process, an 8-channel 10-bit 200 kS/s SAR (Successive Approximation Register) ADC (Analog-to-Digital Converter) IP core for touch screen SoC is realized. In the D/A converter design, a 7MSB (Most-Significant-Bit)-plus-3LSB (Least-Significant-Bit) R-C hybrid conversion approach is utilized to reduce the area of the converter, and by reusing the MSB resistor string, the matching requirement of the capacitors is alleviated. With a low-offset pseuso-differential comparison approach, the input offset of the comparator is reduced. In the layout design, capacitor array symmetrical layout routing approach and resistor string dummy surrounding method are utilized to improve the matching performance. The area of the IP core is 322m267m. This converter operates with a 2.5 V analog supply and a 1.2 V digital supply. With the input frequency of 1.03 kHz at 200 kS/s sampling rate, the SFDR (Spurious-Free Dynamic Range) and ENOB (Effective Number Of Bits) are measured to be 68.2 dB and 9.27 respectively, and the power dissipation is just measured to be 440W. The design results prove the applicability of this converter to embedded SoC.
2010, 32(12): 2999-3002.
doi: 10.3724/SP.J.1146.2010.00077
Abstract:
In this paper the characteristic of a type of oxide cathode for HIRFL-CSR electron cooler is discussed, which includes the measurement of the DC emission current density and the lifetime of the cathode. The thermic uniformity of the cathode surface and its decomposing, activating process are also tested. The results show that the cathode has good emission uniformity with 0.5 A/cm2 at 750℃～800℃and its lifetime is over 18000 h.
In this paper the characteristic of a type of oxide cathode for HIRFL-CSR electron cooler is discussed, which includes the measurement of the DC emission current density and the lifetime of the cathode. The thermic uniformity of the cathode surface and its decomposing, activating process are also tested. The results show that the cathode has good emission uniformity with 0.5 A/cm2 at 750℃～800℃and its lifetime is over 18000 h.
2010, 32(12): 3003-3007.
doi: 10.3724/SP.J.1146.2009.01618
Abstract:
For the issue of location ambiguity of high PRF(Pulse Repetition Frequency) signal in three satellites TDOA (Time Difference Of Arrival) passive location system, a new method based on direction information acquired by adding interferometer on main satellite is presented in this paper. Considering ambiguity may occur in direction finding with interferometer, a method is proposed to eliminate location ambiguity by using ambiguous direction information, which is also practicable without ambiguity in direction finding. Firstly, the distance parameter defined in this method of every location is computed, and then false locations are eliminated by threshold detection. This method is able to reduce the number of ambiguous locations, and improve location ambiguity effectively.
For the issue of location ambiguity of high PRF(Pulse Repetition Frequency) signal in three satellites TDOA (Time Difference Of Arrival) passive location system, a new method based on direction information acquired by adding interferometer on main satellite is presented in this paper. Considering ambiguity may occur in direction finding with interferometer, a method is proposed to eliminate location ambiguity by using ambiguous direction information, which is also practicable without ambiguity in direction finding. Firstly, the distance parameter defined in this method of every location is computed, and then false locations are eliminated by threshold detection. This method is able to reduce the number of ambiguous locations, and improve location ambiguity effectively.
2010, 32(12): 3008-3012.
doi: 10.3724/SP.J.1146.2009.01550
Abstract:
The range migration of the airborne bistatic SAR is determined jointly by its geometric configuration and the motion parameters of the transmitter and receiver, so it has two-dimensional space-variant properties under general configuration. The space geometric model of the airborne bistatic SAR for general configuration is firstly proposed. Via Taylor polynomial expansions to the bistatic range, the expressions of the range walk and range cure in time domain are derived. Making use of the uncoupling expressions both the transmitter range and the receiver range, the range cure compensation condition is derived. By simulations, the variant properties of the range migration and the asymmetry variant characters of the residual range cure after the migration compensation are validated.
The range migration of the airborne bistatic SAR is determined jointly by its geometric configuration and the motion parameters of the transmitter and receiver, so it has two-dimensional space-variant properties under general configuration. The space geometric model of the airborne bistatic SAR for general configuration is firstly proposed. Via Taylor polynomial expansions to the bistatic range, the expressions of the range walk and range cure in time domain are derived. Making use of the uncoupling expressions both the transmitter range and the receiver range, the range cure compensation condition is derived. By simulations, the variant properties of the range migration and the asymmetry variant characters of the residual range cure after the migration compensation are validated.
2010, 32(12): 3013-3017.
doi: 10.3724/SP.J.1146.2010.00119
Abstract:
With the traditional motion compensation methods, the low-frequency motion-error can be completely compensated during SAR imaging process in the theory. But the high-frequency motion-error cant be estimated and compensated. So when the platform of SAR system is not stable, the high-resolution imagery is hardly obtained. In the paper, a new compensation method is proposed to solve this problemcompensating the high-frequency motion-error based on sub-aperture autofocus after traditional motion compensation. With this new method, not only the imaging results are greatly improved, also the operation quantity?is greatly cut down. The experiment results using both simulated data and measured data validate the effectiveness of the method for the compensation of high-frequency motion-error.
With the traditional motion compensation methods, the low-frequency motion-error can be completely compensated during SAR imaging process in the theory. But the high-frequency motion-error cant be estimated and compensated. So when the platform of SAR system is not stable, the high-resolution imagery is hardly obtained. In the paper, a new compensation method is proposed to solve this problemcompensating the high-frequency motion-error based on sub-aperture autofocus after traditional motion compensation. With this new method, not only the imaging results are greatly improved, also the operation quantity?is greatly cut down. The experiment results using both simulated data and measured data validate the effectiveness of the method for the compensation of high-frequency motion-error.
2010, 32(12): 3018-3021.
doi: 10.3724/SP.J.1146.2010.00118
Abstract:
The aim of Sythetic Aperture Radar (SAR) imaging is to detect, identify and track the target in the complicated scene. In the paper, a new method is proposed to detect the man-made target in the complicated background. First, the improved Lee filtering algorithm using information entropy is proposed to eliminate the effect of noise and keep the detailed characteristic; then, a threshold-based segmentation technique is employed to do a cursory target detection in the high-resolution SAR image; finally, the characteristic (the size and the shape) of target is picked up based on the Directed Gradient Power (DGP). After above operation, the target is accurately detected. Meanwhile, the measured data is used to validate the effectiveness of the method.
The aim of Sythetic Aperture Radar (SAR) imaging is to detect, identify and track the target in the complicated scene. In the paper, a new method is proposed to detect the man-made target in the complicated background. First, the improved Lee filtering algorithm using information entropy is proposed to eliminate the effect of noise and keep the detailed characteristic; then, a threshold-based segmentation technique is employed to do a cursory target detection in the high-resolution SAR image; finally, the characteristic (the size and the shape) of target is picked up based on the Directed Gradient Power (DGP). After above operation, the target is accurately detected. Meanwhile, the measured data is used to validate the effectiveness of the method.
2010, 32(12): 3022-3026.
doi: 10.3724/SP.J.1146.2010.00103
Abstract:
Compressed sensing provides a new technique for solving an underdetermined system of linear equations according to sparse nature of its solutions. In this paper, the SAR side-lobe reduction technique based on compressed sensing is proposed. Under the high signal to noise ratio condition, the side-lobe can be suppressed at no cost of resolution of the image, and super-resolution can be also achieved when the scatters are sufficiently sparse; At low SNR is not only a better suppression of the side-lobe, but also inhibit the effect of noise. One-dimensional, two-dimensional and actual data processing verified the validity of this algorithm.
Compressed sensing provides a new technique for solving an underdetermined system of linear equations according to sparse nature of its solutions. In this paper, the SAR side-lobe reduction technique based on compressed sensing is proposed. Under the high signal to noise ratio condition, the side-lobe can be suppressed at no cost of resolution of the image, and super-resolution can be also achieved when the scatters are sufficiently sparse; At low SNR is not only a better suppression of the side-lobe, but also inhibit the effect of noise. One-dimensional, two-dimensional and actual data processing verified the validity of this algorithm.
2010, 32(12): 3027-3031.
doi: 10.3724/SP.J.1146.2010.00232
Abstract:
This paper deals with the issue of 2D Fractional Kalman Filter (2DFKF) and its applications to image enhancement and recognition. With the introduction of 2D fractional differential, 2DFKF recursive equation is first presented. Next, a state space model of a image given and, based on this, the 2DFKF algorithm is proposed. Finally, an example is given to demonstrate the effectiveness of proposed algorithm and the simulation result shows that the details of the image are enhanced, while the background noise of the image is efficiently attenuated.
This paper deals with the issue of 2D Fractional Kalman Filter (2DFKF) and its applications to image enhancement and recognition. With the introduction of 2D fractional differential, 2DFKF recursive equation is first presented. Next, a state space model of a image given and, based on this, the 2DFKF algorithm is proposed. Finally, an example is given to demonstrate the effectiveness of proposed algorithm and the simulation result shows that the details of the image are enhanced, while the background noise of the image is efficiently attenuated.
2010, 32(12): 3032-3036.
doi: 10.3724/SP.J.1146.2010.00034
Abstract:
This paper studies the issue of joint estimation of frequency, two-dimensional (2-D) Direction Of Arrival (DOA) and polarization of coherent sources using a uniform linear electromagnetic vector sensors array, and a Spatial-Temporal Polarization Smoothing Algorithm (STPSA) is proposed. The STPSA is used to decorrelate the source coherency and eliminate the effect of noise with the two time-delayed sets of data collected from different subarrays and different sensor components of vector sensors, and then the propagator method is applied to estimate the sourcefrequency, 2-D DOA and polarization. Comparing with the existing estimate methods, the proposed method requires neither the eigen-decomposition into signal/noise subspaces nor the iteratively searching, hence, has lower computational load. Furthermore, the proposed method can offer the improved estimation performance by increasing the intervector sensor spacing much farther apart than a half wavelength to extend aperture without any additional disambiguation processing. Simulation results show the effectiveness of the proposed method.
This paper studies the issue of joint estimation of frequency, two-dimensional (2-D) Direction Of Arrival (DOA) and polarization of coherent sources using a uniform linear electromagnetic vector sensors array, and a Spatial-Temporal Polarization Smoothing Algorithm (STPSA) is proposed. The STPSA is used to decorrelate the source coherency and eliminate the effect of noise with the two time-delayed sets of data collected from different subarrays and different sensor components of vector sensors, and then the propagator method is applied to estimate the sourcefrequency, 2-D DOA and polarization. Comparing with the existing estimate methods, the proposed method requires neither the eigen-decomposition into signal/noise subspaces nor the iteratively searching, hence, has lower computational load. Furthermore, the proposed method can offer the improved estimation performance by increasing the intervector sensor spacing much farther apart than a half wavelength to extend aperture without any additional disambiguation processing. Simulation results show the effectiveness of the proposed method.
2010, 32(12): 3037-3040.
doi: 10.3724/SP.J.1146.2010.00052
Abstract:
By using shift sequence set, a new construction of ZCZ (Zero Correlation Zone) array pairs set is proposed. Compared with known results, the obtained ZCZ array pair sets include more array pairs. For computing the size of zero correlation zone, the difference matrix of shift sequence set is introduced. Furthermore, more constructions of shift sequences are presented. The proposed technique can be used to construct the ZCZ punctured arrays set.
By using shift sequence set, a new construction of ZCZ (Zero Correlation Zone) array pairs set is proposed. Compared with known results, the obtained ZCZ array pair sets include more array pairs. For computing the size of zero correlation zone, the difference matrix of shift sequence set is introduced. Furthermore, more constructions of shift sequences are presented. The proposed technique can be used to construct the ZCZ punctured arrays set.
2010, 32(12): 3041-3045.
doi: 10.3724/SP.J.1146.2009.01629
Abstract:
A short-wave FH/OFDM system based on joint iterative demodulation and decoding is proposed. By combining the advantage of NSD technique and iterative demodulation and decoding, the method could compensate the SNR loss caused by differential detection and improve the reliability of short-wave communication system. Numeric simulation results show that the proposed method could effectively improve the system BER performance, while it reachs to10-5 when SNR is 5.5 dB and data rate reachs to 8.192 kbps. Joint iterative demodulation and decoding is a practical algorithm for FH/OFDM systems.
A short-wave FH/OFDM system based on joint iterative demodulation and decoding is proposed. By combining the advantage of NSD technique and iterative demodulation and decoding, the method could compensate the SNR loss caused by differential detection and improve the reliability of short-wave communication system. Numeric simulation results show that the proposed method could effectively improve the system BER performance, while it reachs to10-5 when SNR is 5.5 dB and data rate reachs to 8.192 kbps. Joint iterative demodulation and decoding is a practical algorithm for FH/OFDM systems.
2010, 32(12): 3046-3050.
doi: 10.3724/SP.J.1146.2010.00102
Abstract:
In RISC-DSP processor, the instructions with dynamic execution cycles cause difficulty to data dependence detection. Based on distributed dependence detection model, the detection operation is expressed by the calculation of the dependence relationship set. The relationship set at next cycle for each pipeline state are calculated. In the mean time, current pipeline state is detected based on the current instruction distribution and signals from each function module. Thus the instruction dependence can be determined one cycle in advance. Logic optimization based on its set operation is also proposed and implemented in our designed processor MediaDSP64. Synthesis result shows that with slight increase in hardware resource and power consumption, the detection operation can be hidden from the original pipeline critical path, and its timing delay can be decreased by about 30%.
In RISC-DSP processor, the instructions with dynamic execution cycles cause difficulty to data dependence detection. Based on distributed dependence detection model, the detection operation is expressed by the calculation of the dependence relationship set. The relationship set at next cycle for each pipeline state are calculated. In the mean time, current pipeline state is detected based on the current instruction distribution and signals from each function module. Thus the instruction dependence can be determined one cycle in advance. Logic optimization based on its set operation is also proposed and implemented in our designed processor MediaDSP64. Synthesis result shows that with slight increase in hardware resource and power consumption, the detection operation can be hidden from the original pipeline critical path, and its timing delay can be decreased by about 30%.
